How Will 100 Million Independent Workers Find & Manage Work In 2030
34% of the US workforce (53 million people) is already working independently in one form or another (freelancing), transacting $715 billion in GMV, primarily offline and through their own networks. By 2030, conceivably, trillions of dollars in GMV from this market would have gradually moved online — who will be the market participants facilitating that ? How do we augment humans and improve how they are plugged in to the labor market ? What are the current business models in this space, where and how have they succeeded and failed ? What are the key challenges faced by independent workers ? Let us make the world a better place for generations to come.
Tl;dr
PS: I started writing this in late 2016, published initially in 2017; and then re-published. Leaving it as-is, with no content edits since:
- 53 million people are already freelancers in the US, facilitating $715 billion per yr. 34% of the US workforce is already a freelancer, equating to about 53 million people. This is expected to grow to 60 million, or 40% of the workforce by 2020 (Intuit study). Is it inconceivable to project this to at least a 100 million independent workers in the US alone by 2030 ?
- Advances in AI/ML/increase in computing power will limit the jobs available to independent workers.
- 92 million 15–35 yr olds in the US — known as the millennials generation (out of which 38% are freelancing => 35 million millennials are freelancing).
- Over 65% of users don’t download any new apps in a month, and of those that do, over 80% abandon it within the first month => your app breaking out are very slim. Android, on Samsung devices, leads iOS.
- From a funding and M&A perspective — the on-demand labor startups are characterized by a lot of promise but significant lack of M&A. The big heavy-hitters Uber and Lyft skew the fundraising. CB Insights data tracks that (included above).
- Real-world connections are the primary way freelancers find work.Online job boards and classifieds sites, especially Craigslist are also increasingly important, but the bulk of the freelancer-customer interaction has not yet moved to an efficient online labor marketplace.
- Efficient, worker-friendly models will prevail. Current worker classification system is not working. Current system requires the workers to be either be employees (W2) or independent contractors (1099). There is no third category which exists as yet. For startups utilizing independent contractors (1099), there is the threat of worker classification lawsuits and potential of back-dated taxes by IRS.
- Business models of current online freelancer marketplaces involve taking 10%-30% of the worker’s pay as commission.
- Almost everybody has a smartphone, but it is incredibly difficult for new apps to crack through the noise of other apps and user attention span. Stats: 92% of 18–35 yr olds, and 65% of 35+ yr olds in US have a smartphone. 60% Android; 40% iOS. But be careful with that, over 65% people don’t download any new app in a month; and within 3 days the app which is downloaded loses 80% of its daily active users. People spend most of their time on social and entertainment apps.
- Craigslist is the dominant US online marketplace by far, with over 60 million classified ads posted each month, and over 700 million monthly visits. Key insight: How do you make your product work with Craigslist, instead of engaging in a zero-sum game of trying to replace it altogether ?
- Uber is the most efficient labor marketplace — although currently for ridesharing and deliveries only. 15% of the US population has used a ridesharing service. Other labor marketplaces: Upwork — Dominant global freelancer website, over $1 billion facilitated. Primarily for technical work which is done online. Thumbtack: Innovative pay-per-quote business model. Great for finding that first home service provider job, with over $1 billion facilitated. Fiverr: growing. AngiesList: On the downward trend — this used to be the go to site in the US for reviews of home service providers. TaskRabbit is effectively dead. With such high promise and media attention, and extremely low usage numbers, TaskRabbit’s era is over. Less than 4% of US population has used an online task service. Craigslist has 1,000 times monthly active visits than TaskRabbit does. Handy is effectively dead. High promise, and low usage numbers, Handy’s promise has failed to materialize as yet. Fundamentals of the business model which afflicted Homejoy would hurt Handy as well.
- Primary reason why “Uber for X” startups have and will continue to fail: failure to understand relationship-building for services which cannot fundamentally be commoditized. Marketplaces cannot commoditize workers where individual relationship is important between the supplier and the customer — this involves all at-home services. Ridesharing and deliveries — commoditizing workers can work if you are able to scale. Multiple startups in the at-home services space continue to fail/lose momentum because they thought they could be “Uber for X”.
- Primary concern for freelancers is income.
- Worker-focused marketplaces make life better from a tax point of view for freelancers Taxation is a hassle for independent contractors and being on a marketplace which can document the job activity, expenses and issue a 1099-MISC or a 1099-K when needed, helps. Marketplaces alleviate the tax burden on the consumer/business as well who need to employ such workers. Except when/if more than $2000/yr is paid to a home worker, who then gets classified as the homeowner’s domestic employee — making those wages medicare and social security wages with more paperwork for all involved.
- Critical gaps in on-the-job insurance for such workers — falling through the cracks.
- Obamacare/Affordable Care Act has fundamentally enabled the gig economy by providing the ability for freelancers to choose their health insurance and get subsidies on that.
- Peer-to-peer support for workers: The only way a 100-million-worker-scale HR would work is by a peer-to-peer support model, where passionate enthusiasts of a labor marketplace take it upon themselves to share their wisdom/best practices with newer entrants. Hiring a customer service/HR department at a massive scale has to be impractical/costly. The biggest such forum currently only has 60k members (Uberpeople). Facebook Groups and other online localized groups are currently how workers are organizing themselves. Uber and Lyft drivers have been self-organizing themselves based on their city. Workers need to share their stories.
- Worker Classification: Shift to W2 employees for Instacart and others. On-demand worker classification lawsuits. Settlement by Lyft and Uber — startups need to be mindful of what constitutes an employee.
- Increasing experiments being seen to provide tools and services to independent workers.
My son is a few months old. When he is 15 and is looking for his first temp job, he will not find it through an agency or classified sites of today. He will find it through an efficient on-demand labor marketplace, which is aware of his skills/interest and can match him up predictably, and in real-time, with the demand. The user interface for consumers and workers would go through iterations as technology evolves, but what would remain valuable is the network which efficiently and predictably matches supply with demand.
Pew Research did this remarkable study on how are Americans actually using the new sharing economy services. Contrary to popular belief, many of these services are in very, very early stages of actual usage. They have not cracked mainstream yet.
Right now, in 2016, only 4% of consumers in the US have used such a service to get errands done and 15% have used for hailing a ride (Uber).
Compare this with 34% of the workforce which does freelance work (contributing over $715 billion annually in GMV); and almost the entire country having Internet access and 72% having smartphones. By 2030, trillions of dollars transacted by freelancers would have moved on to marketplaces way more efficient than the ones which exist today.
Lets look at what are the key enabling factors, and challenges faced by independent workers. What do the next generation of startups building labor marketplaces need to keep in mind ? What kind of startups servicing this new economy can emerge ? This is the single biggest opportunity of our times and the next several decades — how do we augment humans and improve how they are plugged in to the labor market.
92% of 18–35 yr olds, and 65% of 35+ yr olds in US have a smartphone
According to this Pew Research study, close to 3/4ths of all adults in US and Canada have a smartphone. This is the necessary building block for consumers and workers to transact in real-time.
And out of the smartphone market, iOS-based Apple devices have close to a 40% share, with Android-based devices accounting for close to 60%(Samsung being the leader).
34% of US workforce is already a freelancer in one way or another
This is another key ingredient for an efficient on-demand labor marketplace to rise: the freelance workforce already exists. They don’t have to be convinced, they already are! People want to work on their own schedule, for whom they want, when they want. Freedom is an essential state of being for humans.
Uber is an exceptional case as it drove a new segment of people, who were previously not cab drivers, to start driving part-time. But for others, they would be tapping in to this existing freelancer workforce which currently transacts through other means.
How big is that workforce ?According to this two year old report by Freelancers Union, also covered in the Harvard Business Review, 53 million Americans, or 34% of the US workforce, were working as freelancers. At that point, this workforce was estimated to be adding $715 billion to the economy through their work. (Updated report in 2015 cites this as 53.7 million now, up by 700,000 freelancers from the prior year).
Is it inconceivable to imagine then that by 2030, at least 100 million people in the US alone will be freelancers, working on-demand, for multiple employers/consumers, based on supply and demand and efficient matching ?
Advances in AI/ML/increase in computing power will limit the jobs available to independent workers
Advances in AI/ML and computing power will lead to a substantive shift in types of jobs available to humans.
This is a longer-term trend. Does the job involve cleverness, helping others and requiring negotiation ?
Highly likely to be automated: Telemarketers, retail sales, cashier, tour guides, waiters, bartenders, taxi drivers.
Least likely to be automated: High school teachers, electricians, chiropractors, physical therapists, dieticians.
