Mistakes, more mistakes, and a couple of smart moves as a first-time founder
After about fourteen months of mistakes and lessons learned the hard way, my team is about to launch a new service that connects software product managers with target users in their potential markets to help them validate that they’re building the right things. The target users charge an hourly rate for their advisory services.
The purpose: to help product managers validate that they’re building things that the market really wants and that solve real problems for them. We want to make it so easy to get market validation that enterprise software PMs have no excuse not to do it. At the same time, we want to give everyday experts a way to monetize the expertise they have with tools they use at work. As a 10-year PM myself, it’s a business venture as well as a passion project for me.
I decided to write this piece because, I’ve made plenty of mistakes and a handful of good moves along our journey that can benefit others. This isn’t a story about runaway success — not yet at least. I’m not writing this to brag, as will be clear. I benefited from contributions of many other founders to the entrepreneur community here and elsewhere, and I wanted to pay it forward.
Longer read, 20–30 minutes. Hope you get value out of it.
The Idea
A few years ago, I asked my then boss at a VC-backed startup what he thought made for the best product managers. He replied that in his experience, the best indicator of success was how much time they spent talking to the market. I’ve found this to be true too, but it’s easier said than done for many of us.
Getting out and talking to the market takes both hustle and contacts. In my experience, most PMs aren’t naturally good at the rolodex-building part of the job. They may be great at writing user stories, or thinking through the tech, but most aren’t great at getting out there and engaging the 90% of their potential market that hasn’t entered the sales funnel yet. That means they aren’t learning how the market is surviving without their product today. And they aren’t learning about the problems those people really experience. Despite this lack of validation, they go on building products and marching out new features. No wonder the #1 reason startups fail is that people don’t need their product.
After switching into ad tech from another industry, I was “network-poor.” I desperately wished for a database I could query to find, say, “10 Adobe Audience Manager users in marketing roles at e-commerce companies,” or “twelve Optimizely users currently working for online publishers,” so that I could learn about how they did their jobs, what was hard or tedious, how their existing tools fit their needs, and so on. Making friends at conferences was slow-going and really expensive, so I finally started cold emailing people who looked like good prospects and paying them $100 or more for an hour of their time to talk about exactly those things, and lo and behold it worked.
Most people like sharing their experiences and expertise, I’ve found. They especially like being paid to do so. The big problem was how long it took me to find the right people on LinkedIn or some other network, reach out to them, wait for responses, and schedule meetings. Panel recruitment was practically a full-time job, and a tedious one at that.
It dawned on me that product managers might be willing to pay for a service that made this easier. After chewing on the idea for months, what became UserMuse was born. If there’s one thing the last ten years have shown us, it’s that no matter what you do, someone out there is trying to make an app to help you do it. Why not make some money and help them do it right?
BACKGROUND RESEARCH
Skipping ahead a bit in my iteration on the idea, I decided that the value proposition stemmed from three things: * collecting product reviews from users * the “social graph” we’d accumulate of where enterprise software was being used (our primary focus) * facilitating connections between people. Now it was time to see who else had already thought of this. First came the software review vendors. There are analyst services like Gartner, Forrester, and others along with numerous crowd-sourced software review sites which vary considerably in quality.
My POV was that reviews of products in a vacuum are of limited value without knowing how an organization uses them, whether they’re happy with it, how widely deployed the tool is, and so on. I also think that vendors get happy customers to write reviews on those sites — there should be waayyy more horror stories out there given some of the disastrous rollouts I’ve seen. So there was and is room in this space to be different and provide value. But I didn’t think there was enough meat left on the bone to build a business on that alone.
Second, the “social graph of enterprise software” concept. This was indeed being attempted by several companies (and over the course of building our product, I would find out about more). That was disheartening, and like anyone else I wished my competitors who had a big head start didn’t exist. But after digging deeper into these companies, their public statements who their customers were, I felt better. We each saw the market opportunity differently and had different strategies for monetizing. I reasoned that the competitors’ services and ours would likely evolve in different directions, which left plenty of room for us to be successful.
