Running a dev shop — part 2: Bootstrapping and specialization
This is the second part of a series on running a dev shop. See the overview of the posts below.
🎮 Part 1: Myths and favorable facts
🥾 Part 2: Bootstrapping and specialization
🎤 Part 3: Talent pipeline and employee life cycle
📣 Part 4: Attracting customers
📈 Part 5: Growth
Bootstrapping — fake it until you make it
One of the greatest things about dev houses is that you don’t need much to start one. A person with coding skills, basic communication skills and willingness to work is enough. A great way to start a dev house is by freelancing and building a personal brand to later migrate it towards a company brand.
Ethworks started as a small side hustle when a random person reached out to me to build a prototype for a start-up. It was based on a new technology, called Ethereum. The founder had tracked me down on AngelList, using the search keyword “blockchain”, which I had put there a while earlier and totally forgot.
One thing led to another, the prototype was created, and I got intrigued by this new technology I got to use. Without much hesitation, I bought a ticket for the upcoming conference on the topic. Many years later, my wife confessed she thought it was a crazy idea, to go to a random conference out of the blue.
Just a day before the conference, I created a website for a dev house specializing in the technology I was just starting to learn. Earlier, I had also extracted a small library from the project I worked for, and published it on GitHub. On the way to the airport, I printed myself some business cards.
At the conference, I met a guy who was looking for developers and became my first customer. The little experience I had and the library was enough to convince him, given that Ethereum was 8 months old and there were not many experienced developers in the field.
We set about it through and through, we brought about first prototypes and helped the customer raise $30 mln in the upcoming ICO bubble. The project didn’t turn out to be very successful in the long run, mainly due to business reasons, but it was a learning and forming experience, which laid foundations for our skill and the brand to become recognizable. Soon, we were to start working with some of the most prominent projects and best known brands in the exploding crypto/blockchain space.
This period led me to create another library for smart contract testing called Waffle, which is used today by over 70.000 projects. Waffle would help me tremendously grow the company as well as my personal brand.
👷♂️ Technical co-founder
Our Ethworks venture was launched by myself and my wife Natalia.
I was coding and soon hired the first talented student to become our first employee. Natalia, who is a designer, would soon hire the first designer to create the world’s first blockchain design team for hire. We didn’t have any business or operations person in the team, we were closer to 10 people.
In the early days, we tried to outsource all the work we could, however we did our own sales and marketing. Natalia would make sure her design work was published on the popular design website — Dribbble and that it was properly described and tagged around blockchain-related terms. Natalia’s portfolio brought us projects like the original bitcoin.org website, whose previous version had been designed by Satoshi Nakomoto — the creator of Bitcoin himself. When a call for such a designer was announced on Twitter, Natalia’s team was the only high-quality designer that was proposed with experience in blockchain projects. Natalia’s team later went on to design the world’s first layer 2 application, and did work for many leading brands in the industry.
I was attending conferences, building my social media presence and working on open-source projects. My network later led me to acquire customers like MakerDAO — world’s first Decentralized Finance project or WorldCoin — the Sam Altman’s idea mentioned in the previous blog post.
All of these would have rather been impossible if the company had been founded by fresh business graduates or MBAs. The best advertising strategy and sales techniques wouldn’t pay the bills as well as being part of the blockchain community.
In general, the full potential of a highly specialized dev house can be unleashed by a technical founder(s) who can pick the right specialization, drive the marketing and sales for the company as well as attract great talent.
Mediocre or average-motivated employees are unlikely to unleash full potential.
That’s why I believe dev houses to be the best type of business for engineers and specialists, who understand (or are willing to learn) their area of expertise and actively work in their specialization, and only then leverage those to build marketing and sales strategy.
The specialization
“If you can’t be first in a category,
set up a new category you can be first in.”
The 22 Immutable Laws of Marketing — Al Ries, Jack Trout
One of the most common mistakes dev shop founders make is to start out as a general service provider. Enthusiastic and bold, “Hello there, global software market. Here’s yet another web and/or mobile development agency!”
Instead, in the case of a small, low-budget undertaking with limited expertise potential and marketing resources, it is better to specialize. Choose a narrow, well-defined area to be broadened slowly and gradually (or not at all!) as the company grows. This is for a number of reasons.
🦹 Superpower
A good specialization is a superpower that allows a small, few-people company with a limited budget to do things that would normally be available only to big companies with much bigger resources. How so?
- 💶 Value for money. Specialization allows for delivering high value for reasonable money. By doing similar things over and over again, your team becomes efficient and is able to provide great expertise. It will take much more time and effort to figure out your way to do new things and do them right. If you do something thrice as fast as the competition and charge twice as much — it is a win-win situation, in which you earn more and the customer pays less.
- 📺 Perception. It’s not only about the real value which you want to deliver; it is also about how you want to be perceived. In the eyes of your clients, a narrow specialization makes it easy to present yourself as a go-to expert on a particular topic.