Millennials (18–34 yr olds) are the largest segment of the workforce, and they love freelancing (38%)
More than one-in-three American workers today are Millennials (adults ages 18 to 34 in 2015), and this year they surpassed Generation X to become the largest share of the American workforce, according to new Pew Research Center analysis of U.S. Census Bureau data.
There are 92 million 15–35 yr olds in the US — known as the millennials generation (out of which 38% are freelancing => 35 million millennials could be freelancing).
A back-of-the-envelope calculation, $500/month in casual, on-demand work earnings for millennials already freelancing translates to $210 billion/yr in GMV what a marketplace is potentially leaving on the table by not engaging this segment of the population ($500*35m * 12 mo).
Other trends to understand about the milllennial generation — grown up with technology, more debt and lower income, interested in access (car, house) instead of buying, 1/3rd stay with parents still, putting off marriage and kids later on in life, 60% of millennials prefer to rent instead of own, want maximum convenience at lowest cost, brands are shrinking — their social networks are important in making purchasing decisions, quality is key but price is more important than for other generations, more into health/exercising/eating well and related apps,
“25 YEARS FROM NOW, CAR SHARING WILL BE THE NORM, AND CAR OWNERSHIP AN ANOMALY.”
- Jeremy Rifkin, Author and Economist
More from the exhaustive Goldman Sachs piece about understanding the millennial generation:
The Millennial generation is the largest in US history and as they reach their prime working and spending years, their impact on the economy is going to be huge.
Millennials have come of age during a time of technological change, globalization and economic disruption. That’s given them a different set of behaviors and experiences than their parents.
They have been slower to marry and move out on their own, and have shown different attitudes to ownership that have helped spawn what’s being called a “sharing economy.”
They’re also the first generation of digital natives, and their affinity for technology helps shape how they shop. They are used to instant access to price comparisons, product information and peer reviews.
Finally, they are dedicated to wellness, devoting time and money to exercising and eating right. Their active lifestyle influences trends in everything from food and drink to fashion.
Friends, family and professional contacts is the primary way freelancers find work
- Real-world connections (35%): word-of-mouth, personal and professional contacts. This is listed as being the primary way, with online social networking playing an increasing role (29%).
- Online job boards (29%)
- Online classifieds such as Craigslist (25%): Over 80 million classified ads posted each month, with over 60 million unique visitors each month in the US and over 50 billion page views per month. According to this research, Craigslist entry has had a devastating impact on local newspapers classifieds business, reducing the classified ad rates by ~21% and saving classified-ad buyers about $5.4 billion from 2000–2007.
This article in Forbes said this about Craigslist: “a business model which, by any rational standard, is completely insane”. It includes: $25 for job postings in six of its largest US markets and $75 for job listings in SF and $10 fee for brokered apartment listings in New York City. And that’s it — all other listings are FREE.
- Local newspapers (12%): classifieds section.
- Employment agency/staffing firm (15%)
- Google search: Users finding providers organically through keyword search.
- Specialized freelance marketplaces (16%)
- Sharing economy/on-demand apps (11%)
Understanding Craigslist: 1000x wider reach than TaskRabbit, the ex flag-bearer of the gig economy
After eBay, Craigslist is the only real Internet scale open marketplace in the US currently. With TaskRabbit, there is a 1000x difference in reach between the two. Kijiji in Canada does what Craigslist does in the US: unknowingly nipping other emerging marketplaces before they can bloom.
Not only does Craigslist have a huge audience, it is also highly engaged, much better than any other marketplace.
Working with Craigslist vs trying to over-run it
“For me, Craigslist is all about local face to face transactions. I don’t take Paypal, I don’t ship things, and I always meet folks in public places. To me, this is common sense.”
Real-world, offline, person-to-person transactions are being facilitated at a massive scale through Craigslist, for free. While eBay has an annual GMV of around $80 billion; how much GMV is transacted through Craigslist is not known as it is offline — spread across real estate, jobs, gigs, items and other listing types.
Craigslist pursues a minimal monetization strategy, not beholden to any venture investment or growth path, seeking to serve the community. Some answers by Craig Newmark, Craigslist founder, on Quora put this into perspective.
A trend which has emerged over the past several years is that of new marketplace/on-demand products focusing on replacing Craigslist in a particular category (such as Uber and Lyft for ridesharing; or Airbnb for homesharing). Idea here is to move the marketplace in a particular category from Craigslist over to a new platform (sometimes co-opting and then dumping Craigslist in the process).
Ok, maybe for a few well-capitalized startups which had other factors going for them this has worked out well — but how many long-lasting brands can be created by unbundling Craigslist ? Craigslist still gets almost a billion monthly active users and makes over a $100 million a year; and its founder doesn’t care about making a billion. It is not ideal, but it is good enough.
How do you compete with something free, simple and already pervasive ?
How about startups which recognize that Craigslist is not going anywhere anytime soon, and instead use and augment its market instead of being in a zero-sum game of owning the marketplace entirely ?
Airbnb’s Craigslist growth hack
Airbnb executed what is now known as one of the most famous “growth hacks” in Internet startup history — when during the early years in its history — it made it feasible for its users to quickly cross-post their Airbnb listing to Craigslist (which doesn’t have an API for this purpose), exposing it to the wider network of Craigslist users. Those potentially new users then still had to book through Airbnb if they were interested in the listing.
Today it doesn’t do that anymore — it doesn’t need to and Craigslist doesn’t allow it either, going so far as to not letting any listing which has the phrase “airbnb.com” in it (try it!).
Now that everybody knows about it, this ruins it for other startups hoping to pull off a similar stunt. Craigslist can act a lot sooner now, sapping the momentum out of a startup’s growth plans. If a much larger # of people in your category are on Craigslist compared to your walled marketplace, why would they choose you ? How do new vertical marketplaces grow then if they can’t ride off Craigslist ?
But hey, Craigslist has its challenges. A simple search for “craigslist robbery” turns up the below:
Business models of popular labor marketplaces across online and offline markets
Some marketplaces try to offer a more structured approach than Craigslist and they charge freelancers/service providers for usage of their platform (and in some cases, they charge consumers to use their platform too). These services aggregate demand and supply online and match independent contractors with customers for service fulfillment. These can be grouped based on skill-level/licensing requirements:
Online freelancers — geolocation irrelevant
Examples include writers, developers and designers.
- Upwork (formerly elance/oDesk): Charges 10–20% commission to its workers. Claims it facilitates $1 billion of work annually for freelancers using its platform. $175 million in venture funding. CEO believes within six years will hit $10 billion/yr in revenue.
- Fiverr: Charges 20% per transaction, facilitates 5 million transactions/month; most at $5 but now allows freelancers to list their own price point. Its founder believes 97% of freelancing happens offline; and US will have a 45% gig economy workforce by 2020, so HUGE opportunity awaits. Clearly, its investors believe so as well — funding it with $110 million till date over 5 rounds.
- Freelancer.com: Charges 20% to freelancers; also charges customers 3%; amongst other complex charges. Claims it is the largest global marketplace of freelancers with almost 20 million users globally. It has grown through acquisitions of sites such as scriptlance, rent-a-coder, etc. Headquartered in Australia.
- Guru.com: Charges 8.95% to its freelancers; which they add to what they charge the customers. Claims it has paid $200 million to gurus.
- LinkedIn’s ProFinder: Charges $59.99/month as part of its mandatory LinkedIn Premium Business Plus subscription for service providers to be able to respond to more than 5 project proposals. Consumers can post jobs, see how they are linked in to the professional and also contact through custom forms to request a project proposal from them.
- PeoplePerHour: Charges 15% + complex charging to its freelancers. UK-based, claims has facilitated over 40 million pounds in earnings for freelancers on its platform.
Home service providers: Offline, hyperlocal
$400 billion home services market. Examples include electricians, plumbers, painters, landscapers, decorators and others.
- Houzz:
Thanks to Houzz’s vast reach, homeowners around the world are discoveringAbrams’s work as they surf their phones, tablets and web browsers, looking for appealing remodeling ideas. More than 25 million users a month visit Houzz’s site — an audience far beyond what old-fashioned print advertising ever could have created for designers like Abrams. The payoff: thanks to Houzz, Abrams has signed up clients in New York and Baltimore. He’s also chatting with prospects as far away as Rancho Santa Fe, Calif., and Dubai.
All told, 8,900 people officially follow Abrams on Houzz, so they can systematically take a look at his work. They ping him with questions — and he puts in hours each week addressing their queries. It’s time well-spent: email exchanges lead to new contracts. Abrams still reads so-called shelter magazines such as Home & Garden and Interiors that used to be the bibles of his industry. But he says he doesn’t advertise in them anymore. “I get 65% to 70% of my business from Houzz,” he says.
