Skip to contentBoring Marketing LoopIn development

Marketing strategy

How to write a value proposition you can test

By Boring Marketing Loop · Published

How to write a value proposition in one sentence, how long it should be, and how to test it with five people before it goes on your landing page.

What is a value proposition?

A value proposition is the reason a specific customer should choose your product over what they do today. It names the customer, the result they get, and why your product gets them that result better than their current alternative. For a startup, a value proposition is also a claim you have not proven yet, so the useful way to write one is in a form you can test.

Most guides to writing a value proposition give you a formula and a list of famous examples from large companies. Those examples worked because the companies already knew their customers. This guide shows how to write a value proposition in one sentence, check that each part has a source, and test it with real people before it goes on your landing page.

Why most value propositions do not work

In a 2006 Harvard Business Review article, Customer Value Propositions in Business Markets, James Anderson, James Narus, and Wouter van Rossum reported that "there is no agreement as to what constitutes a customer value proposition." They also found that "most value propositions make claims of savings and benefits to the customer without backing them up." Customers, they warned, will likely dismiss such claims as "marketing puffery."

The article is twenty years old and studied business markets, but the problem it describes is easy to spot on startup landing pages today. "Save time." "Grow faster." "All-in-one platform." Each phrase could appear on a thousand other pages. None of them names who the product is for or what it replaces.

A value proposition fails for one of three reasons:

  • It names no customer. "Teams" or "businesses" is everyone, so the reader cannot tell whether it is for them.
  • It names no alternative. Every customer already does something about the problem, even if that is nothing. A value proposition that ignores the alternative cannot say why yours is better.
  • It claims a benefit without a reason. "Save 10 hours a week" with nothing behind it reads as puffery.

The three parts of a value proposition

April Dunford, who writes about product positioning, works in a fixed order in her quickstart guide to positioning (2021). She starts with the competitive alternatives, then the capabilities that make the product different, then the value those capabilities create, and then the customers who care most about that value. She puts it this way: "the unique value that you can provide to customers is completely dependent on your differentiated features."

For a one-sentence value proposition, that order gives you three parts. The template below splits the second part into two lines.

  1. The customer and their situation. Who has the problem, described by something they do.
  2. The result compared with their alternative. What gets better for them, compared with what they do today.
  3. The reason. The specific thing your product does that the alternative does not.

A value proposition template

Write it as one sentence first, then shorten it for your headline.

For [customer] who [situation or trigger],
[product] [result they get],
unlike [what they do today],
because [what your product does differently].

Here is the template filled in for a hypothetical tool. The details are examples, not research.

For independent consultants who send three or more proposals a month,
FollowUp gets every proposal a reply or a clear no,
unlike a spreadsheet and calendar reminders they forget to check,
because it drafts the follow-up email and sends it on day three
unless they stop it.

The full sentence is too long for a headline, and that is fine. It is a working document. The headline comes from it: "Every proposal gets a reply. Follow-ups drafted and sent for you." Each word in the headline now traces back to a part of the sentence that you can test.

Check where each part came from

Before you test the value proposition, write down the source of each part. This takes five minutes and shows you which part to test first.

PartOur exampleSourceStatus
CustomerConsultants sending 3+ proposals a month6 interviewsObserved
SituationProposals that go quiet after sending4 of 6 interviewsObserved
AlternativeSpreadsheet and calendar reminders5 of 6 interviewsObserved
ResultEvery proposal gets a reply or a clear noOur guessGuess
ReasonDrafts and sends the follow-up on day threeWhat we plan to buildGuess

The numbers in this table are illustrations. In this example, the customer and the alternative have support from interviews. The result is a guess, so it is the part to test. If you have no interviews yet, start with our customer persona template, which uses the same guess and observed labels.

How to test a value proposition

Strategyzer, the company behind the Value Proposition Canvas, says in its description of the canvas that "fit is a claim until customers confirm it." Here are three ways to check the claim, from cheapest to most demanding.

  1. Say it to people and ask them to repeat it back. Send the one-sentence version to five people who match the customer description and ask, "What do you think this does, and who is it for?" If they cannot answer, the sentence is unclear. This tests clarity, not demand.
  2. Use it as the opening of an outreach message. Send the headline version to 30 people from a list like the one in our guide on how to find customers for a startup. Count replies, and note which part of the sentence people respond to.
  3. Put it on a landing page with one next step. Count visitors from one source and how many take the step. Our guide to testing demand with a landing page covers how to set up the page and the measurement.

Write the decision rule before each test. For example, "If 4 of 5 people describe the product correctly, move to outreach. If 2 or fewer do, rewrite the result part." The numbers are an example. Pick ones that would change what you do.

Each test checks a different thing. The first checks whether people understand the sentence. The second and third check whether the result matters enough for them to act. A clear value proposition can still fail the second test, and that tells you the result you promised is not the one they care about.

Put the value proposition at the top of the page

Jakob Nielsen's 2002 homepage usability guidelines say to "start the page with a tagline that summarizes what the site or company does, especially if you're new or less than famous." The guideline is more than twenty years old, and it still holds for a startup. A visitor who does not know you has no other way to tell what you do.

On our own home page, the headline is "Test your startup idea." The line under it says we are building "a workspace for small marketing tests" and tells the reader what they would do in it. The headline is short, and the line beneath it does the explaining. We have not measured how well it works yet, so it is still a guess on our side too.

Value proposition examples: before and after

These are hypothetical rewrites, showing how naming the customer, the alternative, and the reason changes the sentence.

BeforeAfter
"The all-in-one platform for freelancers""Send an invoice from the same page as your time log, so you stop retyping hours into a spreadsheet"
"Save time on social media""Write a week of LinkedIn posts from one recorded call, instead of starting from a blank page each morning"
"Smarter scheduling for teams""Clinics fill a cancelled appointment from the waitlist by text, without the front desk calling anyone"

Each rewrite is longer, and each one tells the reader who it is for and what it replaces. You can shorten it later, once a test shows which part people respond to.

Keep the value proposition attached to the tests

A value proposition changes as you learn. The risk is that the version on your landing page drifts away from what your tests showed, and you no longer remember why you wrote it the way you did.

We are building Boring Marketing Loop to keep those connected. The workspace asks nine Customer & offer questions about who you help, what you offer, and what they use today, and each answer is marked as a guess until people respond. A test records the message or link, the sample size, an end date, and the signal you chose before it started. After the result, you decide to continue, change direction, stop, or gather more evidence, and the next test is planned from that decision.

Marketing Loop is in development. If you want to know when it is ready, join the build updates.

FAQ

What is the difference between a value proposition and a tagline?

The value proposition is the full argument: customer, result, alternative, and reason. A tagline or headline is the short version people see first. Write the value proposition first and cut the tagline from it.

What is the difference between a value proposition and positioning?

Positioning is the wider set of choices behind the value proposition: which alternatives you compare yourself with, which customers you focus on, and which market category you place yourself in. The value proposition is how you state the result of those choices to a customer.

How long should a value proposition be?

As a working document, one sentence that may run to 40 words. On the page, a headline of a few words plus one line underneath. If you cannot write the long sentence, you are not ready to shorten it.

What is a unique value proposition?

A value proposition that states something only your product does, compared with the customer's real alternative. In practice, every value proposition should be unique in this sense. If a competitor could put your sentence on their page unchanged, it needs a better reason.

Write the long sentence first

Fill in all four lines of the template today. Then add the source and status table, find the part marked as a guess, and test it with five people this week. Shorten the sentence into a headline only after a test shows which part people respond to.