- 📣 Marketing. Addressing a small niche is cheap and easy, every blog post you write can be shared in a group with a specific audience. The SEO and ads competition is lower. The twitter and conference community — easy to identify and interact with.
- 👨👩👧👧 Community There is a lot of competition to solve problems, build products and open-source frameworks or libraries in a big market. However, small communities lack quality software and good practices. This exactly is an opportunity for a small dev shop to shine.
💰 Specialization as a high-margin business
At some point with Ethworks, we ran a small mystery shopper scheme to benchmark our services against other dev shops in Poland. To our surprise, it out that our rates were higher than anyone else’s on the market we reached, even though we didn’t really push for high rates in the first place and instead grew them organically from customer to customer.
I don’t need to tell you how it reflected on our profit margins.
Technological cycles
“We try to look for these technical vectors that have a future, and that are headed up, and, you know, different pieces of technology kind of go in cycles. They have their springs and summers, and autumns, and then they, you know, go to the graveyard of technology. And, so we try to pick the things that are in their springs.”
Steve Jobs
A highly disruptive technology in its early spring is by far the best specialization you can settle for. It will grow to become an industry on its own. It will give you time to learn and contribute to the community, when it is still small and hungry. Your company can grow organically together with the technology and community.
How to pick one? You have to make a bet.
👨💻 Web3
Today I would not start a Ruby or JavaScript dev shop. I wouldn’t even start a blockchain or Web3 one, although I see a lot of companies pursuing that way.
Instead, I would look for more narrow specialization within a blockchain that is small today, but has a great potential to grow into a big field in a few years.
It could be a building on a promising blockchain, like Mina or Aleo, but not pretty crowded yet, like Ethereum, Solana or Avalanche.
👮Security and cryptography
Another interesting web3 sub industry is a security related specialization — companies like smart contract security auditing studio. However that one is already crowded and rather than create another no-name auditing company — I would opt to innovate and create a formal verification studio (that works with Certora technology) — something that nobody does yet and seems to be a promising, unfolding security paradigm.
… or under-the-hood technological specialization like Zero-Knowledge cryptography or Layer2 related specialization (e.g. StarkWare’s Cairo).
Many things to choose from.
If any of the above (or below) resonates with you and sounds like something you would like to go for — let me know — I’ll be happy to chat and perhaps be an angel investor in your dev shop.
👾 AI and deep learning
Two other technologies early in their cycle presenting many opportunities include AI and VR/AR.
With great advancements in image processing and NLP, they open a lot of possibilities for small dev houses that can start by building MVPs based on public APIs like GPT3 and with time build their own models based on whitepapers and open-sourced technologies. You can think about AI as a new paradigm for programming, with a wide range of technological stacks and specializations to shuffle through.
🥽 Metaverse, Rust and Others
Metaverse, still hard to define what it is, yet we can see clear progress in VR and AR related technologies, and we could see Facebook’s acquisition spree 2–3 years ago. Hence it would be another great direction, especially for someone with a background in gaming.
Still, niches early in their cycle include DevOps and new languages like Rust. They present themselves with many different technology stacks and frameworks to pick from.
Now I don’t know what stack to pick for VR/AR or what framework for Rust, as I am not aware how the landscape looks for those, so I would start by investing a little to go to a conference to figure out what the specific specialization would be — look for those promising technologies early in the cycle.
Upselling
Narrow specialization shouldn’t prevent you from upselling services other than the core one. It constitutes the core of your activity, it’s what you hire, train, market, sell and plan for, but it’s not a constraint on what you sell to your customers in general. It isn’t supposed to exclude all other projects, which you may treat as supplementary. At Ethworks, we focus on blockchain applications based on smart contracts and later, we even boiled it down to solely Ethereum-related projects, but we build web and mobile interfaces for them too.
In fact, one of our most popular libraries, useDApp, was set up in response to the frustrations we experienced building interfaces for Ethereum applications. So grow your specialization and choose projects wisely. If something doesn’t align with your specialization, but it seems worth the candle, go for it. Don’t promote it too much, and if you want to — make it sound and well accounted for. Don’t dilute the brand, you are too small. If a project happens to outgrow itself and turn into something which you didn’t expect, and you don’t see it as part of your brand image, you can sell it to another dev shop which it more aligns with.
Over low-hanging fruits, try to attract and pick projects that will grow your brand.
Pivoting
What does it take to find a good specialization? Good research? Guts? Luck? A bit of each. You may land the perfect one at your first attempt, as we did at Ethworks. And much as I would like to attribute it to research and guts, I cannot dismiss the luck factor. Sometimes, you may find yourself in a situation where you need to change your specialization while already running a company. In the startup world, we refer to such changes as pivots. And likewise it works for dev houses.
Let’s look at some basic pivot types.
