Using the UCD to plan your project

Several times over the past month, I have been asked to jump onto a project and give some usability, human factors or just general insight. The common complaint I have coming into these sessions is that the project is not prepared with some of the core answers they'll need to test the hypothesis (the web page, experience, etc).

The first thing you should do is ask yourself, "What problem am I solving for?". A web page is a solution to a question a user has. They have gone through many avenues trying to solve it (search, word of mouth) and ultimately are presented your page as a potential solution. We, as a content provider, must consider that the page we deliver is our hypothesis on what the answer might be. In understanding what the user's problem is, what their question is, we can best attempt to answer it.

The second thing you should do is ask yourself, "How do I know if a user has had their question answered?". This is done through metrics, through user interviews, through a set of measures in which you speculate that a question has been answered. Ultimately, there are limited ways to define satisfaction... though some sites may use something like, "Was this helpful?" "Did this answer your question?", and so on. In the big picture, I have often felt that commentary by end users as well as repeat visits often frame if your solution is in fact meeting the goals.

Finally, the last question you should ask is, "Is there an faster/easier way to answer the question the for the user?". This can be done through usability and heuristic evaluation post launch. You examine the site and look at how the solution is presented and see if there are places were things aren't relevant (contextual unaware advertisement).

This is just a quick post, but I wanted to highlight that if these questions are set at the beginning of the project your chances of success increase. I truly believe in the scientific method when it comes to questions and hypothetical solutions. A user is a question, they represent an impulse or planned need for information. Your web site's job is to address that problem in a way that not only answers the need but does it in the easiest way possible.

Your idea is great, let's make sure your users like your information layout

One of the hardest things any organization can do is come up with the 'next big thing'. The ability to foresee events is something that no one has really been able to do consistently without a pitfall or 8. However, when a product manager does in fact generate an idea, the first thing that should be done is do our due diligence to ensure the idea is in fact plausible.

That's where a user researcher or usability specialist can come into play. It is not their job to dissuade the idea, but to discover the user base that would like to use the idea. For example, you would not market an underwater basket weaving kit to a user who does not or cannot swim. Often, the case with product development is that without research, you are throwing things out there and hoping they stick. However, when leveraging user research you can still throw things out there, but you are throwing them to a group of people who you've discovered might be interested in that kit. The cost savings in finding a general target market as well as using this target market to focus your design approach, is measurable through feedback, surveys and follow ups.

Finally, user research helps your business understand that your kit does not have to work for everyone. It is okay to focus your UI and your approach on a specific market. Remember, the user researcher is not there to find the target market or the valuation, but to ensure that the approach does fall in line with that target. So for all you underwater basket weaving aficionados out there, take it to heart that there is a user researcher ensuring that the UI and the experience fall in line what your needs specifically....

Consistent matters, sort of

Having every page on your site be the exact same can strengthen your overall brand. However, it can diminish the value of the content, from a user's perspective, as they traverse your site. In this post, I'll comment briefly on the value of brand allusion.

Even in a niche market, you can have a multitude of like-branded buckets that allude to your overall brand while not creating an environment of ignorance. What do I mean by this? If you look at 10 items that have the same label, you will eventually start ignoring the label (usually by the 8th page...similar to banner blindness). However, if you allude to your brand without being redundant then you maintain the trust that your user has in your product as well as demonstrating the depth and breadth of your offering.

An example is how ebay and ebay motors are interconnected in a branding way but not in an obvious linking method. Visiting the ebay website, you see their branded logo and their standard information architecture. However, visiting their motors site, you are aware that it is ebay but the linking does not connect the parent (the parent link is offset intentionally to ensure you do not fail out of the site). This demonstrates how two sites are linked through brand allusion and not through regurgitating the same image exactly. It enables the user to clearly recognize they are not within an ebay site, but that they are on an ebay property.

Fighting the good fight (aka hiding behind the user)

In a recent conference call, I was accused of 'hiding' behind the user. During the call, it was demanded that we go ahead and make assumptions about the user to ensure the project time line was met. My response was that 'this is dangerous at best' because as a user expert, I was trained that I am not my user and any assumption may inject bias.

This is symptomatic of the gap between the status quo of the software development life cycle and the iterative approach of the user centered design process. Where the SDLC fails, is that it doesn't account for iterations... in other words, you repeat the SDLC in versions. UCD assumes that you will pass through it several times and is designed that way. So, the challenge lies in assimilating an iterative and qualitative approach within a quantitative serial cycle.

Some solutions are to integrate a system of user based checks and balances within the current SDLC. This maintains the process that the development organization requires to function while not being solely dependent on UAT findings to fill the need.

Focus on what you are solving for

Competitive analysis can be a critical tool in aligning the goals you are trying to achieve with the opportunities that exist in the market you are looking towards. Often this simple sheet can easily describe your roadmap to success through highlighting the gaps in where your competition and even yourself exist.

I often recommend partnering this information with a 'skunk works' best case approach. In some cases, you can actually roadmap to your best case using the offsets between what you are developing and what you would like to see as optimal.

On Usability

It has been quite some time since I last posted. So, please forgive me.

Working recently at a major media site, I still see the maturity of the customer voice having gaps. For me, the concept of user feedback avoidance, deprecating or eliminating task analysis, can be fatal to any project.

For a task analysis, you are looking at why the user wants to do something and the steps they may encounter when trying to complete that 'something'. Without this understanding, application solutions can become 'Swiss army knives'. This means they won't necessarily solve the problem at hand, but hope to be able to solve whatever problem may arise. For this particular knife, that makes sense. In the eCommerce world however, this can lead to diminished value of the product. In other words, you don't do one thing well, you do a bunch of things adequately. In the web world, adequate does not translate well to cash flow.

User input helps provide the evidence that the approach you are taking is the right one and will be used correctly. As in the 'Swiss Army Knife' example, if a user realizes that during use they really needed an all-in-one, then this reinforces the approach. However, if the user only needed to open a wine bottle, there are better cork screws than the one that comes with a 'stock' army knife. The point here is that placing the knife in the user's hand and allowing them to give you feedback and how they think it will be used gives you clarity on the solution. Without that clarity, you are not affirming your direction and could be traveling the wrong way.

Using these simple ingredients, you can increase productivity by understanding the task/tasks that you are solving (or making easier). Moreover, you can then reaffirm the approach you are taking by following up with the end-user.

Unfortunately, in a large media organization there is still a focus on what is visually stunning and eye catching. This dependence on emotional design can create some stunning solutions. However, without the task analysis and user feedback the solution is created to meet the needs of the designer and their leadership. As Norman said, "An object that is beautiful to the core is no better than one that is only pretty if they both lack usability."[1]

1. http://www.jnd.org/dn.mss/emotion_design.html