Lastly, the type of person-to-person networking I wanted to facilitate, for the reasons I wanted to facilitate, didn’t seem to exist. I had been using tools to cold-email people I wanted to talk to with some success, but I couldn’t find anything better. I was reasonably confident about this aspect of the value proposition not being too exploited yet. I eventually decided it needed to be a more prominent focus — more on that in a moment.
(A quick note on evaluating competitors: Competition is less scary than it seems. When you think about it, it’s kind of surprising that people are still launching professional social networks and recruiting tools given how dominant LinkedIn has become in that space. Yet look at LinkedIn’s acquisitions, and you’ll see that competitors are still popping up. Yet if I tried to pitch my friends on a new professional social network, around 100% of them would say, “So your plan is to out-LinkedIn LinkedIn? Good luck.” They’d be wrong though; people find unexploited niches all the time even in the face of dominant players. Glassdoor, to pick another example, has become the place where people go to look up salaries for various jobs. It’s one of the most viewed websites in America. And yet, you still have a tool like Comparably launch just last year. So yes, be aware of the competition, but the existence of competitors isn’t a reason to not to go into business.)
GETTING A TECHNICAL CO-FOUNDER
I spent a while on my own with my idea — thinking about the business model, why would work and why it wouldn’t. I made chicken scratch wireframes to think through the details. Lots of crumpled-up pages later, I felt I had a solid idea for a business as well as a product. It was time to get help.
I’m not a developer, so I needed a technical co-founder. I made a list of every engineer and technical manager I’d ever worked with to define my network. I had about thirty names and then ranked them. I wanted people who were in the DC area, where I live, and with whom I’d had an especially good rapport in the past. Full-stack developers were ideal, but I prioritized their fit with me and their general smarts above everything else. I soon realized that the list of people with whom I thought I could start a successful business was short — just three.
I approached my #1 target and gave him my pitch, but it wasn’t a good time for him. He thought it was a “great idea” though, which is a little like being turned down by a girl for the prom but then hearing her say that “you’re a really nice guy.” My #2 target had just started in a new role at another startup when I pitched him. #3 was always a long shot and in the midst of starting his own company as it turned out. Three swings and misses may not sound like much, but pitching somebody on starting a business with you and laying out your vision is a really personal experience. You put ideas you’ve invested a lot in emotionally already out there, along with your enthusiasm for the business and whatever charisma you possess. Having all that rejected, however politely, stings.
I was starting to look at co-founder speed dating sites and thinking about outsourcing (ugh), when I remembered that I’d forgotten about another perfect candidate. This was mainly because he’d moved out of the area a few years before to take a great new job (I’d been a reference for him). He was whip-smart though and we’d worked well together in the past. The distance wasn’t ideal and nearly every piece of advice I’ve read says founders need to be co-located. But what if the alternative is a stranger, or no business at all? I said screw it and reached out to see if he was open to working on something new.
The guy happened to be in town for like 18 hours on his way back home from South Africa, with a layover coincidentally in DC. I gave him the pitch over breakfast at the airport at 6 AM before his flight, and a few days later he signed on. Did I commit a cardinal sin that doomed my company at that moment? Who the hell knows. What I do know is that working with people I like, respect and trust makes me eager to work. And if I’m not eager to work, the business dies anyway.
One of the things I learned while looking for a co-founder is that from the first moment you open your mouth about it until the moment the business either has a successful exit, dies, or fires you, you are in sales mode. Not to sound craven, but with just about everyone you meet you’re trying to get their attention, talents, or money — and sometimes more than one of those. Just the way it is.
