It is no secret that I have a little bit of a crush on Edenspiekermann. I have been quite inspired by their writings on agile and I am a very big fan or the ‘rules’ they outline in their manifesto. I was therefore very excited when Rober Stulle and Michael Börner of the Berlin office agreed to meet me for a coffee, when I was recently in Berlin.
As I have written in an earlier blog post we at Hello Group are currently undertaking a change in the way we work. We are working on shifting from a more traditional design agency setup (which usually subscribe to a waterfall philosophy) to a structure that resembles those found in design studios and start-ups (Which usually have a more agile and lean approach).
I therefore set up the meeting with the two knowledgeable Germans as a way getting a little insight into how they managed to make this transformation. What I learned was that they have an inspirational agile approach that merges the design and development tracks of product design, but also that even they haven’t managed to completely solve the puzzle of incorporating the lean approach into their projects. Let me explain:
If you look at how most agencies talk about their process, they talk about 3 to 4 sequential phases. There is usually a research phase, a concept phase, a design phase and a development phase, which ends off with an arrow pointing back to the research phase (which usually is mostly for show). This is what Espi has gotten away from. The digital landscape is simply to complex for this, and you need a process that allows you to constantly recalibrate. This recalibrating is constantly happening at Espi when their designers and developers work collaboratively in teams to create the complex digital products that inhabit our lives today.
Working truly agile though, means that you don’t know the exact outcomes of a project when you start. This ‘black-box’ is difficult to sell as agency, since most clients just want to buy a website or an app. They don’t want to buy something as intangible as a process, a team or a vision. What Espi has done to accommodate this is to break the project up into separate phases – and this is where the sequential waterfall mentality appears again a little.
Their first phase, which they call the ‘Creative opening’, consists of the research and high-level concept phases. This package can be purchased at a fixed price and it is from this foundation that the development phase is structured. The initial backlog is created from the business intelligence and the user research, and a high level roadmap for development is then created. This allows the client to get a better idea of what should be designed to solve their actual challenges and what this will approximately cost.
Working with agile in this way, where the development and design process is merged, allows Espi to deliver a lot of value. Their developers and designers work together with the client to ensure that what is designed can be developed and what is developed solves actual business problems.
Getting lean and mean.
Working with design and (full stack) development in an agile process is in my opinion the foundation for creating great products. The problem with this approach though is, that it doesn’t really ensure that the product you build is going to solve the problems you set out to solve. It is impossible to predict how the product is actually going to perform when real live users start fiddling around with it. What you thought would make users jump for joy can leave them utterly unimpressed. The agile process doesn’t solve this problem.
To reduce the risk of deploying a useless product, I believe you need to do user validation continually throughout the development process. In our meeting Robert, Michael and I talked about this and how difficult it is to convince a client to pay for continual user validation throughout the process. It was not from a lack of wish that Espi didn’t have this as a natural step in their Scrum sprits yet. But it was almost nice to know that even some of the forerunners hadn’t figured everything out yet.
We also talked about how difficult it is as an agency to keep your hands on a product when it launches. All wisdom and knowledge points to the fact that you need to keep improving and iterating on a product once it has been deployed in a live environment. But for some reason it is hard for clients to find budgets for this. Unless something is actually wrong with the solution most people regard it as ‘done’ when it is launched. If you are lucky you get to do a phase two, but that usually doesn’t start up before months later.
Robert and Michael have been working on trying to build this loop-back into their process and what they are currently doing is to set up a 100-day meeting when a product is launched. As the name suggests, they set a meeting 100 days after launch to evaluate. They do this with all the key stakeholders and look at how the design is actually solving the problems they set out to solve.
The real secret sauce
Flexibility and agility is vital in a design process – but unfortunately no process framework will ensure this. What I believe to be Espi’s real secret sauce is their focus on client relationship and trust. (just check out this video). Without trust you will not be able to do agile, you will not be able to do lean and you probably won’t be able to create much of real value.
To do great product design in an agency setting you will need a partner relationship with the client and not a vendor relationship – and this is what I see Espi talking about continually.