- December
Posted By : Adrian Chan
Lean UX and lean startups: is trial and error the best way forward?

Lean UX is meant to be the user experience design approach best suited to the lean startup. Lean startups are meant to be the best business approach suited to the startup industry. And agile development, it goes without saying, the best development model.

It’s hard to contest these on the basis of budgets, market conditions, and even to some extent the needs of social tools early in development. Developers need users, not just to test out product stability and functionality, but to provide feedback also. By starting small, startups get to test out their designs for MVP.

But I can’t help wonder whether this method is the best we can come up with. For in some respects it is less a method and more a justification of trial and error by means of factors unrelated to design. Developers, and designers, surely have expectations of what will happen when their social tool is first used. Entrepreneurs surely have expectations of the same. Perhaps developers and designers base their insights on personal or industry observations. While entrepreneurs might be more informed by the business impact or problem-solution assumptions that they have made.

While there is nothing egregiously misguided about either of these approaches. They’re eminently practical. But given that so many social startups either fail outright, or soon pivot (and sometimes more than once), you can’t help but wonder if there’s a something missing.

From personal professional experience, as well as excessive use of social tools, it strikes me that the lean startup approach lacks a certain amount of spherical mass when it comes to social design. Not that users and social are an afterthought. But that they are an assumption. It is assumed, on the basis of success elsewhere (logical, yes), that users can be expected to adopt and embrace this in the same manner they use that (fallacious).

If we have learned anything, it is that social is not just technology and technique, but also culture, adoption, beach-heading, community management, and so much more. The assumption that what works in one context will work equally well in a different one is perhaps logical. But the conclusion drawn from it — plow ahead and do more or less what they did, but marginally better — is false. Users don’t make a move for the marginally better.

The lack of courage comes into play then when it comes to thinking both differently and better. One might blame this on the industry’s disposition towards financial risk-taking; which is accompanied by an aversion to risk in the user experience department.

That entrepreneurs might be more keen to spend large sums of money on things that more or less exist elsewhere in the marketplace, but shy from the risk of new or different user experiences, makes no sense. And certainly isn’t a design methodology. It suggests to me only that the industry has more confidence in the proven use of existing social tools than it does in its understanding of users, what they want, do, and why.

Every successful new social tool took a risk on user behaviors and social outcomes. Users are entirely capable of new experiences. They can be understood, and their responses to new social products can be anticipated. So, too, can social outcomes — or what happens when user adoption scales.

Perhaps it’s just me, and my own bias. Perhaps it is that neither design nor engineering pass through much social theory, anthropology, or psychology. Perhaps it is just how the industry works — risk on financially, risk off on design. From where I stand, however, the less lean startup methodology could use some fleshing out.


  • I think you may want to re-read the Lean Startup book… Just saying.

  • Have no problems with books. This is about what companies are actually doing.

  • Interesting post Adrian, but I think you’ve got the wrong end of the (lean) stick 🙂

    I wouldn’t say Lean is only suited to startups, rather anyone looking to creative innovative products. Rather than creating safe products, it creates an environment where innovation and experimentation are encouraged. There’s less risk when you’re using lightweight, informal methods to learn about your users.

    There’s no silver bullet with any of these approaches and obviously they need to be adapted to your circumstances. We incorporate elements of lean into an agency environment which isn’t easy as there’s the client to keep happy as well as the customers. It’s all about finding a common ground and gaining trust.

    I think you’ve misunderstood what lean is about. Lean UX focuses on reducing the amount of documentation, with more focus on the experience being designed. Lean Startup is about reducing the risk out, not the creativity. No successful product can be created by just following a step-by-step process – it needs a visionary, creative inspiration, a great facilitator (and let’s not kid ourselves, lady luck) to guide the vision through from start to finish.

    I disagree with Eric Ries in that he sees design as waste in an MVP. It all depends on your definition of ‘minimum’. If you hack something together and it’s poorly designed with little thought to the UX then you won’t know whether it’s the product they don’t like or the implementation of it. In my eyes there needs to be a balance. Just enough to get the answers you need without compromising on quality.

    Also I don’t see lean as just a way to do something slightly different to what’s out there in the marketplace already. I think the more groundbreaking the product, the more important it is that you test our your ideas at an early stage, rather than investing large amounts of time and money in building a product that nobody wants.

  • I may have misconstrued my argument. I have nothing against disposing of excessive documentation, process overhead, and suchlike. If they lead to wheel spinning and waste time, they’re not serving the purpose. The point here is to design and develop at the speed with which the market and industry move. 

    However, design matters of course, and is for all intents and purposes unavoidable. Design occurs whether it’s designed or accidental, thought out or duct-taped together. In social products, design should also account for social outcomes — how users will engage with new social features, how social dynamics might change, and so on. These are design factors in social that are worth taking into consideration — before, during, and after development. 

    So what I meant to critique — the trial and error aspect of lean — was meant not as a critique of lean but as an argument for better accounting for the dynamics of social systems. You want things to work — in social, that’s somewhat complicated. It’s a good idea to think that through before rushing into development. 

  • I think we’re in agreement on that 🙂

Leave a Reply