Designing for end-users you don't have access to

9 Feb 2012 - 4:43am
4 years ago
5 replies
2104 reads
Henri Huttunen

Any practical tips on how to design in a user-centered manner for a part of your user base you just cannot get more than second-hand accounts for?

In addition to second-hand accounts we of course have access to event logs that record user behavior, but this tracking cannot be tied to a this elusive user group with any reliable accuracy (i.e. we know they're there, we just don't know which ones they are).

So this is a classic theory vs. practice problem, rooted in scarcity: in theory it should be so that everyone's interested in getting the design right and we can just walz in and do an interview, but in business practice it's not realistic to expect access to users with a lot of distance to the software provider (we're talking B2B software).

So, what are your experiences in this scenario?


9 Feb 2012 - 9:35am
Brian Mila

Yeah, that sucks.  If you have access to some SMEs they would be the best proxy.   Or possibly if you have a support group that handles customer complaints or something.   Try to find the interaction points your desired user base has with others and go after those.  You wont get real accurate data, but you may be able to get at some pain points.   Good luck, thats a tough position to be in. 

9 Feb 2012 - 3:37pm

Perhaps you could reach out to the users via a survey or social features?

9 Feb 2012 - 4:32pm
Kristin Johnson

I'm in the same boat as you with an upcoming project. Here are some things we've been doing to get insight:

1) Find industry journal articles, surveys, and other sources of professional insight for this group and comb through for goals, behaviors and attitudes as they apply to your specific project.

2) Evaluate third-party sites and services (when they're accessible) and extrapolate the user needs from the features provided (this, of course, assumes that these sites have done their own research and haven't just thrown things together :)

3) Reach out to friends and family who may know of acquaintances who are in the same professional sphere. I also heard of someone using either Twitter or Linked In to pose a general question to the audience in question, and they were surprised the the number of responses they got. Another way to do this would be to post a brief survey/questionnaire with your most critical questions and include that link in your outbound communication and let them provide feedback that way.

Good luck. Let me know if you come up with other methods. I'd love to hear them.

10 Feb 2012 - 4:03am
Henri Huttunen

Thanks for all the replies! Glad to see I'm not alone with this. I'd imagine anyone who's ever battled in the trenches of UX "guerilla warfare" has come across this scenario at some point. We're getting good momentum going for UX in the company, but still have a ways to go before we can really apply all the good stuff. In a sense, we have to make it with "halved rations".

Right now our initial personas are built on internal stakeholder interviews, i.e., conjecture. We're expanding this to those users we do have access to, but according to our current understanding we are unlikely to get views on representatives of 50-60% of the user base.

The problem of access lies mostly in that the elusive group has a large share of corporate VIPs and senior management. Replies to online questionnaires and chances for personal interviews are all very scarce.

So perhaps I should rephrase the original question as "How to design emphatically for end-users using only desktop research?"

As a sidenote, resulting from the challenges faced in persona creation, I've noticed I'm thinking more and more about ways to quantify user interaction and behavior more granularly. So, in lack of personal access to the users, a UX performer would keep a very close eye on the information generated by what users do in the system and use a variety of statistical tools to analyze that information.

Sometimes, when I'm in my happy place, statistical analysis of our event logs confirms or disproves the conjectured behavior of our personas. 

10 Feb 2012 - 2:59pm
Kristin Johnson

Well, given the additional detail you provided, you could also try a goal-based approach. Obviously, you're going to have to guess at those goals ("I, Senior Manager of Group X, want to maintain my coveted reputation as an on-time, under-budget guy"). But if you got enough down as part of a brainstorm and then figured out which ones most directly impact your product/project, you could use those to shape the final direction.

And while you might not have access to a large swath of execs, you  must have some at your own company. Do whatever begging/borrowing/pleading you have to do to get 15 mins of face time with a few to validate your top (assumed) goals. (And think of the networking possibilities.)

I don't know if it's possible to use your existing metrics as another way to validate the goals...

Fun times. Stay brave. 


Syndicate content Get the feed