Paradoxically, the more successful you are at evangelizing usability in your organization, the higher the likelihood that you'll have to change your strategy. The approach that takes your company from miserable usability to decent design is not the one you'll need to get from good to great.

A company progresses through a series of UX maturity levels as usability becomes more widely accepted in the organization and more tightly integrated with the development process. If you are the company's leading user advocate or usability manager, one of your main jobs is to prod the company to the next level.

Early Evangelism: From User Advocate to Usability Group

Starting point: One or two people in the company care about usability, but working on usability activities is rarely their main job. As a result, they start small, typically by doing a little user testing on the side.

Desired goal: Establish an official usability group with a manager, a charter, and a budget to perform usability activities.

In the early maturity stage, few resources are available and the company is not truly committed to usability. About halfway through this growth process, the company typically has a few full-time usability specialists, but they won't officially "own" usability because they don't have a recognized spot on the orgchart.

Under these circumstances, it's impossible to support the full user-centered-design (UCD) life cycle, and it would be futile for the few lonely usability specialists to try to do so. A company can't be forced to jump through multiple maturity levels in one push.

At this point, the strategy should be as follows:

  • Usability methods. Do small, qualitative user tests. Don't bother with advanced methods, such as field studies and benchmark studies, because you won't have the time or money (and might lack the expertise).
  • Lifecycle stage. You perform usability at whatever point the project manager calls you in. This is typically late in the project, when the manager realizes that the user interface is in trouble. We all know that it's better to start usability activities early — before design begins — so that user data can drive the design's direction. Unfortunately, this only happens if the project manager is a fervent believer in usability, and that won't be the case when the organization is at a low maturity level.
  • Choice of project. Work with projects that want to work with you. When you don't have the organizational mandate to own usability, you can't impose yourself on projects that might need you, but don't realize it. You're much more likely to get your recommendations implemented when you deliver usability findings to managers who want them.
  • Methodology growth. Set up a usability lab, establish procedures for recruiting test participants, and define standardized test tasks for your domain. Building this infrastructure will let you roll new tests with little preparation, and thus do more of them.

The key word for early evangelism is to be opportunistic in allocating your scarce resources. You can't follow the recommended usability process in all its glory because your organization lacks the commitment required. Instead of fighting windmills, go for the easy wins.

Luckily, a company without a systematic usability history will have much low-hanging fruit for you to pick. Even quick-and-dirty usability activities performed at the "wrong" project stage will often improve a website's final user interface by 100% or more (slightly less for other types of interfaces).

Once these easy usability wins generate substantial business wins for the company, management is usually interested in funding an official usability group.

(Much more on early-stage evangelizing techniques in the full-day course The One-Person UX Team Tool Box.)

Late Evangelism: From Usability Group to UX Culture

Starting point: There's an official usability group with a manager, a charter, and a budget to perform usability activities.

Desired goals: Establish an entire user experience department, with several specialized groups for different user-centered activities. Generate total company commitment to a formal UCD process — owned by the user experience department — for all development projects.

Typically, usability becomes "established" in a company without giving the usability group the power to fully own the total user experience. The usability group is often viewed as a service organization that supplies usability expertise to project teams at their managers' request. At this point, usability groups rarely have enough resources to supply all projects with the services they need to meet the recommended UCD process.

As the company matures, more resources become available for usability, but some prioritization is still needed. At the previous stage, priorities were opportunistic; they must now be more selective. Rather than chase easy wins, you must build spectacular wins for usability to convince executives to move the organization to the desired goal state.

At this point, the strategy should be as follows:

  • Usability methods. Continue to perform simple user tests to clean up bad design. You'll likely do this forever. Now, however, you should spend more resources on deeper studies that generate new insights and thus inspire new features or paradigm shifts in the design. Conduct early studies such as field studies and competitive studies that set the design's direction. Collect usability metrics from benchmark studies. Track this data over time to quantify usability's return on investment.
  • Lifecycle stage. Press to start usability activities earlier for each new project. For key projects, ensure sufficient resources to conduct both early, pre-design studies (for big-win insights to shape the project) and multiple rounds of iterative testing (to polish the design beyond simply fixing its worst flaws).
  • Choice of project. Focus on high-impact projects where the benefits of substantial design improvements will have enormous monetary value and high visibility for senior executives. Better to support the full life cycle of a few important projects than to do lower-impact, last-minute testing for all projects. Adopt a killer intranet app like the staff directory as your baby to create an example of breathtaking user experience that everyone in the organization can experience everyday.
  • Methodology growth. Once you've performed enough comparative, field, and iterative studies, generalize the lessons and write usability guidelines for your organization's particular type of user interface designs. Develop and maintain official UI design standards. Guidelines and standards leverage your usability findings beyond the individual studies undertaken by your group.

The goal of late-stage evangelism is to fully integrate usability with development so that it becomes second nature to start projects with usability activities, before design begins. The organization needs a usability culture. All managers should understand the basic steps in the UCD life cycle — if nothing else, because that life cycle has been mandated as the way projects get done. For this to happen, executives must have seen several examples of the added value created by full-fledged usability, as opposed to last-minute user testing.

Because user testing is so cheap and so profitable, it's easy to get caught at a mid-level of organizational maturity, where user testing is common but deeper research is avoided. To get past this, you must show that 100% improvements are nothing compared to what's possible. Achieving a 1,000% improvement in project outcomes requires more work than the 100% improvements that come cheap, but that's what you have to do. The way forward is to focus on a few high-impact projects, and make them smash hits.