😴 Unconscious pivoting
Unconscious pivoting is a common mistake in running a business. It usually happens when you slowly start to change your specialization, without any consistent and coherent strategic effort to leverage it. You may end up pivoting to a worse specialization or lose predictability in the business.
🔎 Zoom out
Zoom out — is one of the most abused moves. A zoom out happens when you pivot your business into too broad a specialization. You do too many things and spread yourself thin, losing your unique competitive advantage. You risk turning highly skilled employees into mediocre jacks of all trades, and giving up narrowly-targeted marketing for the sake of a diluted and overly generic brand image.
A proper zoom out should be done consciously and gradually, with a proper strategic backing from the whole company.
🔍 Zoom in
On the other hand, there’s the zoom in, which narrows the specialization. I consider it to be the most underappreciated pivot.
A great illustration here could be the year 2019 for Ethworks. We built our initial brand and we were in a great position — talking to founders of many fast-growing blockchains with huge budgets to spend, marketing power to share, and grants waiting to onboard us into their technology. We were so thrilled! But we were soon to learn it was fool’s gold.
As we start applying for grants and invest heavily in learning new technologies, we soon discover we are not treated as well as by our usual customers. We would get lower hourly rates (while still spending unbilled time on training) and there was a lot of doubt from customers if we can actually deliver. It was humiliating for me and for my team. But then, these big budgets, attractive projects, top brands, the marketing power — I couldn’t help thinking about them.
And then it dawned on me that what we were doing was wrong. We were unconsciously zooming out, which was a bad practice for a company that small (20+ people at the time). So we decided to do a zoom in instead and we explicitly and exclusively focused only on Ethereum projects. In our communication, we replaced the general “blockchain” with the more specific “Ethereum”. We would reject leads looking to work in other blockchain technologies, before even getting into a deeper conversation. This move made our marketing efforts more powerful and reduced sales efforts. Soon, there were more high-quality Ethereum deals than we could handle, even though we were going through a fast-growth phase.
👩💼 Change customer segment
As they grow, many dev shops want to switch the customer segment. Companies that bootstrapped themselves working with SMBs and startups look to engage with bigger, longer, high-budget projects for customers like scale-ups and enterprises.
These are some of the most difficult pivots. Marketing efforts are different, sales cycles are much longer, as well as different customer expectations and challenges around maintaining relationships with clients. All this maps into a very different company structure.
With startups, the decision making is fast and the founders are keen to work directly with internal and external teams. Scale-ups and enterprises call for different approaches to customer service — with dedicated project managers, or accounts that ensure smooth communication with multiple stakeholders.
One can build a great business leveraging both these markets, but will likely struggle to make a sustainable business serving both. Pick one and make a long-term effort to build quality service for your target.
🎬 Starting from scratch
Many dev houses with generic services doing general web and mobile services might find it hard to zoom in or out into a new fresh space. For them, one interesting option remains — create a new company within a company. To do so, restructure the existing company to work without a few key people, while allowing this group of people to bootstrap a new brand with new specialization from scratch.
🥷 Killer mistake
A killer mistake I have seen dev houses make over and over again is mixing the two pivots unconsciously — and the most commonplace one is to do the following mix:
🔎 zoom out and the
👩💼 customer segment change,
🏜 while crossing the valley of death at the same time.
What is the valley of death? I’ll discuss it in part 5 of this blog posts series.
It is an easy way to get a prosperous company into trouble.
This happens especially a lot when a successful, around-40-people-company founders feel like they want to grow and attack a symbolic 100-people size company. Driven by ego and lack of understanding of strategic level management, or the true source of their success so far, they lead a company into a trap.
Instead, a company could split this transformation into 3 separated moves made consecutively. Soon, they might discover that 3 different transformations are not necessary, but just one or two done well can provide better results.
🤔 How to pivot
Pivot is a strategic move that will alter marketing, sales and day-to-day operations. Therefore pivot should be led, decided upon and executed by founders, who should make informed decisions based on market observation. A pivot is a hypothesis to be verified — which might take a few months to execute, before signals from the market become clear — whether new specialization is the one that market desires. The best signals would include new converted projects, but for a small company with few leads and new projects coming only every few months — proxy signals might be a good reason to keep moving forward. That would include increased interest from customers, successful social media posts or an accepted talk at a conference.
You’ll find more on how to provide strategic support in part 4 of this blog post series, about the customer pipeline.
What’s next?
Even the best specialization, marketing, and sales, won’t work miracles without access to great engineers and specialists who will work on customers’ projects.
Not only do we live in a world with a deficit of great engineers, the supply and demand is even tricker in a new fast growing specialization, like the one you would like to pick. Moreover, you want to pay reasonable salaries if you want to make a profit on top of them. How to solve this conundrum?
Read the next post in the series, where we will dive deeper into hiring and retaining great talent.Thank you for reading!
If you like this post — hit 👏 button below.
To get notification about the next parts click follow below.
If you want to continue the conversation and see more stuff like this, you can find me on Twitter: @ethmarek.
