B2B brand strategy: how positioning actually gets decided

Positioning gets decided in four moves, and the order is what makes them work. Follow all four and you end with a position that strangers can repeat back correctly on first read, the test that shows it will hold up outside your own building.

You need lost deals, five strangers, and your leadership team in one room

Three inputs, none of them a workshop template: a list of deals you closed and lost in the last two quarters, enough to reach 10 to 15 win interviews and 8 to 12 loss interviews; five to ten people who fit your ICP and have never been in your pipeline; and your leadership team with the CEO in the room. Book the buyer conversations first; they set the calendar.

This piece covers how a position gets decided. When to reopen one belongs to this pillar's cornerstone, Brand refresh: when and how to renew a brand without rebranding.

Name the alternatives your buyer actually considers

Write down what your buyer would do if your product disappeared tomorrow, and keep only the options they genuinely weigh.

This goes first because every later decision hangs off it: your differentiators are only differentiated against something. April Dunford settled on it after two years of testing candidates from arbitrary starting points, and her verdict on any other opening move is that you end up with "positioning that sounded good in the office, but it didn't work with customers because it wasn't differentiated."

She splits the buyer's options into three, and two get collapsed constantly. Stay with what they have, usually a spreadsheet or an incumbent. Pick a new vendor off a shortlist they built themselves. Or drop the purchase entirely, which Dunford separates from the status quo because the mechanism differs: "the prospect has not been able to confidently make a decision."

Then cut every name your buyers never mention. Dunford's shortlist rule belongs to the customer: "if a competitor rarely or never lands on a short list ... they simply aren't a competitor you have to position against."

Pull the reasons out of deals you already won and lost

Interview whoever made the buying decision on recent wins and losses, and let their answers overwrite your team's theories.

Product marketer Zack Alami is blunt about the default: "Most PMMs build positioning on guesswork." His program runs 10 to 15 win conversations and 8 to 12 loss conversations per round, wins within 90 days of close and losses at 30 to 45 days, with themes repeating after five or six in each group. Go to whoever signed or pushed hardest for the purchase; implementation contacts cannot say why it got bought.

Two questions carry the weight: on wins, whether they'd still have chosen you without your biggest feature, which separates core value from a nice-to-have; on losses, what would have needed to be different for you to win.

Read price answers carefully. "A loss that feels like 'they wanted a lower price' is often actually 'they didn't see why our solution justified premium pricing,'" Alami writes, filing that under positioning and never under pricing. His follow-up does the work: at identical prices, would they still have picked the other vendor?

Make the call in one room, with the CEO in it

Get leadership and the CEO into one session, put the evidence on the table, and leave with one written position.

Ellie Victor, CEO of the positioning consultancy ZOOM Marketing, makes that a precondition, per Renegade Marketing's account of her method: "If your CEO is not the sponsor of the positioning process, not just an approver but an active participant, do not start. This is not a marketing exercise. Marketing shepherds the process. The CEO owns the outcome." Sponsorship means sitting in the session; signing off afterwards fails it.

Open with a fast diagnostic you can run in the same session. Hand every leader an index card, ask them to write the category the company competes in and what they most want the market to associate it with, then read the cards aloud. Ellie Victor asked a room of senior B2B CMOs at a CMO Huddles Strategy Lab what a scattered result means: "They're all executing against their own idea."

Build from those cards before importing an outside answer: a position developed elsewhere, then presented as finished, gets rejected by the CTO for missing the product vision.

Apply two tests before the room breaks up. Victor's test, as Renegade Marketing frames it: "If you removed your brand name from the statement, could a competitor say it? If so, it's not a position, it's a description." Then check it against her three conditions: you hold the advantage today, at least relative to your most direct competitors, your roadmap keeps building it, and competitors cannot or will not claim it.

Test the wording on buyers who have never met you

Show the finished statement to five to ten ICP buyers outside your pipeline and ask three questions, unexplained.

Greg Rosner, founder of PitchKitchen, calls it the Three Questions Test: who is this for, what problem does this company solve, and what makes their approach different from other options you have seen. His case for strangers is that everyone inside the company crossed the comprehension threshold long ago: "A message that survives an internal debate has been optimized for internal buy-in, not buyer comprehension."

Read the answers as a pattern: one wrong answer is noise, but three of five wrong in the same direction is Rosner's threshold. His most-reported result is buyers clearing the first two questions and failing the third, meaning the category landed and the differentiator didn't.

Victor scales the idea to a panel of ICP-matched prospects who score each candidate: winners typically land in the 80 to 90 percent range, and anything below 50 percent is usually dead. Renegade Marketing's rationale for scoring prospects here: "customers have already bought from you, so they'll be generous."

You know it worked when strangers can repeat your position back

The exit signal is same-day and concrete: five ICP buyers who have never met you answer all three questions correctly on first read, and every leader's index card says the same thing. Rosner's guidance: two rounds of testing is usually enough, and identical drift after the second points at the underlying logic below the wording.

Market movement is slower, and none of these practitioners publishes a measured lag between shipping a position and seeing win rates move. Alami runs win/loss rounds twice a year, which is where you find out.

Two failure modes undo an otherwise sound decision

Letting internal consensus stand in for buyer evidence. Rosner's framing is that a room with no external test is running an opinion contest, and opinion contests have a predictable winner. Wynter's 2025 survey backs the mechanism: when its 100 leaders explained their own indistinct messaging, the top reason they gave was fear of risk and herd mentality, at 36%, ahead of a genuine lack of differentiation.

Approval by committee after the fact. Victor's pattern, reported by Renegade Marketing: a small team develops the position and presents it to leadership for approval. "The CRO finds something missing, the CTO says it doesn't match the product vision, and the CEO defers to whoever spoke last. The project stalls, gets revised by committee, and produces something long enough that nobody can remember it." The fix is sequencing, which is why the CEO sits in the room from the start.