- Thumbtack:
Innovative business model of pay-per-quote based on demand (small jobs quotes are cheaper; bigger jobs quotes are more expensive) — providers quote for jobs. Top providers get 1 in 10 jobs they quote for. To start with, providers must buy 10 “quotes worth of credits” (eg if first quote cost 9 credits, will be asked to buy 90 credits) — and then in packages of 20, 50 and 150 credits. Each quote is between 2 to 9 credits. Each credit is worth $1.67 (bulk buy reduces per credit price to $1.5). So a 9 credit quote costs 9*$1.67 = $15.03; while a 2 credit costs $3.34. Before you dismiss Thumbtack for its tough to grok business model, consider that it now claims $1 billion annual revenue for professionals using its platform, with 200,000 active professionals, doing 5 million projects each year across 1100 services. Thumbtack is valued at over $1.25 billion by venture capital investors, and has raised well over $250 million in funding till date.
List of negative reviews from contractors and customers highlighting problems with Thumbtack. Key issues: Worthless leads, incompetent contractors.
Advertiser-driven — where workers have to spend their hard-earned $$ on premium online placements
- Angie’s List: Providers pay for advertising, freemium model for consumers, where they pay subscription fees to access value-added features. Estimate $345 million /yr in revenue. Advertiser-driven. Workers pay for advertising to Angie’s List network of homeowners.
With the coming changes, it will be much harder for consumers to find information on companies who are not advertisers. All A and B advertisers will appear first. There will be only one or two pages of total results. Non-advertisers will be limited to their name and rating. The service description and number of reviews will be removed. Angie’s List will run advertisements for competitors on the pages of those that do not advertise.
- Yelp: Providers pay for advertising their profiles, for not showing competitors ads on their profiles, among other ways Yelp can charge them. Advertiser-driven. Workers pay for advertising to Yelp’s network of consumers.
- Amazon Home Services: Charges 15% service fee to its workers.
- Google Home Services: In SF Bay area only. Even though they have invested $100 million in Thumbtack (apart from $250 million in Uber) through their Google Ventures, they are supposed to be entering this market.
- Porch.com:
Along with generating revenue from professionals paying for more prominent placement, Porch charges a booking fee to professionals when homeowners purchase their services from the Porch mobile app. The app, available only in Seattle for the past month, includes only professionals who have been verified by Porch and who have agreed to pre-negotiated prices for set services.
My reconnaissance mission to Lowes revealed that the store will remain a direct competitor with my company, and they will recommend themselves first for projects. If pressed further for information, Lowes representatives may then refer consumers to Porch.com, a website listing all of my other competitors. This marketing style does not sit well with me…Also, lack of Porch.com signage throughout the store, and poor displays of the signs that do exist, indicate that the Porch.com brand is not as widely promoted as a paying advertiser would expect. Of course, this is just the information that I gathered in my local Lowes.
Freemium model: $100/month with Porch Premium to get a feature like this “Leads from your profile page go directly to you instead of other professionals”: https://enroll.porch.com/
Porch.com charges contractors separately for each trade category in which they wish to advertise based on zip codes selected for their area.
3.2 M home professionals, 1.5M unique monthly visitors, 100M in funding raised in less than 18 months
Lowe’s: 1.7k stores nationwide, 15M customers/week, 1.2M professionals included in searches by Lowe’s associates last year
Porch lets people search for home services professionals by browsing pictures of past home-improvement projects along with the price tag, user reviews and other details. Associates at all 1,720 Lowe’s stores nationwide now use Porch as their in-store tool to help shoppers find home pros.
Non-specialized, general labor workers
On-demand drivers — people and things
There are vertically-focused startups trying to “unbundle Craigslist” by providing a better, mobile-driven user experience. They have focused on utilizing a 1099 independent contractor general labor workforce and few have battled lawsuits concerning whether those workers are employees or not. Examples include food and other stuff delivery drivers, rideshare drivers, grocery shoppers, cleaners, movers and dog walkers, among others.
General / Miscellaneous errands:
On-demand personal services
- TaskRabbit: Charges 30% commission to its workers.
- DogVacay: Charges 15% to its workers.
- Care.com:
Basic Membership is free and allows families to view and post jobs, receive newsletters full of expert tips and advice, and access a wealth of content on all things care-related. Premium Members pay a subscription fee ranging from approximately $37 for a monthly subscription to $147 for an annual subscription and enjoy all these services, plus the ability to view contact details and make contact with an unlimited number of caregivers, purchase background checks, and access references for providers.
Care.com is the world’s largest online destination for finding and managing family care, with 11.0 million families and 8.6 million caregivers* across 16 countries, including the U.S., UK, Canada and parts of Western Europe, and approximately 800,000 employees of corporate clients having access to our services. Spanning child care to senior care, pet care, housekeeping and more, Care.com provides a sweeping array of services for families and caregivers to find, manage and pay for care or find employment
Sittercity.com: serves more than half a million users in cities across the United States. Sittercity’s scope of offerings include sitters, nannies, pet sitters, and senior care; more than a million caregiver profiles nationwide.
Sittercity offers background checking on sitters and care providers. Care.com, on the other hand, no longer offers any type of background checking.
Comparison: http://www.iparentinglife.com/sittercity-vs-care-com-which-is-best/
Rideshare/delivery drivers:
- Uber: Charges 25–30% to its workers, depending upon market and other factors. 50 million rides in the US in March 2016 were on Uber. Also from that Bloomberg article, Uber was profitable by $0.19/ride in the US market.
- Lyft: Charges upto 25% to its workers. 11 million rides in the US in March 2016 were on Uber. “Top 20 markets account for 80–90% of all ride-hailing rides”.
- Postmates: Charges 20% of the delivery fee to its workers.
- Handy: Charges 10–15% to its workers. Raised almost $100 million in venture funding over the past 4 years.
- Homejoy: Now dead, used to charge 25% to its workers. Amazing writeup of why it failed, but more on that later.
Challenges for labor marketplace startups: commoditizing workers
Labor marketplace startups can either create value for its participants or suck value from them.
From a customer point of view, if worker A is replaced by worker B for the next job and it makes little to no difference to the customer, then the startup’s marketplace is where all participants are deriving value. The marketplace in this scenario is commoditizing the workers, and that is ok. It can seek to optimize and scale that to remain in business — otherwise if the workforce is not busy enough, and thus not earning sufficient money, they will lose interest and leave.
But, if the customer really liked the service provided by worker A, and wants that specific worker to return for the next job — then the marketplace at that point is a barrier, sucking value from the ecosystem. The worker and the customer have no need to go through the marketplace for subsequent jobs, and they will find ways around it. Most marketplaces seek to punish this kind of behaviour, which they call disintermediation — but it is a fundamental aspect of the design of their business model.
People start caring who is servicing the request closer it gets to their home. There is no data I have for this so it is my opinion against yours. But here is how I view it:
- Local, < 60 mt ridesharing — simply do not care if it is Jack or Joe taking you from point A to B, as long as it is quick! UberX
- Curbside/mailbox deliveries: Impersonal / like your mailman, in the mailbox: You don’t care who delivers your mail to your mailbox/outside the door as long as you don’t need to interact with them. A reasonable amount of trust in the system is sufficient. UberDelivery
3. Ring-the-doorbell deliveries no judgment required:
- Restaurant food — UberEATS
- Deliver stuff bought pre-selected and bought such as clothes
- Household supplies
- Others ?
Each one of these categories has a dominant player: Uber. It has the customers, drivers and data which it can leverage to provide an optimal experience to all participants. Startups don’t have breathing space here except if they catch on to a particular niche/value-added service. Example: HopSkipRide — ridesharing for kids only — with additional background checked drivers. But, do you want a different driver every time for your kids ? Would you trust a brand name with your kids or the individual servicing the request ? Is this really a business where one driver can be swapped for another easily ?
The below categories are what you cannot commoditize, and these are great only for local, lifestyle businesses but not for venture-backed high growth startups. These are not the business models which will last for marketplaces /on-demand startups — even if venture capital is used for paid user acquisition for now.
Why ? All these services involve relationship-building, which the service provider does with the customer directly. They involve someone coming inside your home to provide you a service, using some level of judgment and knowledge about you/your home. The customers in these cases prefer for someone they have already worked with and like to come back again, as opposed to a new person every time for the same task — who will then have to learn their house and needs again, and build that level of trust again. Marketplace and on-demand startups are good for the first introduction here only. Thumbtack is successful because it is in the first-introduction business only, with its pay-per-quote business model.
Business models of charging 15% (such as TaskRabbit does) every time for even repeat customers from the same “tasker” are unsustainable. Why ? People are not stupid. An off-hand, cash only transaction works well enough.
