Too Much, Too Early

prototypeLast week, we were in a discussion about a new mobile web interface for a client. We were looking for feedback on the design. The initial design was presented and we expected to get feedback about the usage of screen space and the navigation model or flow if you like. Instead we received a lot of feedback about the lack of brand identity and proper use of colours. Looking back at the situation the design was already too much like a realized interface, which hinted towards the wrong discussion. We should have had a much rougher prototype of the interface where the design was sketchier to get the focus on the right (lower) level of the design, not on finalizing it.

A two-step rapid prototype design would have fitted us much better so it seems. We needed a low-fidelity prototype for this kind of initial flow discussion and brought in too much detail too early. A sketchy look and feel would have given the participants the feeling that anything could still be changed, that every part of the design was open to discussion. Nothing was fixed yet. They might all be changing it all.

The design at hand was clearly more towards a high fidelity prototype where the feedback is expected to be around actual colour usage, branding and pixel perfectness. Such a two-step, low- followed by high fidelity approach, would have given much more feedback on initially the layout and navigation and would be perfectly followed by the second step of actual brand guidelines and perfect fit on multiple screens.

It’s clear that Too Much, Too Early is just as bad as Too Little, Too Late…

Here proposed is a two-step approach on the design, but I would really love to hear how many levels, or steps, you usually take (or expect) on such a design and room for feedback. How many iterations do you share (or want) for gaining feedback and how many do you use internally without the external feedback loop?

Arnd Brugman

About

Creative business technology management consultant with a major drive and passion for people value. Innovator with broad business and technology knowledge based on twenty years of experience in IT and fourteen years of management.

More on Arnd Brugman.

Related Posts

Your email address will not be published. Required fields are marked *

1 + 5 =


  1. Etienne September 18, 2014 Reply

    I can absolutely back this. Using the right prototyping approach for getting the right questions answered at the right time is crucial, whether you evaluate with experts, users or even clients. The first general questions you need answered are those that are most expensive to change when implementation has started. From experience these are generally the dimensions of Concept, Functionality and Navigational Flow. If this is what you need reviewed, don’t build the rest.. It can wait until prototype fidelity increases.

  2. Carl.henrikson September 18, 2014 Reply

    Earlier I had exactly the same experience as you describe in the article, with clients focusing on the wrong things when presenting a mockup that is too detailed. Later, I decided to always make prototyping a two-step process with lofi first. usually I find that two lofi iterations are enough before going to hifi. After that several smaller iterations are made with the actual app on actual devices to improve usability.