Jobs to be done: the real job your SaaS must solve

Your product has ten features. The one across the street has three. And yet it's the one signing customers. The reason sits in an idea from Clayton Christensen, the Harvard professor behind disruptive innovation theory: people don't buy a product, they "hire" something to get a specific job done in their life. The jobs to be done (JTBD) framework starts there, and it completely reorders how you build your SaaS.

What "jobs to be done" actually means

The best-known example comes from a fast-food chain that wanted to sell more milkshakes. After months of tuning taste and price, nothing moved. Christensen and his team asked a different question: what job are people trying to get done when they buy this milkshake? The answer was surprising. Half the milkshakes were sold before 9 a.m., bought by people alone who immediately drove off. The job wasn't "drink something tasty". It was "make a boring commute go by and hold off hunger until lunch", as Harvard Business School tells it.

So the milkshake's real competitor wasn't another milkshake. It was a banana, a donut, or the boredom of the drive. Once the job was understood, the fix became obvious: make the milkshake thicker so it lasts longer, put it in self-service so people move faster. The lesson for you: as long as you reason in features, you compare your SaaS to the wrong competitors and miss what actually drives the purchase.

People don't want a drill. They want a hole in their wall.

A job is neither a vague need nor a technical task. It's a progress someone is trying to make in a given situation. "Run my payroll without errors", "hand in a client report that looks pro", "know if my ad is profitable before spending again". The product is just a means. If another means does the job better or faster, it wins.

Why frame the job before writing a line of code

The reflex when you can build is to add one more feature. It's reassuring: you control your editor, not a stranger's reaction. Except that's exactly where most products die. They're well made, but they do nobody's job.

The numbers are brutal. Among the startups analyzed by CB Insights, 43% of those that shut down cite poor product-market fit as a root cause: they built something the market didn't want badly enough. And according to Christensen's work relayed by MIT Professional Education, roughly 95% of new products launched each year fail, often because they start from the founder's idea rather than the customer's job.

Framing the job first is the vaccine against that. You stop asking "which feature do I add?" and start asking "which job do I move forward, for whom, and better than what?". That single shift saves you months spent polishing a product nobody hires.

How to formulate your SaaS's job

Good news: formulating a job doesn't require heavy methodology. A formula, a few conversations, and you hold the essentials. Here's the sequence to run this week.

The three dimensions of a job (and why you miss them)

A job is never purely functional. When someone hires you, three things play out at once. Most founders only work the first and then wonder why their "logical" product doesn't convert.

Take a reporting tool for freelancers. The functional job is "generate a report". The emotional job is "stop being embarrassed by my patched-together deliverables". The social job is "look like a real agency to my client". If your sales page only speaks to the functional part, you leave two thirds of the buying motivation on the table.

From the job to your offer, persona and message

Once the job is framed, everything else follows. That's where JTBD becomes a distribution tool, not just a product one. Here's how the job feeds every decision.

The job becomes your compass. Every feature idea passes through a single filter: does it do the job better, or does it flatter my urge to code? That discipline is what separates a product that finds product-market fit from a rich product nobody adopts.

Your checklist to frame the job

Tick them off as you go. The goal isn't theoretical perfection, it's having a job sentence clear enough to test with real people this week.

How it ties into the rest of your strategy

The job to be done isn't an isolated exercise. It directly feeds your marketing persona, which answers "who lives this job most often". It also gives the raw material for your value proposition: a promise anchored on the progress sought always converts better than a feature list. And before you even code, it structures your market research: you're not asking "do people like my product", but "is this job painful enough that someone will pay to get it done".

Framing the job is half the road. The other half is finding the channel where people are already trying to get this job done, and speaking to them in their words. That's exactly what the diagnostic below is for.