1. Ring-the-doorbell deliveries with judgment required:
- Groceries: It’s an experience and an outing buying your own groceries. How do you provide a consistently high quality of produce-selection. When you get a bad apple delivered — is it because the shopper didn’t look hard enough as you have or all apples were bad ?
- Prescription drugs: Will you be able to trust 100% the drugs you are getting are the same ones filled by the pharmacy ? Isn’t there room for manipulation here by unscrupulous actors by the time it gets to your order ?
2. Step-inside-my-house services:
- Cleaner: Do you want a different one every time ?
- Electrician: Do you want a different one every time ?
- Plumber: Do you want a different one every time ?
- Handyman: Do you want a different one every time ?
- Landscaper: Outside the house, but do you want a different one every time ?
- Snow removal: Do you want a different one every time, who then has to learn about your driveway’s particulars.
3. Personal attention:
- Personal: Massage, hair stylist, doctor — Do you want a different one every time ?
- Kids: baby sitter, tutor — Do you want a different one every time ?
- Pets: pet sitter — Do you want a different one every time ?
- Elders: nurses, caregivers — Do you want a different one every time ?
What’s going on with TaskRabbit — despite over $36 million in funding and a 7 year history
“Too many of the Rabbits, it turns out, were not happy bunnies — they were underpaid and did a poor job, despite company rhetoric to the contrary. An increasing number of them simply failed to show up for their tasks. As a results, customers also failed to return….Market forces have resulted in a convergence of companies on a few services which had been the most used on their platforms.”
The sentiment amongst “Taskers” on Facebook groups like these (New York City Task Rabbits) is generally:
- There aren’t enough jobs.
- There are too many taskers in concentrated areas — leading to reduced prices.
- They can’t do it full-time.
- Slows down over the winter.
Overall, general sentiment seems to be this is not quite cutting it. Where are the happy taskers, similar to the relatively happy Uber and Lyft drivers I found online, who are at least making money and posting screenshots of their payouts ?
Challenge #1: Stable & Good Enough Income
Freelancers top concern is the ability to find work (and to acquire skills required for that), to find it frequently enough and to make sure the clients’ payment goes through. These are all about having a predictable, meaningful income at the end of the day.
Uber is the most efficient labor marketplace currently — for people who can legally drive, have access to a car and have some time to spare
In the on-demand labor economy world, for independent contractors who have a car, best earning potential is on Uber. It has volume, steady demand from customers and ability to boost earnings by surge multipliers based on low supply and high demand.
However, even Uber, as good as it is, is not good enough when it comes to drivers take-home pay. BuzzFeed did an in-depth analysis with Uber data: “How much Uber drivers actually make per hour ?” Key finding:
Uber data suggests that drivers overall in three major U.S. markets — Denver, Detroit, and Houston — earned less than $13.25 an hour after expenses in late 2015, according to calculations based on more than a million trips. This is the methodology they used.
This chart is from an Uber blog post — trying to convince drivers that even though Uber reduced the fares for the clients (23% in this case for Chicago), driver earnings still went up (by 12%) because more customers took advantage of the lower fares. Higher volume = more driver earnings. See this too.
Of course, the driver now has to work more hours to earn the same money as before, which makes the driver somewhat unhappy. But this market is highly liquid — there are enough providers and customers in it to facilitate transactions quickly enough. There is incentive for the drivers; bonus in the form of price multiplier surges which rewards them for low supply and
high demand. Uber and Lyft have long advertised that they charge only a 25% commission of the fares they earn. However, these figures ignore the “Booking Fee” that both companies assess on top of each ride, which goes solely to the TNC. The resulting effects of this fee and the 25% commission mean that when passengers step outside of their Uber and only see a $5 charge, the driver will only get around half of that.
To make matters worse, Uber and Lyft have consistently cut prices so that most drivers are now driving twice as far to earn the same $10 dollars that they made only two years ago. The result is that drivers are forced to drive longer hours while incurring heavier expenses.
Of the 37 UberX rides we sampled, our driver rarely paid anything close to a 25% commission to Uber. In fact, the lower the total fare, the higher the commission was for Uber. Uber has used the line in the past that ‘fare cuts means higher earnings for drivers’ but after 3 years of fare cuts, we have yet to meet a single driver who feels they’re making more now than they did in the past. This data shows that not only are drivers earning less, but they’re also spending more in expenses (gas and depreciation on their vehicle) to do it. And although we used Uber trips as a baseline, Lyft’s numbers are only slightly better.
And as simon rothman of Greylock Partners points out, Uber didn’t just buy the supply and the demand — they bought the entire network. These are the most insightful paragraphs from that post, quoting here:
“If we rewind the clock to 2012, Lyft launched a peer-to-peer ridesharing service and Uber was a black car marketplace. Lyft and Sidecar’s P2P taxis were a bigger idea in a bigger space. And like most marketplaces, it was a winner-take-all opportunity, or at least a winner-take-most. At the time, all three companies were only in SF and everyone was racing towards liquidity, with sights to have a right of first refusal on every other city in the country thereafter. Lyft had a big head start, but Uber was trailing right behind.
Uber began to aggressively put its war chest to work to achieve liquidity as fast as possible. This meant enabling ~5 minute pickups for riders and ~$25 hourly earnings for drivers. First, it guaranteed drivers hourly rates and spent money to acquire drivers. With drivers on the road, Uber focused on filling their cars through paid channels and incentivizing referrals. In a nutshell, the company subsidized fares and gave away free rides until there was enough demand and drivers could earn enough on their own. City by city, Uber implemented this playbook — buying drivers, buying passengers, subsidizing rides — to shave minutes off the pickup SLA and increase driver earnings, propelling Uber to liquidity.”“…The period of cheap capital and billion dollar checks has ended. In this capital constrained market, buying scale is no longer going to be a credible lever for the next generation of startups.”
“..liquidity isn’t the most important thing… it’s the only thing. Building a marketplace is similar to building two separate companies simultaneously with each being codependent on the other. The reality is that’s really hard to achieve. The cold start problem of getting supply and demand to the same place at the same time for the same product or service is hard. Brutally hard. If you don’t have enough buyers, the sellers won’t come. If you don’t have enough sellers, the buyers won’t come. This is what kills most marketplaces — they don’t just die of natural causes.”
On-demand startups are not able to scale like Uber leading to churn of workers and users
No other startup can even match Uber due to the amount it has raised (over $10 billion) and spent in acquiring drivers, customers and buying the network in the markets it operates in. If this is the best, and Uber can’t be replicated, that makes “Uber for X” a fundamentally unsound business model over the long-term. What this really means is that “Uber for X” startups can’t provide sufficient liquidity of suppliers and customers in their marketplace. Suppliers who are not earning enough will not be good representatives of the “Uber for X” startup, in turn leading users to abandon the service eventually, after free credits are gone.
Case in point: Homejoy — on-demand cleaning service. Suggest reading this excellent analysis of why it failed. I disagree however that its lawsuit struggles over worker classification were the reason it shuttered. I believe the real reason it died is because it’s independent contractor workforce did not care enough about it. They weren’t deriving sufficient value from the marketplace which reflected in their interaction with the customers. Many other on-demand startups will die for the same reason: lack of liquidity (linking to another good read).
Commission rates charged to workers are way too high
Commission rates charged by these marketplaces have been ridiculously high. For example, TaskRabbit takes 30% commission of what the customer is paying to the worker doing the errand. If it is a repeat customer for the same worker, then the commission is 15%.
Yes, there are people who will work for almost anything, but is the value provided by the marketplace sufficient to justify a 30% commission ? Money which could have gone in that low wage earner’s pocket, who is trying to make ends meet, is instead going into the “marketplace”. And 15% for repeat business with the same customer ? What is preventing the worker from forming a relationship directly with the customer, skipping the marketplace altogether ? Threat of eviction from the marketplace ? Are there enough, Uber-scale jobs being offered in the marketplace which can make high individual commission rates palatable ?
Still, insufficient pay, not scheduling, turned out to be the most common reason workers left their jobs.
People who are working the hardest here, the ones actually connecting the dots and getting the job done, rain or shine, are getting squeezed. Since the startup is chasing dreams of scale and having more liquidity in their marketplace — the path to earning more money for these providers has been set as: work more, if the startup has been able to scale. The narrative is Our commissions are high, but that is OK as we can give you more volume of work which will result in more overall earnings for you, you just need to put in more and more hours.
From the point of view of the worker, what is the difference between an offline agency of previous years and an online on-demand / service marketplace connecting customers with workers today ? Where has the innovation been for the workers and who/how has made their lives significantly better than before, where they are able to earn more ?
Top complaints from workers included not being able to find enough work, not understanding legal obligations and taxes, and being unable to optimize schedules to maximize earnings.