One other lesson to share here: No matter how uncomfortable it may be, have the conversation up front with your co-founder about who owns what, and how you’ll settle disputes. Obviously, the point is to work together, but disagreements will arise. Here’s an excerpt of the email I sent my new co-founder on this point, and it did the job:
…I see myself as the final owner of product strategy and marketing, and I see you as the final owner of technology implementation and app design decisions. In practical terms, it means that when we need to make a decision on, say, which product categories to target first, it’s my call if we can’t agree. Likewise, when we face a choice between rolling out something fast and cutting architectural corners versus implementing it in a way that will save time down the road, it’s your call if we disagree. If we disagree on the angle for marketing direction, I make the final call. If we disagree on how a new feature should be implemented, your call…
However you do it, have this conversation. It will save you time and frustration later.
PART-TIMING IT
The whole nights-and-weekends thing is something I had some practice at. I’d been writing a blog for a while and prior to that wrote a book that together took up a big chunk of my free time. And at the places I’ve worked, my experience has been that most engineers have their side projects too (the means of production are in their hands, after all). So while burning the candle at both ends can be draining, the enthusiasm at first makes up for a lot. One more reason why co-founder fit matters so much.
We started by reviewing and refining all of my ideas about the business, then tore down and rebuilt the wireframes and workflows, one painstaking micro-feature at a time. As our expectations around data volumes and UX crystallized, we started making technology decisions. I wished to God we could have knocked it all out in one month rather than several, but we weren’t in a position to abandon our paychecks. Nights and weekends it remained.
Like a good partner, he pushed my thinking in lots of areas where I needed it and challenged my assumptions about user behavior. I ran some more surveys to test some of my assumptions about user uptake (Google surveys are great for this) and adjusted certain features as a result. We were making real progress, but this phase was just going too slow.
A few things stand out to me about this period (which continue to this day):
- I found myself sympathizing with my former bosses and CEOs quickly. Not being the CEO gives you the luxury of insisting on proper methodology. As the CEO though, I have to fight the urge to take shortcuts so that I can see new features. If you’re in this situation, remember that progress can be illusory if you’re not thinking things through properly.
- Second the part-time phase is hard as a people-manager. When you’re all in full-time and your hands are taped to the wheel, motivation and management are relatively easy. But holding people accountable and pushing them harder when they’re already tired and facing yet another night with only a couple hours’ sleep is tough. This got a bit easier as we brought more people to work on the product, because they could encourage each other. With my co-founder, not being in person to lift each others’ spirits probably took a toll here.
- Boosting spritis matters. Look for those milestones wherever you can. Getting our logo finished was hugely galvanizing, for example.
For us, the part-timing phase was necessary, but I wish it had gone faster. Had I accepted that having all the skills we needed on board more quickly would help, I think it would have been faster. I’m skipping some of the additional frustrations of finding the next developers to contribute, but thankfully I ended up with a team of competent people who are great in their lanes.
If I were to do this again knowing what I know now, I think this year-long period could have been shortened to five or six months, if not less, even with everyone going part-time. All of that is time we’ll never get back.
THE HALF-PIVOT
I had been going back and forth for a while on whether to emphasize the social graph aspect first or the P2P connections when it came to our product. I was having trouble telling a coherent story that explained both. Then, well into the design and development of our MVP, I came across more competitors who were attacking the “social graph for enterprise software” angle and doing it well. That should have compelled me to acknowledge what I already knew: we needed to get the networking aspect of the product right first. Everything else would flow from that. But it didn’t.
I was actually on vacation with my family in Colorado when I woke up one morning and the answer was suddenly clear. It was like my subconscious interrupted my thought stream to say, “What the fuck are you waiting for? Make the call.” So I did. And it allowed me to downsize our MVP product to make it easier to get to market. It also made things much simpler from a prioritization and messaging perspective. The elevator pitch was now crisp and clear.
The downside? My co-founder wasn’t as thrilled with the shift as I was. Probably for unrelated reasons (I think), he made the decision to take a new position with his company in London. This time though, I had a replacement waiting in the wings, and reached out the day after I head about my co-founder’s decision to move to London. Maybe my pitch has gotten better, but this time it only took a round of drinks to convince this guy to start working with us. I was as relieved as I was happy.
