Process Transformation

A little over a year ago, THINK set out in earnest to build the agency of the future. This isn’t the sort of thing that happens overnight. Nor does it happen without a great deal of discussion, contemplation and consideration.  In order to grow the agency to meet the emerging and future demands of our clients — the requirement to scale digital into meaningful parts of the business, the desire to innovate and to deliver something imminently useful to customers — we have had to rethink our process. The implications on how we work, our offerings, who we hire, the clients we choose to work with, and how we are organized internally are all affected. A year into this process, we have learned a number of lessons.

Question Everything

One of our first tasks was to do a deep analysis of our processes for research and discovery, experience design, creative design and development. We looked at how we worked, where knowledge transfer happened, the types of artifacts we produced, as well as the inflection and pivot points where projects take shape and ‘the magic happens.’ Our process has always been collaborative, but we saw opportunities to more deeply embed shared representations and the decisions they elucidate into our work.

We also began to really consider the ways in which the digital and physical worlds have been changing.  And blending together. As I mentioned recently in my “THINK. Make. Learn. Disrupt.” blog post, the late 90’s through the mid-00s saw much change, but there were also many constants dictated by technology. Monitors had a 4:3 aspect ratio and there were really only three screen resolutions to worry about. People accessed the Internet sitting in front of a monitor and a keyboard using some type of pointing device like a mouse or track pad. Interactive modalities were fairly standardized and limited as well. People understood how to click and scroll. Most companies creating interactive digital experiences followed a fairly standard waterfall approach to problem definition, discovery, design and development.  This is the era we grew up in as an agency. More than that, it’s an era we helped define and it has served us well.

The essence of the “waterfall” approach is getting one stage “right” before moving on to the next.

But as we move beyond the monitor and the mouse into a world of variable viewports, changing physical contexts, an influx of connected devices and a rapidly expanding lexicon of interactive modalities, this way of working seems dated. The artifacts also seem dated, and perhaps even irrelevant.  Why would we use the same processes and artifacts in 2013 that we used in 2003? We are now designing highly experiential and rich interactions that span a myriad of devices and contexts. Our content must be as adaptive as our experiences.  Our approach must be as responsive to this new landscape as the digital experience we are creating.

To quote Hugh Dubberly, “Our processes determine the quality of our products. If we wish to improve our products, we must improve our processes; we must continually redesign not just our products but also the way we design.” This has never been truer than it is today.

Sketching, Prototyping and Co-Creation

One of the first changes we made was a move to incorporate more prototyping into our workflow with a goal of creating interactive experiences sooner and iterating more frequently.  The results have been incredible.  Working this way has changed the dynamic of our team, removed disciplinary silos and barriers and created a culture of collaborative co-creation. Our projects are more satisfying to work on, our teams are happier and our outcomes are better. We’ve also become far more engaged with our clients.  We work in real-time with them, showing them functional options and interactive elements across screens very early in our projects. This allows them to make better choices sooner. It also means that they are more vested and excited by the work. It seems a bit unexpected, but prototyping and frequent co-creation of ‘the thing’ — be it a website, app, or device — have improved our account management and project management processes as they have changed our internal work processes.

Interestingly, one of the outcomes of this shift seems to be a resurgence of paper, pen and whiteboard drawing. Working in real time, in a highly collaborative team environment means more sketching and dialogue around those sketches. At times, it means moving from a hand sketch directly into code that can be assessed and modified as a proof of concept. It means designing interactions by building them and testing them. It means creating a shared understanding of the desired experience for all stakeholders much, much earlier in the project. For us, it has also meant that our developers may be sketching alongside an experience designer or a creative designer. It’s a beautiful thing to behold.

The wireframe has not disappeared for us, but it is no longer the default first page-level artifact for us either. In fact, I’ve recently completed a 12-week project without ever opening Omnigraffle or Visio. All of the experience moved from hand-sketches directly into interactive prototypes. The credenza by my desk has a stack of about 150 – 200 hand sketches, often quite detailed, that document the experience and the iterative nature of the work.  We are in flight on another large project that is using an agile development approach. There were very few wireframes generated for Release 1. Everything was sketched and prototyped. This project now has a very solid foundation with defined interactions. As we’ve moved into the Release 2 sprints, the need to sketch and protoype has diminished considerably and wireframes have become a much more effective and appropriate artifact.

In both cases above, the impact on requirements cannot be overstated. Agencies have historically seen requirements as precursors to the design effort, as well as fixed and eternal for the life of the project. Requirements after all are either met or not met during QA, acceptance testing, and client approval. But in a sketch- and prototype-based process, requirements are allowed to shift, morph, and even be eliminated (or pushed to later releases) as the requirements definition is updated alongside the project artifacts. Whether from the THINK internal design team or from client interaction (or even from vendor, API, and platform choices), the ongoing nature of design allows requirements to be co-created. Defining requirements while the project is in motion does not have to add time to delivery — instead of a time intensive highly speculative and disjointed ‘requirements gathering’ period at the outset, we distribute this effort (and the contributors) throughout the design sprints.


Another exciting outcome from the last year is a new understanding that process is not a prescriptive, one size fits all affair. Rather, it can and should be viewed as a flexible tool palette that should be adapted to the needs and constraints of each project. Our Director of Strategy, Zach Pousman, recently came across a tremendous collection of process frameworks (PDF) compiled by Hugh Dubberly. For us, it’s a bit of process porn that’s tantalizing to peruse (Zach and I are both a bit geeky that way).

PDCA stands for plan-do-check-act cycle of continuous improvement, a standard principle of quality assurance and management.

The earliest process framework Dubberly documents is from 1939. It’s a plan-do-check-act (PDCA) cycle for continuous improvement that is not altogether different from our own “THINK. Make. Learn.” approach. Simply scanning this document is a rich visual exercise that speaks to the flexibility and adaptability of process.  While I am reluctant to pick a model from this collection and call it “right” or “preferred”, I will say that the Nigel Cross illustration on page 25 conforms to how I have begun to visualize our new processes at THINK. I believe our outcomes are more focused, but we also have more frequent and active cycles of divergence and convergence with more spirited discussions and debates.

The overall process is convergent, but it will contain periods of deliberate divergence

One outcome of our careful thinking about design process has led us to a conclusion: that there isn’t one process that will always be right. Depending on the nature of the assignment, we take some time to be flexible and select the right design process to fit the nature of the problem to solve.

Continual Iteration

I think we are well on our way to becoming the agency of the future. We are embracing new technologies, methods and processes in unprecedented ways and it’s exciting to be able to participate and influence these changes at THINK. We’ll continue to document and discuss our evolution and we’d love to hear how you and your company are adapting to the sea of change in both the digital and physical world. It’s a great time to be on a journey like this.