Except for a few exceptions, the on-demand and service marketplaces which have emerged in the past few years have sought to provide the absolute best value/convenience to the customer and focused on their own growth, while treating service providers as an afterthought, as a cog in the wheel.
This has resulted in unhappy providers who don’t have any loyalty to the marketplace they work for as independent contractors; which then reflects in the service the customers get, which in turn impacts the growth of the startup itself once paid user acquisition wears off. Due to the structuring of these so-called 1099 workers, such startups are unable to provide guidance to them, otherwise risk them getting classified as employees which would further jeopardize their business model and Internet Scale dreams. This article in Wired captures the disillusionment amongst the workers of on-demand startups.
Half of the respondents said they planned to stop working for on-demand companies within the year, citing insufficient pay, lack of enjoyment of the work, or simply because they no longer had the need to work the job
Worker-oriented marketplaces do not exist. Look around on the Internet and there are sufficient examples of unhappiness for each one of these services.
Fundamentally new business models will be invented to address these gaps. That’s when we might see true transition happen from offline+Craigslist to on-demand labor marketplaces.
Challenge #2 : Health Benefits
Health insurance in the US is increasingly expensive and complicated even for those with full-time jobs where employer covers the cost partly, let alone for independent contractors who have to foot the entire cost themselves. How much worse ? Let’s see first what it means to have health coverage in the US if your employer covers it for you.
Employer-covered health insurance
Employer-sponsored insurance covers over half of the non-elderly population, 147 million people in total. To provide current information about employer-sponsored health benefits, the Kaiser Family Foundation (Kaiser) and the Health Research & Educational Trust (HRET) conduct an annual survey of private and nonfederal public employers with three or more workers. This is the seventeenth Kaiser/HRET survey and reflects employer-sponsored health benefits in 2015.
- Work in a small firm — likely to not get any benefits: Workers working for small firms (3 to 49 employees) are most likely to not get any benefits from their firm — with only 54% of such firms offering benefits. Compare this with over 89% rate for firms with more than 50 workers.
- Worker in big firm with over 35% low-wage workers — likely to pay more towards premium: Workers in firms with a higher percentage of lower-wage workers (at least 35% of workers earn $23,000 a year or less) contribute higher percentages of the premium for family coverage (41% vs. 28%) than workers in firms with a smaller share of lower-wage workers.
- Annual premium: Average annual premium for single workers is $6251; and for family coverage is $17,545. Lowest average premium, in high-deductible health plan with a savings option (HDHP/SOs) for single coverage: $5,567; and family coverage:$15,970
- Premiums have increased significantly compared to inflation and rise in workers earnings from 2010–2015.
- Over $1000 deductible: Among covered workers with a general annual deductible, the average deductible amount for single coverage is $1,318. 81% of covered workers have a general annual deductible for single coverage that must be met before most services are paid for by the plan.
- Additional copayment cost: Almost 68% of covered workers pay a copayment (a fixed dollar amount) for office visits with a primary care or specialist physician, in addition to any general annual deductible their plan may have.
- Cost sharing for prescription drugs: This is based on the tier of the prescription drug.
- And even more additional costs: Most workers also face additional cost sharing for a hospital admission or an outpatient surgery episode. After any general annual deductible is met, 65% of covered workers have a coinsurance and 14% have a copayment for hospital admissions.
- Choose your poison-based health insurance: If you can’t afford high premiums, you end up choosing a plan with high deductibles, which means you have to pay that before the health insurance kicks in. Either way, the lower wage worker is screwed.
PPO plans remain the most common plan type, enrolling 52% of covered workers in 2015, although a smaller percentage than 2014. Twenty-four percent of covered workers are enrolled in a high-deductible plan with a savings options (HDHP/SO), 14% in an HMO, 10% in a POS plan, and 1% in a conventional (also known as an indemnity) plan
Independent contractor health insurance
Despite the high cost and complexity, insurance is essential given the private health care system. Without it, here is an example what it will mean if an independent contractor break his/her leg somehow:
The Affordable Care Act (ACA, also known as Obamacare) has been a key enabler of health insurance for on-demand economy workers. ACA mandates employers extend coverage to only full-time employees (more than 29 hours/week), while offering subsidies for individual insurance buyers. In June 2015, the US Supreme Court issued a decision preserving federal tax credits tied to ACA. This has been huge for Uber, as acknowledged by its CEO:
The best being done by current on-demand companies is to connect their contractors with healthcare exchanges. Uber, Postmates and TaskRabbit have partnered with Stride Health which offers free recommendations to individuals on what healthcare plan would suit them the best (Stride offers this for free even without these partnerships).
Stride Health, a San Francisco-based insurance broker service that makes health care recommendations tailored to specific needs such as gender, medications, income, and preferred doctors. Those who sign up for a plan via Stride’s website or mobile app receive year-round support, including access to customer service reps who will sit on hold for hours with an insurance provider.
There are well over a hundred plans to choose from on the healthcare exchanges and Stride aims to simplify that. However, it does require the user to be sufficiently literate and smart to be able to navigate through the carriers, networks, and metal tiers. For example, I generated a recommendation for myself below — with the profile of a 21 yr old male in Houston with no prior conditions and this recommendation popped up.
In terms of benefits, Uber offers a website called “Momentum” which offers rewards and perks to drivers.
https://get.uber.com/cl/momentum/
Lyft offers a website called “Accelerate” offering similar connections.
Unfortunately, by and large, the on-demand independent contractor is on his own for health insurance in US at the end of the day. Even if he is able to navigate through the complexity and range of choices; affordability and being able to pay for the premium, deductible, prescription drug cost sharing, co-payments and other out-of-pocket medical costs for himself or his family; all lead to one path: being able to earn more $$ and more consistently.
Challenge #3: Exposure to risks — insurance
Workers in the on-demand/freelancer economy are by and large not covered under any liability insurance policy and neither do they have worker’s compensation insurance provided by any marketplace — they are on their own and exposed to significant risk as a result. There are however guarantees generally available for the customers, centered on the worker’s negligence. Ridesharing companies like Uber do offer insurance to their “driver-partners”, but they don’t really cover drivers in “Period 1” (when they are online but are not on a ride yet).
Thumbtack
..The Thumbtack Guarantee is limited to $1 million, and it applies only to customers..
The Guarantee covers up to $1 million for property damage directly and primarily caused by the negligence of a professional hired on Thumbtack.
The damage must be a result of services rendered based on your original request. Add-on projects or future projects not requested through Thumbtack, even if completed by a Thumbtack pro, aren’t covered.
The Thumbtack Guarantee is not insurance and it does not insure you or any other person. It does not cover all types of property. It does not replace your homeowners, renters, property, professional, or any other type of insurance.
This means you should still research and comply with all rules and best practices when you complete projects using Thumbtack pros. If you need additional assurances beyond those provided by the Guarantee, you must purchase insurance or other coverage to provide it.
Upwork
Upwork Payment Protection helps give security and peace of mind to clients and freelancers working together on both hourly and fixed-price projects.
Upwork Payment Protection consists of two sub-programs:
Fiverr
None
TaskRabbit
TaskRabbit has a $1 million guarantee (with caveats of course, such as, if the tasker steals over $10k of item, then not covered). These all *require* the tasker to have been negligent. What if the tasker was not negligent ? Some opinion on how to close the gap in insurance for TaskRabbit with renters’ insurance:
What if I hire someone from TaskRabbit to clean my home and they injure themselves while cleaning?
If the tasker injuries him or herself as a result of negligence on your part, your renters insurance will cover it. Other injuries will be covered up to the limit in your policy’s medical payment section.
What if I take a moving or housecleaning job on TaskRabbit and accidentally injure someone while on a job?
If you injure someone while performing a “task,” it will not be covered. The policy does not provide bodily injury coverage while conducting a business activity.
What if I hire someone to walk my dog while I’m away for the weekend and the dog bites someone at the park?
If your dog bites someone at the park, the policy will cover it assuming there was negligence on the part of the dog handler. This will vary state by state according to leash laws.
What if I don’t have renters insurance?
TaskRabbit has a similar guarantee to AirBnB, in that it protects both its clients and taskers up to $1,000,000 in the case of losses or damages. TaskRabbit will cover both damages to a client’s home and medical injuries sustained by a client, tasker, or third-party. In both cases, the tasker must be found to be negligent in the incident for TaskRabbit to cover it.
If a tasker is injured and the client is found negligent, the client’s renters insurance policy will cover the tasker’s medical bill. If you plan on letting a tasker into your home, we suggest getting a renters insurance policy.
What if the client does not have renters insurance ? How does the tasker recover ?
Freelancing
What if a laptop that I use 50% of the time for personal purposes and 50% for freelance writing work gets stolen from my apartment?