My co-founder potentially leaving raised some questions though. Not the least of which being, what happens to the company structure and the equity allocation with one of the original team members leaving? Thankfully, we had an answer to this, something I honestly believe made the company possible.
DIVVYING UP THE PIE
I’m going out of order here a little, but I spent a crazy amount of time figuring out how the hell to organize the company and divide the equity amongst the early team. I can’t tell you how many nights I spent reading other entrepreneurs’ experiences on Reddit, Quora and other content forums. Laugh all you want, but I have no idea how many hours I spent researching and thinking about this. There may be no single right way to divide up the pie, but there are lots of ways you can doom your company and poison relationships with your team by fucking up the cap table from the start. No pressure!
Naturally, as soon as my co-founder joined, I fucked up and verbally committed way more equity to my co-founder than made sense given our responsibilities. Maybe it was the elation of finally having a partner, but it was bad judgment, and several of the people I asked for input on this weren’t shy about telling me how badly I’d botched it. As graciously as I could, I talked to my co-founder about our expectations for contributions, when we wanted to go full-time, how much cash we could put in, and decided that a different split made sense. It was embarrassing, and disproportionately time consuming. Relationships matter a lot to me, and I didn’t want him to think I was trying to burn him.
I played with different startup equity calculators you can find online to figure out what the split should be. I asked people who had been in this position. I thought hard. I drank a lot.
The main challenge is this: The early life of a startup is volatile. You don’t know yet know what the product will be, much less what the business will be, which means you can’t possibly know how much value everyone involved will contribute to the final outcome. You might go along for months and then realize that the best thing for the business is to take the product in a totally new direction. Certain new skills might be needed while others are now obsolete. Adding to the volatility is the fact that people are…people. They change their minds. They can lose interest in things they were once passionate about. Their lives change as they get married, have kids, etc. You don’t want to tie up equity in people with irrelevant skills or who are no longer interested in the business, but that’s the risk you take when bringing people on board. Allocating equity in fixed percentages to early employees involves predicting the future. And I suck at that as much as the next guy or girl.
I spent months thinking about how to divide up the company in a way that was both fair and protected us from the risk of people leaving with big chunks of the company in hand. My team was unbelievably patient while I did this. Finally around 11 pm one night I was on my laptop reading something and I came across an ad for the book Slicing Pie by Mike Moyer. It’s about exactly this predicament, and read it that same night (thanks Kindle store!). I hate to call anything a silver bullet, but…it resolved like every issue I was wrestling with. I won’t relay the methodology here, but the gist is that everyone on the team has an hourly “rate” and tracks their hours spent working as well as cash they put in. You add up everyone’s contribution based on their (hours worked) * (hourly rate) to determine their share, and this adjusts dynamically over time based on everyone’s contributions. Of key importance is that the model has a built in way to handle people leaving that doesn’t involve them leaving with a big slug of equity.
Because the framework is rooted in fairness, it’s hard for people to really object to it. I’ve actually been happy to see as we’ve gone along how the other team members’ share of the pie has risen (and yes, their gain comes out of my hide). Everyone has even more incentive to bust their asses with the dynamic approach. Also, I reached out to Mike Moyer when I had some specific questions and he responded with answers quickly. I had our lawyer read his book too (on his time), and we baked it into the lightweight employee agreements we drafted for everyone.
I’ll write a follow-up someday about the transition from this model to a more conventional split when we get there. But for this weird, volatile early startup stage, the slicing pie model was a godsend. I’d love to hear from others who have tried this.
RESEARCH PART 2
To this point I’d run several surveys validating that the people we needed to attract to our service would be interested in it. I wanted more though, and by random chance, my mother-in-law (a long retired teacher) gave me a book she thought I’d like called “Ask” by Ryan Levesque. I especially love the first survey methodology he describes for learning about your audience. The first question you pose to your target market is the most important and the simplest. It’s also open-ended, which is key. We asked product managers:
What’s your #1 single biggest product management challenge right now?
