Value Proposition Canvas: designing a value proposition

How to fill out the canvas so that it actually works

The right side of the canvas is the customer profile: jobs, pains, gains. The left side is the value map: products, pain relievers, gain creators. Work strictly from right to left. A team that starts with the left side — 'we already have a feature, let's come up with who it's for' — almost guaranteed wastes the budget solving a problem that nobody has.

In practice, almost every second canvas breaks down here, and it breaks down for the same reasons.

• Job is replaced by a feature. ‘Set up a push notification’ is not a customer task, it's a button in the interface. The task sounds different: ‘to find out about important messages without opening the app every five minutes’. The difference between these formulations determines whether the product will solve a real problem or just accumulate settings.
• The canvas is filled out without a single interview. All pains and gains are hypotheses of the founder, written in a beautiful font. The benchmark for the B2C-product is a minimum of 5–10 deep interviews per segment, for B2B — 8–15, because the decision-making cycle is longer and several roles are involved.
• The card is filled with 12–15 pains of equal importance. The product attempts to address them all at once and ultimately fails to convincingly close any of them. The value proposition blurs into a generic phrase like 'convenient solution for business'.
• The hypotheses are tested 'by eye'. Validity is confirmed by retention at week 4, NPS, CSAT, conversion from trial to payment — not by the founder's opinion that 'the idea is definitely viable'.

Working rule: minimal viable Fit is a confident closure of 2–3 of truly critical client issues, not a match by all items on the map. Chasing one hundred percent coverage of all pains almost always results in an overloaded, incoherent product — a classic example of 'we do everything for everyone'.

VPC in B2B: each stakeholder has their own canvas

In B2B one common canvas is almost always lying, because the purchase decision is made not by one person, but by three to five roles with different success criteria. Let's take a typical CRM implementation deal: a sales manager wants to enter call data faster, a department head wants to see quarterly revenue forecast, and a CFO wants to see return on investment within 6–9 months. If the product addresses only the first role, the deal will stall at the budget approval stage, even if frontline users are thrilled with the interface.

The right approach is to build a separate canvas for each role and then look for intersection points, rather than one averaged 'customer' map. The intersection point usually becomes the basis of the message for the landing page and sales presentation — something that is equally important to both the user and the person who signs the invoice.

Three levels of Fit by Osterwalder

Alexander Osterwalder describes product-market fit not as a one-time event, but as three sequential checks, and confusing them is a common mistake among startups.

1. Problem-Solution Fit. The hypothesis that the problem is real and the solution theoretically addresses it. Verified through interviews and a prototype, without code.
2. Product-Market Fit. The product is created, and the market actually buys it repeatedly. A classic benchmark is the Sean Ellis test: if at least 40% of users answer 'very upset' to the question 'how would you feel if the product disappeared,' this is a signal of sustainable PMF. Less than 40% — the market accepts the product, but doesn't want it.
3. Business Model Fit. Unit economics is positive, and customer acquisition channels are scalable — they work not only on personal connections and founders' enthusiasm, but on a clear, repeatable mechanism.

Osterwald separately describes four types of value in gain creators: necessary — without which the solution simply does not work; expected — the basic industry standard, without which the product is not even considered; desired — a pleasant bonus that increases loyalty; and unexpected — that very wow effect, due to which the client tells others about the product without being asked. In most canvases we see in practice, the third category is empty: teams bring the product up to the industry standard and stop, although it is precisely the unexpected benefit that often triggers organic word-of-mouth marketing.

How VPC differs from BMC, JTBD, and positioning

These four tools are constantly confused and used as interchangeable, although each has its own area of responsibility and its own moment in the process.

The logic of the linkage is simple: JTBD provides raw observations about the customer (a classic example from Clayton Christensen — people 'hire' a milkshake not for taste, but to make a boring morning commute into an engaging activity), VPC turns these observations into a structured product-need alignment, BMC checks whether this construct is viable as a business, and positioning wraps the validated value in a specific message for the market. VPC without BMC creates a product that customers like, but doesn't generate revenue. BMC without VPC builds an economically elegant model around something that nobody really needs.

What the canvas will never show

The tool has blind spots that are rarely written about — but they come at a high cost.

First: VPC does not adequately describe the emotional and status value. A luxury brand does not solve the customer's 'pain' in the rational sense of the word — it sells belonging and self-identification. Attempting to describe a bag priced at 300 thousand rubles through pain relievers ("eliminates the inconvenience of carrying things in your hands") looks absurd precisely because the purchase decision is inherently emotional, and the canvas is designed for rational jobs.

Second: the canvas does not suggest which specific job to pursue if the segment has ten of them. This is a strategic choice of niche and market positioning, not an analytical task solved by filling out forms. Here, brand strategy applies: choosing one job out of ten determines what the product will become in two to three years, and without a conscious decision, the canvas will simply record a random hypothesis that happened to appear in the document first.

Third: VPC requires a relatively stable, well-studied segment. On early, rapidly changing markets, filling out the canvas gives a false sense of certainty — the data becomes outdated faster than the team can react to it.

A separate practice — Competitive Value Map: overlaying value maps of several market players onto each other. For example, if you compare the maps of services for team collaboration such as Miro, Notion, and Trello, you can see that all three address the job 'structuring information visually,' but almost none systematically address the job 'synchronizing an asynchronous team across different time zones.' Such white spots at the intersection of maps often become the basis for new positioning — not because they guessed the perfect niche, but because there is currently no competition for the customer's attention there.

Why this is a living document, not a one-time presentation

Filling out the canvas before launch and archiving it — a common and failed practice. The client's pains and priorities change along with the market, competitors, and the product itself. VPC review is necessary after every significant shift: a major competitor's launch, pricing policy change, new feature release, retention drop, or NPS falling below target values. Many product teams have a rule to align the canvas with new data once per quarter — not rewrite from scratch, but check which hypotheses are confirmed by interviews and metrics, and which ones should be discarded. It is this regular review that distinguishes the canvas as a working tool from the canvas as a pretty picture for an investor presentation that was filled out once and never opened again.