Most policies will limit the amount paid out for any computer if its used at any time for business purposes. One company we contacted would pay out $2,500 max for said laptop. Most other insurance companies also have similar limits in place.
What if it was stolen from a Starbucks?
That same company will lower its payout to $250 if said laptop is stolen from an off-site location. Most other insurance companies also lower payouts for business electronics stolen from off-site locations.
What if I accidentally lose it (left it in a taxi, on the train, etc.)?
Most insurance companies told us that this situation would require an investigation, but that if it was ruled to be lost or misplaced, rather than stolen, then they would not provide any coverage.
What if I was freelancing on site for a business and that laptop or my phone was stolen from that office space?
If your laptop is stolen from an off-site location, it would be covered under the theft portion of your policy. Payout would again be limited if the laptop is used for business purposes in any way. The same limitations apply to your phone.
Uber
Someone pulls up an app on their phone, types in the address of where they need to go, and orders your car. You turn around and drive to pick them up.
While you’re driving, you look over to your phone to make sure you’re still heading in the right direction. While you’re looking the other way, you hit a car. It’s not a big accident (luckily), but it’s bad enough that the other driver calls the police (and you have to cancel your pick-up).
The police get there, they take down all of your info, including the fact that you were driving for a company called “Uber.” You submit your claim to your auto insurance company and go about your life.
A few days later, your insurance company calls you. Unfortunately, your claim has been denied because you were performing a business activity. You’re now on the hook not only for your own car repairs, but the other driver’s, too. Oh, and by the way: your car insurance policy has been canceled.
You contact Uber to see if their insurance policy is going to cover the damages. Nope, says the woman on the other end of the phone. Since you didn’t have a passenger in the car, they’re not liable. Consider yourself lucky you didn’t hurt anyone.
Do you notice a key gap here ? During the time the driver is in “Period 1” — online but not accepted ride yet, the driver has very limited insurance coverage. Without a custom ridesharing policy or commercial policy, the drivers are operating in a grey area and risking their insurance carrier finding out.
Below is from an insurance provider’s website where they are now offering a custom ridesharing commercial policy where they cover this gap in “Period 1”:
What if I drive a car for Uber or Lyft and a passenger steals my backpack or phone from the car?
All theft of personal belongings from your car are covered under the theft portion of your policy.
What if I don’t have renters insurance?
You’re out of luck — car insurance will not cover theft of any item that be easily removed from the car (CDs, phone, aftermarket stereo, etc.). Renters and homeowners insurance are the only ways to cover these items.
Challenge #4 : Simplified Tax Filing
Disclaimer: This is not tax advice — just an effort to highlight the complexity faced by freelancers in dealing with taxation.
In the US, if you hire a freelancer directly, and pay him/her more than $600 during the course of a tax year, then the following happens:
- Freelancer needs to file a W9 form with you — this is so that you have the right information.
- At the end of the year, you need to issue the freelancer a filled out 1099-MISC form based on the W9 you received.
- You also need to file this 1099-MISC form with IRS (you can e-file this federally and to 34 states at the same time).
- The freelancer needs to use the income from the 1099 form as part of his/her 1040ES tax filings.
- The freelancer also needs to pay estimated tax quarterly if tax owing for the year/or for the prior year was more than $1000. There are penalties for not paying it quarterly in such cases then.
For most marketplaces utilizing third party payment providers to actually pay the contractors, like Uber or Lyft, they qualify as a “Third-party Settlement Organization”, which enables them to file only 1099-K form for contractors who were paid out more than $20,000 AND had more than 200 transactions during the course of the year (along with providing it to the contractor). They don’t file 1099-MISC for regular driver payouts above $600/yr — saving a lot of overhead. In the case of referral bonus and other miscellaneous payments, a 1099-MISC may be issued (as Lyft does). Payable explains this well over here.
A service like TaskRabbit does not send any tax form at all, instead letting their payment provider Braintree Payments deal with all tax reporting. This is Upwork’s tax info center.
From freelancers point of view, regardless of if they are issued a 1099 or not, ANY income earned must be reported, otherwise they can face penalties.
This then requires the freelancer to accurately and diligently track itemized list of all activity over the course of the year and earning associated with each. Most marketplaces provide a dashboard where this information is available. For non-marketplace based freelancing, the freelancer could use web-based bookkeeping software solutions such as that offered by FreshBooks and Bench.
For tax deduction purposes, tracking business expenses is very important as well: mileage driven, mobile phone use, stationary, business cards, etc.
This is a high burden on freelancers — complex tax requirements and the need to accurately and regularly track all activity which led to the generation of the income and associated expenses. Failure to accurately report income risks penalties from the IRS.
But as is hopefully clear from the above, for both customers and freelancers, it reduces the administrative burden a bit by using a marketplace to transact instead of transacting directly. Customer doesn’t have to issue 1099-MISC for work over $600 during the year; and the freelancer can get itemized list from the marketplace instead of having to maintain it.
How can this be made 10x better for freelancers/independent contractors ?
Householder worker earns > $2000/yr — then considered your employee by IRS with social security and medicare tax due
IRS $2000/annual earnings for a home worker makes him/her your employee with Household employees:
Do You Have a Household Employee?
You have a household employee if you hired someone to do household work and that worker is your employee. The worker is your employee if you can control not only what work is done, but how it is done. If the worker is your employee, it doesn’t matter whether the work is full time or part time or that you hired the worker through an agency or from a list provided by an agency or association. It also doesn’t matter whether you pay the worker on an hourly, daily, or weekly basis, or by the job.
Example.
You pay Betty Shore to babysit your child and do light housework 4 days a week in your home. Betty follows your specific instructions about household and child care duties. You provide the household equipment and supplies that Betty needs to do her work. Betty is your household employee.
According to the Internal Revenue Service, a household employee is defined as an individual you’ve hired to work in or around your home and who works under your direction. The most common examples of household employees include a cleaning person, maid, nanny, or driver. Babysitters, home-health aides, and private nurses also generally fall under this category. If you manage the work and how it gets done, the worker is your employee.http://www.paychex.com/articles/finance/household-employer-tax-guide
You are responsible for paying taxes on any employee who meets these criteria and who earns wages of $2000 or more per year. In the eyes of the IRS, this is true regardless of whether the employee works part-time or full-time, lives outside the home, or was hired through a staffing agency or some other job-placement firm.
You may be responsible for any or all of the following:
- Payment of social security, Medicare and federal unemployment insurance
- Withholding (at the request of the employee) state and federal income tax
- Withholding state income tax (depending on your state regulations) and payment of disability and
workers’ compensation insurance - Filing paperwork with appropriate government agencies
If you pay your household employee cash wages of $2,000 or more in 2016, all cash wages you pay to that employee in 2016 (regardless of when the wages were earned) up to $118,500 are social security wages and all cash wages are Medicare wages. However, any noncash wages you pay don’t count as social security and Medicare wages.
If you pay the employee less than $2,000 in cash wages in 2016, none of the wages you pay the employee are social security or Medicare wages and neither you nor your employee will owe social security or Medicare tax on those wages.
Challenge #5: Help Desk for 100 Million People
If by 2030, there are over a 100 million people making themselves available as part of the on-demand economy, and they have questions concerning their income, benefits, taxes, or regulations — how do they get serviced ?
Brands which are loved by the workers are seeing rise of online peer groups (such as this one uberpeople.net) where members provide quality advice to each other, for free, moderated by peer-appointed group leaders. That is the only scalable way a customer support function works here. What we are seeing here are initial versions — this group has close to 60,000 members, with over 1 million messages posted already. How will this scale 1000x times and still provide value to its members ?
Challenge #6: Laws & Worker Safety Net
Future US elections would be contested with on-demand/sharing economy being a key pillar of the debate. It will no longer be a niche/emerging trend, but instead it will be the dominant labor market reality at the time. There would be several efficient labor marketplaces. There would be tangential startups servicing this workforce. Full-time, salaried jobs with one employer would be a small minority. With a significant proportion of the population participating in this new economy, regulations and laws will evolve to make the life of on-demand workers better.
Forward-thinking politicians like Senator Warner are already on point here:
“This next generation, where they are in the ‘sharing economy,’ the Millennials, 80 million strong, they have no safety net at all: no unemployment, no workman’s comp, no disability,” Sen. Mark Warner, D-Va., told Capital Download. “Somebody may be doing very, very well as an Etsy seller and Airbnb user and Uber driver and part-time consultant … but if they hit a rough patch, they have nothing to stop them until they fall, frankly, back upon government assistance programs.”
