In The Lean Start-up, Eric Ries defines a start-up as ‘a human institution designed to create new products and services under conditions of extreme uncertainty’.
COVID-19 is resulting in a dramatic restructuring of our economic and social order. Nobody can really tell what this will mean for business and consumer behaviours. So virtually any business developing digital products – especially those developing or evolving products in response to the crisis – is functioning under extreme uncertainty.
It’s a simple fact of life that start-ups accumulate technical debt at a tremendous pace. The ‘build-measure-learn’ approach, described in The Lean Start-up and embraced by most successful tech start-ups today, pays no heed its accumulation. The emphasis is totally centred on ‘building the right thing,’ and rapid experimentation to work out what that is. ‘Building the thing right’ comes later, once you’re sure someone actually wants to buy it.
At the moment, taking additional time to perfect a new digital product or feature is almost certainly the wrong approach. Chances are you no longer have a perfect understanding of what your customers want. And by the time you get something to market, things will probably have shifted again.
For the short to medium term at least, the emphasis needs to be on speed, experimentation and learning, rather than building things perfectly. Create and test hypotheses rapidly to learn, then iterate them based on that learning. When you hit gold with a product or feature that makes a meaningful difference, technical debt won’t matter at first. If it becomes problematic later, you can deal with it then. And of course, for all the good ideas that don’t work out, you’ll never have to repay the debt.
If you’re still not convinced, here are three reasons why embracing technical debt can be a good thing:
1. Respond to customer feedback fast
When you have a hypothesis for solving a customer problem, the quicker you can test it in the real-world the better. If the hypothesis is fundamentally flawed, you can find out before you waste too much time on it. If it just needs modification, again speed is of the essence.
Embracing technical debt, and building ‘quick and dirty’, lets you get that feedback with the least possible investment. Getting insight as early as possible will lead you to the right finished product more quickly, even if it does take time to factor out the debt.
2. Experiment with more radical ideas
When you’re working with Minimum Viable Products (MVPs), you can have more ideas on the boil at the same time. This reduces risk and creates space to explore radical concepts alongside more conventional features.
Outlandish ideas might fail. It doesn’t matter. If you’re working fast and not spending hours perfecting things to avoid technical debt, any wastage will be minimal.
Conversely, experimentation could reveal a breakthrough feature that gives customers exactly what they need and earns significant competitive differentiation. Even with technical debt, it might be good enough for its intended purpose, especially if it’s only relevant for a limited time while the COVID-19 situation is ongoing.
3. Cut out the cost of delay
In the current market, uncertainty is high. Nobody can say for sure when the kids will go back to school, social distancing will end or non-essential businesses reopen. We don’t know how much the current measures will remain part of the new normal when that does happen either. If you’re developing an idea that’s related to or impacted by these factors, time spent refining it ahead of launch will be wasted if (or when) the goal posts move.
You can’t predict the future, but you can ensure you’re able to adapt to it quickly. So release something – anything (almost) – as early as possible; make sure you can measure customer response; and harness that real customer feedback to improve it.
Things are changing rapidly at the moment and the potential cost of delay is high. Waiting for feature completeness or adherence to engineering principles could make a great idea a total flop if it’s launched too late.
Perfection can come later
In a changeable world, ongoing experimentation and learning is vital to ensure digital products have enduring relevance and value. The COVID-19 situation takes the everyday volatility of the digital economy to a new level. To respond effectively, product teams will need to relax their engineering principles and step back to their start-up roots. When it’s harnessed deliberately and prudently, technical debt is not a big deal. And it can always be designed out later.
Eric Ries has written extensively about the need to embrace technical debt. In his words:
“It’s more important to make continual progress than to build the ultimate design”.
I couldn’t agree more.
In our work with scale-ups we help contain and manage the inevitable chaos and challenges of rapid growth to maximise future performance. Our dedicated scale-up squads have the energy, enthusiasm and expertise to improve underlying infrastructures so our customers can grow with confidence.