Replace “product management” with whatever your concern is to do this for yourself. I wrote the survey on SurveyGizmo and promoted it on my blog. I also ran a LinkedIn ad targeted to product managers linking them to the survey to get more responses. LinkedIn ads are kind of expensive, whether CPM or CPC, but the data was worth it. It also validated what I believed all along. Most product managers said their #1 challenge was not having time to talk to the market and do the proper market research to validate their roadmaps — and they said this unprompted no less.
Here’s a tactic of my own that I’d recommend. Sangeet Choudhary has a saying about how you don’t need to put your product in the New York Times, you just need to be in the New York Times of your niche. I wanted to make a little splash in the PM community with our findings. I decided that the Mind the Product blog would be a good place. I sent one of the editors an email offering to write up our findings in an exclusive post for their site if they would promote our survey to their readers to get us more responses. The editor agreed, and we got more responses as well as a built-in audience for the follow-up post in which we would disclose our findings (which of course would mention UserMuse).
The second post with our survey findings went live, and we immediately got our first several dozen sign-ups on our brand-new landing page we rolled out the day before the post went live. It put our name in one of the most respected blogs in our field — and thus in front of a lot of serious product managers who clearly have the problem we’d like to solve for them. It was one of my better moments this year.
Doing research is important, and if you can find a way to make the research double as a growth mechanism, that’s even better. And while it’s a little scary to publish your proprietary insights, I’d do it again in a heartbeat. Everyone has their proprietary ideas and insights. Most of the time, they don’t care as much about yours as you think.
EVANGELIZING
We all have our strengths, and believe it or not, writing is one of mine. At the very least, it’s something I do a lot of. I’ve written a modestly popular weekly blog called Smart Like How for a while that has given me another forum to promote UserMuse. Writing frequently also allowed me to build up relationships and slowly (with lots of false starts) learn the ropes of social publishing and building an audience.
The relationships and skills I’ve built through blogging also let me to do a portion of our PR without relying 100% on others to drive awareness. By becoming a contributor on sites like Elite Daily, Quartz, Fast Company and others, I’ve been able to push out content that mentions and/or promotes UserMuse to massive audiences. Having your own e-mail list is currency when you’re trying to jump start awareness.
Content marketing won’t be the only way we promote UserMuse, but it’s probably the way we’ll stick with for right now. Auren Hoffman, former CEO of LiveRamp, RapLeaf and others, wrote something I found really compelling about marketing. I’ll paste an excerpt here rather than try to paraphrase. It’s great advice.
Marketing is one of the toughest things to do well because there are just too many options. 95% of marketing professionals employ too many marketing tactics and essentially become adequate at many things and great at no things. In marketing, spoils go to the company that is great in a particular marketing tactic…If you try a tactic and it starts to work, focus all your marketing energies on just that one tactic until you become of the very best people in the world at employing that tactic. Then keep working on that tactic and hire more people around that tactic so you get better and better at it. Then, when the tactic is working great and you are known as one of the world leaders in that area of marketing … then do it EVEN MORE. Triple down on it…Even multi-billion dollar companies likely have under 5 marketing tactics that work.
Most importantly for me as a CEO, writing frequently throughout the last 14 months or so have forced me to become more comfortable with letting my personality show rather than trying to be some facsimile of a “serious” entrepreneur. Writing something like, “I paid a woman $175 an hour to talk to me about underwear, and it led to a new business venture.” is a good example of that. I’ve gotten better at being myself, and more people respond to authenticity.
My last piece of advice for this post: The best way for you to manage people or express yourself is to do so in ways that leverage your natural strengths. The most successful people I’ve come across have literally zero difference between their at-work personality and their everyday personality. They’re basically monetizing who they are. That’s where you want to be, no matter what business you’re in. Writing frequently and at such length kind of forced me to get comfortable being myself, because being someone else is too tiring over hundreds of pages. Do whatever forces you to be your authentic self.
Hopefully I’ll see you on UserMuse sometime. I’ll provide a postscript after we launch to talk about more things I didn’t know I needed to know. Happy to answer any reasonable questions as I can.