And here is another speech by Sen. Warner raising some interesting points for government of the 21st century to consider — key excerpts below:
Today, nearly 80 million American millennials are flooding into the workforce. And whether by choice or economic necessity, many in this generation are piecing together two, three or more on-demand job opportunities, often with no benefits at all
To be sure, the sharing economy is not just comprised of millions of millennials who began entering the workforce in 2000. As many as one-third of American workers now moonlight or have a second job. The sharing economy also includes middle-aged professionals downsized at mid-career, and Baby Boomers hit with a premature end to what they thought was a solid career.
..these new business models are based upon the premise that workers are independent contractors, not employees. That means employers do not help pay the costs of health insurance, retirement, unemployment or workers’ compensation insurance.
That means many workers, even if they are doing relatively well, exist day-to-day on a high wire with no safety net underneath them. That may work for many of them — until it doesn’t. And that’s when the taxpayers likely will be asked to pay the bill.
First, the biggest challenge may be the change in the employer-employee relationship. Are there other options for providing safety net benefits for workers who are not connected to a traditional, full-time employer? Who should administer it? Should it be opt-in or opt-out?
Second, while litigation is underway about whether on-demand workers are independent contractors or employees, this is too important to leave to the courts on a case-by-case, state-by-state basis.
..we also should not overlook the impact of the millennial generation’s combined $1.2 trillion in student loan debt: It is limiting options, opportunities and economic mobility for too many young people.
.. millennials are demonstrating a preference for renting and sharing instead of owning things such as vehicles and houses. That could have huge ramifications for local, state and federal tax structures designed to incentivize ownership.
Economic Policy Institute takes an opposing view point and puts forth the case that Uber drivers are employees, and not independent contractors, and that there is no need for an “independent worker” classification. Worth noting that one of the authors, Ross Eisenbrey, is a former commissioner of the U.S. Occupational Safety and Health Review Commission.
Hillary Clinton has had tough words for the sharing economy: “She “vows to crack down on employers who misclassify workers as independent contractors”, which she says is “wage theft”. Clinton also said that benefits, paid sick leave, and maternity leave are essential to strengthening the middle class. Those are things independent contractors don’t get.”
And Donald Trump has vowed to repeal Obamacare.
Voters will decide and elect US congressmen and women who are in line with their aspirations to legislate accordingly.
Yelp: Where People Turn For Reviews
Step #1: Search for house cleaners in san jose
Step #2: Go through the listings and choose one
Step#3: Click on the provider’s website and then call/email for a booking!
Google: Over 3 billion search queries a day
Below are some screenshots from a new Google experiment in the SF area. Home service professionals can buy ads for premium placements on the results page; Google then allows requesting quotes from 2–3 such pre-qualified providers. User is able to indicate up to three preferred time slots and the request for quote is sent.
Since the fall of 2013, Google has set up an App Indexing effort to encompass data inside mobile apps into its general index. Fifteen percent searches from signed-in Android users now yield results with information inside apps. Apps indexing, though, does not currently include iOS apps, a serious gap…
But for now, the challenge of crawling the apps universe hasn’t affected Google’s search dominance. The statistics remain staggering. Google accepts over 3 billion search queries a day. In the US, two-thirds of all searches use Google — worldwide, there’s similar dominance. (A recent dip in market share is largely attributed not to search quality , but Yahoo’s deal to dislodge Google as the default search engine on Firefox). Even more impressive, Google hosts well over 80% of mobile searches. When Google suffered a five-minute outage in 2013, global web traffic dropped forty percent.
No search competitor has Google’s infrastructure, its deep talent, or its experience. Few have its ambition.
Facebook: 1 billion daily active users on mobile
Approximately 84.5% of our daily active users are outside the US and Canada
Daily active users (DAUs) — DAUs were 1.13 billion on average for June 2016, an increase of 17% year-over-year.
Mobile DAUs — Mobile DAUs were 1.03 billion on average for June 2016, an increase of 22% year-over-year.
Monthly active users (MAUs) — MAUs were 1.71 billion as of June 30, 2016, an increase of 15% year-over-year.
Mobile MAUs — Mobile MAUs were 1.57 billion as of June 30, 2016, an increase of 20% year-over-year.
https://s21.q4cdn.com/399680738/files/doc_presentations/FB-Q216-Earnings-Slides.pdf
Fiverr has a very simple concept. People post small jobs, called “gigs,” that they are willing to do for just $5. It sounds small, but what it really provides is a platform to grow from. Buyers like Fiverr because they can get small tasks done quickly and cheaply, there isn’t a large risk and over time they can find dependable freelancers they can turn to for quality work. Sellers can grow their business by offering basic gigs for $5 and offering upgrades and extras for additional money.
http://launchastartup.com/how-to-hire-online-freelancers-without-losing-sanity/
https://www.toptal.com/resume/luis-martinho
or delivery entirely offline — personally delivered; location matters
mechanical turk, spare5
autobreak down of big tasks into smaller ones
contract: http://www.docracy.com/3152/designer-sample-contract
proposals: https://www.proposify.biz/proposal-templates/e-commerce-proposal-template
Quora question: “How does a business person hire a good developer/programmer/engineer on Freelancer or Upwork?“
Many answers echoed the same sentiments, here are a few highlights:
Yishan Wong, CEO of Reddit answers:
You shouldn’t do this; it will probably result in failure.
I have a friend who is a designer (so, closer to technology and implementation than a business guy; about as close as you can be without being outright technical yourself), and he was hiring developers via eLance [now Upwork]. Even with consultation from friends of his (e.g. me) who were real engineers, it was extremely difficult to find decent engineers who could do the things he needed, deliver reliably, and iterate according to ongoing testing/customer feedback. The end product was merely “okay” — kind of slow, with little glitches here and there.
If you have total technical ignorance and no local (friend) resources to help you, hiring from eLance or oDesk [now Upwork] is almost impossible to do correctly. I would recommend trying another route.
Mircea Goia, another Quora power user adds:
I second what Yishan says…my biz partner, being a business guy and having some ideas in mind, took the eLance [now Upwork] route…lost some money, got some bad days…this mostly with Indian developers (he is trying now Russians).
It’s very hard to find competent AND reliable ones (even if they have 5 stars and lots of projects on eLance — maybe those who gave them projects have low standards?). Reliability and work integrity matters a lot when the developer is 10,000 miles away.
Only 4% of Americans have used an online service to get errands done
According to a recent exhaustive report by Pew Research about the sharing/on-demand economy:
- 4% of American adults have used an online service to get errands done (TaskRabbit, Fiverr, Amazon Mechanical Turk)
- 6% have ordered groceries online (Instacart, Fresh Direct)
- 11% have used home sharing services (Airbnb, HomeAway)
- 15% have used online ride-hailing apps (Uber, Lyft)
- 73% have not heard the term “sharing economy”, and 89% have not heard the term “gig economy”
- ..while, 50% have purchased used/second-hand goods online (eg Craigslist, eBay)
Even more fascinating is this matrix from that report, which points out that even amongst the 50% people who have in the past bought a used good online, only 6% have bothered to use an online task marketplace or, 8% have used an online grocery delivery service or 16% have used a ride-hailing app. By and large, even though awareness about these is there, a vast majority of people are yet to use these services.
Rising trend is for on-demand startups to shift from 1099 to W2 model
Hundreds of startups that rely on freelancers to clean houses, run errands, deliver food and ferry people are under siege.
Lawsuits are threatening to reclassify their contractors as employees, which could drive up labor costs an estimated 20% to 40%. That has put an investment chill over these startups and prompted some founders to switch their business models.
Entrepreneurs and venture capitalists alike developed a fondness for “1099 workers,” referring to the tax form they file, because they don’t require costly outlays such as health insurance, payroll taxes, travel reimbursement, training programs or severance pay.
Marcela Sapone, CEO of New York-based Alfred Club, which goes by Hello Alfred, estimates her household-task company pays at least 20% more on labor costs than when the company had contractors in its early days.
Handy Technologies Inc., an on-demand home-services provider, sought in its user terms of service to pass along workforce liabilities to its customers. If Handy was found liable for any back taxes on its workforce, the provision read, the customers “will immediately reimburse and pay to Handy.com an equivalent amount, including any interest or penalties.”
After an inquiry from The Wall Street Journal, Handy removed the provision.
In a post-Uber world, one of the defining characteristics of technology companies that use software to more efficiently allocate service workers is a two-tiered labor structure.
One part contains the engineers, executives and product managers that build the platform, and the other part contains a much larger mass of contractors. The first part is compensated in stock, which could be worth nothing or a lot, depending on the outcome of the company. The second part is paid by the gig or the hour, and receives no benefits, workers’ comp or unemployment insurance.
