Stop the assembly lines

and let’s try something different.

I’ve had the fortune of making some great websites over the years. Looking back it’s interesting to see what things worked and what things didn’t. It provides a chance to get some insight into how you might improve the process of making things. Almost universally the problems that arose lead back to one thing: a lack of flexibility. There was really only one answer as to why our process lacked flexibility and that was because it was engineered not to have it. Our process for making digital products was not that different from every other design company out there. It followed the sequential phases of discovery, design, development and finally deployment. Every step building on the last leading to one glorious outcome. It’s what everyone did and what our clients demanded, so we got on board. We eventually learned that we needed a process that didn’t punish us when requirements changed. It was inevitable when building a website, so our company sought for a new, better way.

How we got here

The idea of a sequential creation process got its start from manufacturing and construction. It was important to have a very rigid, functional and safe plan before building any type of structure or hardware. It also gave the workers a clear plan to follow without having to think about the big picture. The process could easily be turned into an assembly line and was straight forward to manage. Changes to the design far down the production line were, in most cases, impossible to do. The sequential nature of the process lent itself to being easy to predict and therefor easy to estimate time and cost. It was eventually coined the waterfall method because the analogy best described the act of going from top to bottom while at the same time illustrating the challenges of trying to go back once you’ve gone to far.

The waterfall method made its way into the digital space back when most software had to be perfect because of its close links to the hardware. The computer engineers required a similar precision that construction teams did and the waterfall method provided that. Eventually, software moved further and further away from the hardware yet the engineers were still using the same process. Some of the things that made the waterfall method so great started to give resistance against the needs of the customers who bought the software. The demand for flexibility when making things became apparent and this waterfall method started to show its weaknesses.

While decades have passed, the waterfall method has crept its way into most creative industries even though most of what’s made in these industries won’t ever take a physical form. The structure of the waterfall method allowed creative industries to price their work and scale their agencies effectively. However, the idea of adapting a process that excelled at creating efficient production lines didn’t match the needs of the customer buying the product or the teams of designers and developers who created it. Even still, managers created similar manufacturing constructs within the creative agencies with production teams and lines of artists and developers slamming away at keyboards. There was a clear mismatch in what the clients and talent wanted and what the process provided.

Where we are today

Without a doubt the waterfall method has lead us to some great digital products. Often what you don’t see however are the projects that failed because of it, and the number isn’t small. So many businesses are buying digital products that fail to meet their expectations and they are tired of the wasted resources. Part of the problem is the ubiquity of the waterfall method; it’s just what everyone is used to now. It’s so engrained in the purchasing and execution processes of digital product builds yet it fails to provide the flexibility everyone needs. Business moves quickly and by the time the project ends its requirements may not match what they were at the start. Not being able to adapt to those changes is unacceptable.

Why do we still resort to a process that doesn’t match the needs of the business or the design company? It’s time to move past the idea of digital products being made like physical products in the manufacturing or construction industries.

Let’s try something different

A business is like a living, growing entity. It’s not often fixed in the same place for long. The demands put upon it are always in motion and the customer’s needs shift on top of that. For a digital product, like a website, to be a valuable tool in helping a business thrive it needs to keep up and adapt quickly. For a design company to be successful in achieving this it needs a constant stream of data to make sound decisions from. The following is a flexible process that leads to creating digital products through data-driven goal-directed design.

Assessment Period

An effective design team needs to get a lay of the land before it can even think about creating a quality user experience. The goal of the assessment period is to understand the problem and the business ecosystem before creating a viable solution and roadmap. This assessment period replaces the time a business spends creating an extensive RFP or creative brief. Instead the business can describe their current situation when interviewing design companies. Upon selecting a design company, the team would go through a discovery period together and collect relevant information about how the digital product could best serve the business. Data would be collected about the existing products, the business vision and mechanics and finally the technical constraints the product will be under. This assessment period lives outside of the scope of the digital product and it serves to create a common ground for the business and designers to work from. With real information at hand and a clear understanding of the landscape ahead, designers can set out creating valuable ideas to help the business thrive. The result of this period will be an action plan that outlines the big picture solution as well as the time and costs associated with making it.

Start Small

The outcome of the assessment period will provide the designers with essential data on business goals and past customer behaviour. It’s the designers job to then narrow down the focus to a minimum feature set that gets a customer to their goal. Until the product can show it provides value effectively, its features must be minimized. By starting small and focusing on essential user experiences, the resources are effectively distributed to the most important aspects of the product. If there is pressure to add features, they should always be tested against the project goals. If they don’t aid in achieving the goal, they shouldn’t be part of the product. The goal of starting small is to create a product that goes to market as quickly as possible. As soon as real customers start using the product, essential behaviour patterns will emerge that will inform its future evolution.

Sprints

The waterfall method always finds itself in a head down marathon to the launch date. The result is a big curtain reveal hopefully followed by oohs and awes. The alternative to this is breaking the process down into small chunks of time called sprints. Each sprint has a clear goal associated with it and all members of the design team, including members of the business, works to reach this goal. At the end of the sprint all members would discuss the work done and test the product together. The team would then discuss changes and set goals for the next sprint. The sprints continue through the entire process all the while adapting to feedback, bugs and shifts in requirements. As the product starts coming together everyone has a chance to see and interact with the moving parts.

Measure

Once your sprints have resulted in a sound product it can be released to the wild. The product will be loaded with enough tracking tools to help test hypotheses and inform what is working and what is not working. In most cases this minimal product will be deployed much sooner than the typical waterfall method would afford. This allows the design team to collect the much coveted data they need to make new hypotheses and start formulating solutions to test them.

Evolve

The waterfall method has a very clear start, middle and end. Once the cloth is pulled off the business is handed the keys by the design company and sent on their way. This is counter to what should be happening because the goal of launching the product is actually the beginning, not the end. It’s where the most meaningful decisions will be made about the evolution of the digital product. A good process will continually learn and cycle back creating a better and better product over time.

The intentions of using the waterfall method within the creative industry are good ones. It’s about creating an efficient time box around a solution so it can easily be priced, managed and built. The unfortunate consequence is that it’s uninformed, has trouble adapting to change and as a result can lead to poor solutions. The conflict associated with the poor outcome can lead designers and business owners to point fingers and yearn for something better.

Fortunately there is something better! It’s important to remember that making digital products isn’t like putting together a Lego set. Business is like a living breathing entity that needs constant fuel to operate. The digital product made is a reflection of the business and because of this, it should be able to keep pace with the growth and direction of it.

I am the Co-Founder of GRAND, a digital design company that specializes in customer focused Websites, Apps and Social integrations. Check us out at WeAreGrand.com or connect with me on Twitter @IAmMattQ.