Jobs to be Done as Psychological Framework

When two worlds unexpectedly collide

Introduction

I’m quite skeptical any time “personas” are invoked within an organization. The marketing team spends an inordinate amount of time attempting to distill in far-too-precise terms who the (fictional) core user is for a given product and, almost without exception, summarily discards this work because, well, no one actually knows how to use them¹. 

I was quite surprised and delighted while revisiting the Jobs to be Done (JTBD) framework that Intercom’s CPO Paul Adams says quite explicitly that they do not use personas:

“Personas look at roles and attributes. Jobs to be done looks at situations and motivations. Personas explain who people are and what people do. But they never fully explain why people do something”

What ultimately struck me in this timely reading was not just the utility of the JTBD framework itself but how closely the personas criticism echoes my criticisms in Psychology has a Context Problem last week. 

JTBD as Psychology Framework

Remember again that our base equation for understanding Behavior is: 

B = P x E

To understand behavior, it is wholly insufficient to study the person only. Our behavior is heavily context dependent, and in fact context-resistant behavioral persistence is often a signal of mental illness. Our personality thus represents some base tendencies, coded at the genetic and neurological levels, but the expression of this foundational layer depends on our environment. 

Though Intercom didn’t create JTBD, the company is perhaps the cleanest instantiation of the framework as a company, and they created the Job Story to more cleanly translate the principles of JTBD into their day-to-day working environment. The Job Story itself looks a bit like an equation: 

In this construction we have three core variables: 

• When = context²

• I want to = motivation

• So I can = outcome

Note that a definition of the given user’s traits is nowhere to be found here, because it frankly doesn’t matter. The Job Story specifically, and Jobs to be Done more broadly, is an acknowledgement of two truths with which we should grapple: 

• Any given behavior is highly likely to be expressed by many “distinct” personas. 

• The selection of any single persona necessarily limits the size of the addressable market.

Taken together, Personas as key tool has in my eyes far outlived its utility. So why do they persist?

“I prefer what is not in our best interest”

Early in my adtech career I led many of our business development efforts for performance marketing clients, primarily game developers who categorically cared only about results – a maniacal focus on the triumvirate of app volume, prices, and user value. This was a wholly quantitative endeavor, and as such, any personal characteristics of the given user were wholly unimportant – all that mattered was outcomes. 

As the resident performance expert, I was brought into meetings with our Brand team to ostensibly educate their clients on how mobile performance marketing worked in order to secure test budgets we hoped would balloon into much larger, guaranteed spends. For unlike their Performance-focused counterparts, Brand buys (historically and to this day primarily driven by agencies) are typically top-of-funnel efforts to improve awareness. Success is not performance outcome but guarantees of eyeballs at a given price; Cost per impression versus cost per action.  

On our introductory call with this agency representing a major restaurant chain that shall not be named, I explained how we of the Performance ilk operate, and the types of KPIs on which we would typically focus. My naive brain was not really prepared for the unexpected question that kept coming thrown my way, over and over; a question that my today self would call the persona question – 

“You want to drive downloads for your first consumer app, right?” 

“Yup”

“We can easily do that at x scale and y price point, and you only pay when we deliver³” 

“That’s amazing”

Pause to denote a but is forthcoming…

“But how do we know that each user is a 25-44 year old male who loves skateboarding?”

Now, at this point in my career I was not well-versed in personas nor in more traditional brand marketing. And I was, let’s say, exasperated by this entire line of questioning. 

I tried to protest, less tactfully than I probably would today: 

“Why do you care about demos if the user is high quality?” 

“Well because it’s our core demo, that’s what we’re looking for”

“So let me get this straight. If we deliver you a thousand users, each of whom spends an incremental $1000 more per year because of this app, but they’re all 70 year old grandmas, that’s not what you want?” 

Uncomfortable silence. 

“We would prefer if they were in our target demo”

The Power of Defaults

In business and in sport, the playbook is a key piece of operations. It represents the dependable, repeatable behaviors that all team members are expected to follow with the oft implicit understanding that doing so will yield optimal results. The playbook is effectively an aggregation of defaults. 

Defaults are powerful because behavioral inertia is so difficult to break. Something as simple as switching browsers can be a completely infuriating experience precisely because so many of the defaults within this new context deviate from our formerly internalized defaults⁴. 

At one point in history, personas were viewed as a useful innovation; a necessary transitional vector for companies to adopt a more customer-first modality. And because we’re all mimetic creatures, any innovation can (hypothetically) generate a groundswell of adoption independent of its actual value.

This is not an anthropological tracing of persona lineage, of course — I’m sure there were more principled reasons for adoption, given the absence of other more useful frameworks like JTBD (to say nothing of the relatively data-poor environments in which most mid-century companies operated).

Regardless of the pathway, it is inarguable that personas are a default within the industry today, though a begrudging one. As so many who ever worked in a persona creation process will swiftly attest, it’s super difficult to buck the norm – “We haven’t really found value here, but there must be a reason why everyone uses them, right?!”

This is ultimately what makes Intercom’s decision to completely eschew personas that much more remarkable, and yet necessary. Continuing to dabble with an ineffectual framework or tool can be more damaging that full adoption, because it prevents the team from committing to anything. No, the better — and necessary — approach is to summarily reject the offending framework and banish its invocation from the premises.

1. Or perhaps more accurately, no one knows whether they can actually be used.

2. It's perhaps confusing because when signals a temporal variable, but when in this case is doing a LOT of heavy lifting in describing the environment in which the given user finds herself.

3. I know how trite this sounds, but CPA pricing was, back in 2012-2014, still an oddity.

4. As a begrudging yet long-term Chrome user, I will simply never get used to the close tab X being located on the LEFT in Safari. Infuriating.