The Job-to-be-Done is the Unit of Analysis

What's up? It's been awhile.

There are a few common things folks ping me about:
1. design and user experience research in libraries,
2. links should open in the same window almost always,
3. how I don't think user personas are useful,
4. and the thing that I said once about how "in customer service, the 'customer is always right;' but in user experience design, the user isn't."

The last two make me sound super cranky, but -- hey -- if you want to design for happiness and ensure that the folks paying you don't get their return on investment, by all means ....

In person I would say this ^ with the kind of face and tone to signal I'm sooo not the jerk this makes me sound like, but ah - well, please take my word for it.

Anyway, these last two opinions -- that user personas aren't particularly useful and that the user experience isn't necessarily situated around user _happiness_ -- are rooted in a strategic approach we call Jobs to be Done.

When people first get into and start thinking whimsically about the user experience, the immediate googleables are what you expect: personas, card sorting, interviews, and so on, which underscore our totally sane first impression that good user experience design is about the user's ephemeral experience in the superficially delighted-by-the-user-interface way. This is, in part, the fault of our vocabulary, and why "service design" -- which, I don't care what any of you service designers say, is pretty much user experience design in practice with a _slightly_ different metric of success - but a much better name -- is an increasingly popular label.

So, it makes sense that with this assumption that personas are popular because the user's experience is the unit of analysis. The more you know about the user's personality, history, and influences, the better you can predict that experience.

From where I stand, where budgets are constrained and there's a high-level of skepticism, when the investment in user experience design needs to pay off, this approach is kind of listless. In practice, user experience designers need to show profitability. That is, sometimes, in financial terms, but even when pure dollars aren't of concern there is a suite of success metrics -- numbers people care about, like newsletter subscribers (ahem) -- that need to move up to justify the costs -- human and financial.

It's really hard to make effective design decisions around subjects in constant flux. Delight and expectation are so determined by external pressures and the experience of other people's shit [1], that for the service to adapt constantly and successfully requires huge investment in the systems underlying it, which are often too costly to consider seriously.

Thus, without an alternative, user-centricity (and thus user experience design) has to be put on the back burner. Even though good user experience correlates to good business, for many it's just too expensive to maintain when good user experience requires constant vigil and change [2].

That alternative doesn't need to eschew delight and that ephemeral experience, but shifts the focus of your service or product from the user to the problem he or she needs your service to solve. Folks don't use your library for the sweet decor and bomb collection: they use it to distract from their shitty day; to nail that job interview; to better enjoy their honeymoon in Paris because they can order off the menu. The services you provide are a means to an end; they help get the job done.

There is still a user experience that is - ah - experienced, but that positive or negative metric is abstracted from the overwhelming number of shifting variables that define a person to how well your service or product got that job done.

This provides a stable focal point to which the business of your service can respond while user experience remains an important signal of its success.

The role of the user experience designer is then to identify that job, whether there are enough people who need it done to get into the market, and then how best to do it. Jobs remain unmoving: they need to be done - they're being done right now by someone else.



1. Libraries are still so hard-up about chasing that one-search-box-to-rule-them-all because their users expect Google and Amazon, that they will drop big dollars on indexing software ["discovery layers"] that in many user tests aren't any more easier to use or useful than before.

2. Ironically, vigil and adaptability are the core selling-points behind design thinking.

Thank you so much for subscribing. Feel free to shoot me an email any time.

Your pal,

Michael