Bob Moesta on Jobs to Be Done

Source: Lenny’s Podcast Speaker: Bob Moesta Source URL: https://www.lennyspodcast.com/bob-moesta/

Key ideas

Jobs to Be Done (JTBD) core: people hire products to make progress in their lives. The unit of analysis is not a feature or a demographic but a struggling moment — the context+outcome vector that causes someone to act. Pain-and-gain is the wrong frame; context-and-outcome is correct.
Four forces of progress: Push (F1) and Pull (F2) drive switching; Anxiety of the new (F3) and Habit of the present (F4) resist it. A switch happens only when F1+F2 > F3+F4. Adding features increases F2 but also F3; reducing friction (storage, dining table) often works better.
Six phases of buying: First thought → Passive looking → Active looking → Deciding → First use → Ongoing use. Matching your sales/onboarding process to the buyer’s phase — not your own sales funnel — halved Autobooks’ sales cycle and 4×‘d conversion.
“Bitchin’ ain’t switchin’”: complaints do not predict behaviour change. Only study people who actually made the switch; extract the causal story through interrogation-as-therapy, not stated preference surveys.
Demand side vs. supply side: competition is defined from the demand side (what would someone stop using to hire your product?), not the supply side (product category / ingredient similarity). Snickers competes with protein drinks, not Milky Way.

What is JTBD?

Bob Moesta co-created Jobs to Be Done with Clayton Christensen. The core premise: supply and demand are less connected than most builders assume. Demand is created by struggling moments, not by product launches.

“Build it and they will come” is the lie engineers are taught. The reality: a struggling moment causes demand. Products are hired to fulfil that demand. If there’s no struggling moment, the product is invisible — people are creatures of habit who can’t see a new product unless something pushes them to look.

The Snickers vs. Milky Way example: product-centric benchmarking finds minor differences (peanuts, calories, texture). JTBD reveals that Snickers is hired as a meal replacement in high-output, time-constrained contexts (competes with protein drinks and Red Bull), while Milky Way is hired as emotional recentring after a stressful event (competes with wine, brownies, a run). The two candy bars are barely in competition.

Four forces model

When someone considers switching from an old product to a new one, four forces are in play:

Force	Label	Effect
Push	F1	Context making the current solution untenable — the only reason they’d leave
Pull	F2	Attraction to a new outcome
Anxiety of the new	F3	Uncertainty whether the new solution will work
Habit of the present	F4	Inertia — cost of changing familiar routines and artefacts

Switch happens only when F1+F2 > F3+F4. Most teams focus entirely on F2 (more features, better product). But:

Adding features increases F3 (can it really do all that?).
Reducing F4 is often cheaper and faster.

Condo case: downsizers said they didn’t want a dining table; research revealed the table was “the emotional bank account of their entire life.” Moesta sacrificed the second bedroom to include a dining area → 22% sales increase. Separately, raising the condo price to include 2 years of storage eliminated the friction of clearing out the old home → 30% sales increase.

Six phases of buying

First thought — vague awareness something needs to change
Passive looking — problem-aware, solution-unaware; absorbing context
Active looking — comparing alternatives
Deciding — making trade-offs
First use — peak anxiety of the new
Ongoing use — habit formation

Autobooks case: the sales team’s process was: qualify → demo → close. Mapping the demo to buyer phase instead — (1) story/background for passive lookers, (2) alternatives overview for active lookers, (3) choice-framing for deciders — halved the sales cycle and 4×‘d conversion.

How to interview for JTBD

Who: people who recently made the switch (purchased or churned). Never people who say they want to switch — “Bitchin’ ain’t switchin’.” Stated preferences are unreliable; only past behaviour is causally informative.

How many: 10–12 per job cluster. Patterns repeat around 7–8. Two rounds of 12 is preferable to one round of 24.

Method — extract the story:

No discussion guide. Follow the most information-rich thread.
Use misparaphrase: play back the story incorrectly → they correct and elaborate. The moment someone says yes, the conversation stops.
Use bracket technique at the “edge of language”: “was it more X or more Y?” — neither is right; both prompt more disclosure.
Three layers of language to move through: (1) pablum (surface pleasantries), (2) fantasy/nightmare (exaggerated), (3) what actually happened (the causal story).
Criminal and intelligence interrogation that feels like therapy.

Reference: Never Split the Difference by Chris Voss.

For zero-to-one products: study what people will fire when your product launches. For Facebook Marketplace: studied eBay, Etsy, and Craigslist sellers. Jobs exist before products; find the struggling moment first.

Demand side vs. supply side

Supply side: technology, features, business model — how we build things. Demand side: struggling moments, contexts, outcomes — why people change.

Most companies define competitors from the supply side (similar product category, similar technology). JTBD defines competitors from the demand side: what would someone stop using if they hired you?

Intercom: JTBD revealed four distinct jobs. Rather than one product for all, they turned off features not relevant to each job pathway. One job competed with HubSpot (acquisition); another with Zendesk (support). Pricing model aligned to each competitive frame.

Clustering, not segmenting

Segmentation groups by demographics → averages that don’t reveal jobs. Clustering groups by shared push+pull pathways → actionable job profiles.

Most products are hired to do 3–5 jobs in conflict with each other (speed vs. thoroughness). Identifying these conflicts enables deliberate product trade-offs.

“Choose what to suck at and figure out the trade-offs that you need to make and make sure that your trade-offs map the trade-offs of the customer.”

Struggling moments as roadmap

A roadmap of features is brittle (ChatGPT wasn’t on anyone’s roadmap). A roadmap of struggling moments is durable — a struggling moment can be fulfilled by multiple technological solutions; which one depends on what’s currently available.

When Moesta works with companies on product strategy, he talks about features only for the first 90–120 days. After that, the roadmap is expressed as struggling moments because that’s “the seed for real innovation.”

When JTBD doesn’t apply

No real choice (employer-provided benefits, heavily regulated categories).
Deep habitual categories where people can’t recall the purchase moment (chewing gum). Study the little hire (how they use it) rather than the big hire (why they bought it).
When teams want JTBD to validate a pre-formed hypothesis — this corrupts the method. JTBD is hypothesis-building research, not hypothesis-testing.

Three flavours of JTBD

Moesta/Christensen approach: qualitative, causal, story-based. Starts from the demand side. Hypothesis-building.
Ulwick (Outcome-Driven Innovation): starts from product functions; systematic, quantitative, many steps. Better suited to regulated, complex systems. Only people have jobs; Ulwick’s version can veer into products having jobs.
Christensen (strategic): more philosophical/theoretical framing. Moesta turned the thinking into a tactical method.

Key books and resources

Demand Side Sales (Moesta) — flips sales process to buyer’s timeline
Competing Against Luck (Christensen et al., with Moesta’s clients referenced) — strategic framing
Never Split the Difference (Voss) — interrogation techniques for JTBD interviews
End of Average (Todd Rose) — recommended annually by Moesta