Increasingly, due to the threat of lawsuits from contractors and poor service quality, some startups have pivoted away from using a 1099 independent contractor workforce and instead have hired their workers as W2 employees. This enables them to better control user and worker experience, while paying them benefits and a salary. Some examples here include:
- Task/errand space: BlueCrew and Alfred
- Elder care: Honor
- Online grocery delivery: Instacart
- Logistics: Shyp
- Parking: Luxe
- Food delivery: Munchery, Sprig
- B2B cleaning service: Managed by Q
In New York, for example, Q cleaners start off at $10 an hour and move up to $12.50 after a training period. (Rates vary slightly in each market, with San Francisco cleaners starting at $15 an hour.) Cleaners can become mentors, earning $14, and then advance to quality-control supervisors at an annual salary of $35,000. Managed by Q charges its clients $25 per hour of service by a field worker. New York City’s largest property services workers’ union, 32BJ SEIU, uses an independent contractor agreement with a minimum handyperson wage rate of $26.198 an hour for Class A office buildings.
Except a few, on-demand labor startups are losing momentum and dying
https://www.cbinsights.com/on-demand-overview
According to this CB Insights analysis (some charts from that article referenced below again), there has been over $10 billion invested in on-demand startups since 2010. While majority of this has been Uber, $4 billion+ has been invested in non-Uber, other on-demand startups.
Seed activity peaked in 2010, with majority of the funding in 2015 going to later stage startups in this space, indicating that the “biggest consumer on-demand startups are already funded and in existence”. But — why do such a small % of American consumers use them ?
Despite all this activity, there have been fewer than 10 exits till date in this space, with a lot more startups dying/losing momentum.
CB Insights points out the following startups haven’t raised further funding for a while and might be primed for a move (from this list, Homejoy and Spoonrocket have already shut down; FlyCleaners has had terrible reviews after losing clothes of customers in dry-cleaner runs; TaskRabbit went through a recent CEO change):
Users don’t care about your on-demand app; and when they do, you get ONE chance
Working on a shiny new on-demand app ? Great. There are two problems:
- Users are unlikely to download your app no matter how great it is. 65.5% users in the period below downloaded a grand total of 0 apps. Social and entertainment apps consume their focus.
2. If they do download it, 80% of them will abandon it within first 30 days.
If you don’t have sufficient liquidity in the marketplace the first time they try to use it, they are not coming back.
But Uber and Lyft are doing great. That is after spending billions of dollars on user acquisition.
New buy and sell marketplaces taking on Craigslist
Tools and services for independent workers
As the chart below shows, these are very early thought experiments in this space.
Startups servicing on-demand workers sound like a great idea..except Zen99 tried to do that very thing and failed. They wanted to provide taxation and health insurance advice and simplify the life of the 1099 contractor. What they found out was that the 1099 contractor did not have enough focus on these matters until when deadlines/issues hit them.
But a web developer’s wants and needs are very different from an Uber driver’s. Trying to service too many different groups leads to a product that isn’t a magical experience for anyone. Trying to service too few of them can lead to too small a business. We made the former mistake while trying to avoid the latter.
.. For example, wouldn’t it be great if we could pool individual contractors together to provide group discounts on health insurance? Indeed it would, but the Affordable Care Act made this illegal starting in 2014 (see Freelancers Union shutting down their insurance product).
We also explored recreating the W-2 experience, by acting as an intermediary to help automatically save for taxes, retirement, and other items. But intermediary tools like these fall under money transmitter laws, which are some of the most stringent laws out there. Proper compliance is expensive
In both cases, regulations meant we couldn’t build the products we wanted.
Here is what freelancers and on-demand workers primarily care about: earning more and regularly.
crowded.com
qwil.co: Instant pay, $1.2 trillion locked up in 30/60/90 day invoice cycles — helps workers get paid sooner. Qwil debit card — instant cash out.
payable.com
Bonsai charges $1/invoice; claims used by over 20,000 freelancers
Recap of key points
- 53 million people are already freelancers in the US, facilitating $715 billion per yr. 34% of the US workforce is already a freelancer, equating to about 53 million people. This is expected to grow to 60 million, or 40% of the workforce by 2020 (Intuit study). Is it inconceivable to project this to at least a 100 million independent workers in the US alone by 2030 ?
- Advances in AI/ML/increase in computing power will limit the jobs available to independent workers.
- 92 million 15–35 yr olds in the US — known as the millennials generation (out of which 38% are freelancing => 35 million millennials are freelancing).
- Over 65% of users don’t download any new apps in a month, and of those that do, over 80% abandon it within the first month => your app breaking out are very slim. Android, on Samsung devices, leads iOS.
- From a funding and M&A perspective — the on-demand labor startups are characterized by a lot of promise but significant lack of M&A. The big heavy-hitters Uber and Lyft skew the fundraising. CB Insights data tracks that (included above).
- Real-world connections are the primary way freelancers find work. Online job boards and classifieds sites, especially Craigslist are also increasingly important, but the bulk of the freelancer-customer interaction has not yet moved to an efficient online labor marketplace.
- Efficient, worker-friendly models will prevail. Current worker classification system is not working. Current system requires the workers to be either be employees (W2) or independent contractors (1099). There is no third category which exists as yet. For startups utilizing independent contractors (1099), there is the threat of worker classification lawsuits and potential of back-dated taxes by IRS.
- Business models of current online freelancer marketplaces involve taking 10%-30% of the worker’s pay as commission.
- Almost everybody has a smartphone, but it is incredibly difficult for new apps to crack through the noise of other apps and user attention span. Stats: 92% of 18–35 yr olds, and 65% of 35+ yr olds in US have a smartphone. 60% Android; 40% iOS. But be careful with that, over 65% people don’t download any new app in a month; and within 3 days the app which is downloaded loses 80% of its daily active users. People spend most of their time on social and entertainment apps.
- Craigslist is the dominant US online marketplace by far, with over 60 million classified ads posted each month, and over 700 million monthly visits. Key insight: How do you make your product work with Craigslist, instead of engaging in a zero-sum game of trying to replace it altogether ?
- Uber is the most efficient labor marketplace — although currently for ridesharing and deliveries only. 15% of the US population has used a ridesharing service. Other labor marketplaces: Upwork — Dominant global freelancer website, over $1 billion facilitated. Primarily for technical work which is done online. Thumbtack: Innovative pay-per-quote business model. Great for finding that first home service provider job, with over $1 billion facilitated. Fiverr: growing. AngiesList: On the downward trend — this used to be the go to site in the US for reviews of home service providers. TaskRabbit is effectively dead. With such high promise and media attention, and extremely low usage numbers, TaskRabbit’s era is over. Less than 4% of US population has used an online task service. Craigslist has 1,000 times monthly active visits than TaskRabbit does. Handy is effectively dead. High promise, and low usage numbers, Handy’s promise has failed to materialize as yet. Fundamentals of the business model which afflicted Homejoy would hurt Handy as well.
- Primary reason why “Uber for X” startups have and will continue to fail: failure to understand relationship-building for services which cannot fundamentally be commoditized. Marketplaces cannot commoditize workers where individual relationship is important between the supplier and the customer — this involves all at-home services. Ridesharing and deliveries — commoditizing workers can work if you are able to scale. Multiple startups in the at-home services space continue to fail/lose momentum because they thought they could be “Uber for X”.
- Primary concern for freelancers is income.
- Worker-focused marketplaces make life better from a tax point of view for freelancers Taxation is a hassle for independent contractors and being on a marketplace which can document the job activity, expenses and issue a 1099-MISC or a 1099-K when needed, helps. Marketplaces alleviate the tax burden on the consumer/business as well who need to employ such workers. Except when/if more than $2000/yr is paid to a home worker, who then gets classified as the homeowner’s domestic employee — making those wages medicare and social security wages with more paperwork for all involved.
- Critical gaps in on-the-job insurance for such workers — falling through the cracks.
- Obamacare/Affordable Care Act has fundamentally enabled the gig economy by providing the ability for freelancers to choose their health insurance and get subsidies on that.
- Peer-to-peer support for workers: The only way a 100-million-worker-scale HR would work is by a peer-to-peer support model, where passionate enthusiasts of a labor marketplace take it upon themselves to share their wisdom/best practices with newer entrants. Hiring a customer service/HR department at a massive scale has to be impractical/costly. The biggest such forum currently only has 60k members (Uberpeople). Facebook Groups and other online localized groups are currently how workers are organizing themselves. Uber and Lyft drivers have been self-organizing themselves based on their city. Workers need to share their stories.
- Worker Classification: Shift to W2 employees for Instacart and others. On-demand worker classification lawsuits. Settlement by Lyft and Uber — startups need to be mindful of what constitutes an employee.
- Increasing experiments being seen to provide tools and services to independent workers.
