A CMO at a Series B company told me last spring that they had finally fixed positioning. Two days offsite, a consultant, sticky notes, the whole thing. They came out with a one-page document, a new category name, and a homepage rewrite that shipped in six weeks. Everyone loved it. The CEO put a slide about it in the board deck.
Six months later I pulled their HubSpot data. Win rate: 19 percent before, 20 percent after. Average sales cycle: 94 days before, 97 days after. The share of qualified pipeline that closed as no decision had not moved at all. It sat at 46 percent both years.
The positioning document was not bad. I read it. It was clear, it was specific, it said something true about the product. The problem was that nobody ever checked whether it was right. It was a hypothesis that got treated as a conclusion, and the only evidence anyone collected was that the people in the room nodded.
I have spent over ten years in revenue operations, and this is the pattern I see most often with positioning work. Marketing owns it, treats it as a creative exercise, and hands the output to sales. Sales half-adopts it. Nobody instruments anything. Twelve months later somebody suggests another offsite.
Positioning is testable. The evidence is already sitting in your CRM. Most teams just never wired it up to collect the right thing.
What positioning actually decides
Skip the textbook definition. In practice, positioning answers one question for a buyer who has a problem and a budget: compared to the other things I could do about this, why you?
The best framework I know for the answer side is April Dunford's work in Obviously Awesome. She builds it bottom-up: start with your best existing customers, work out what alternatives they were weighing, find the capabilities you have that those alternatives do not, translate those into value, then decide which segment cares about that value most. It works because it starts from evidence rather than aspiration.
The part most teams skip is the alternatives. Not competitors. Alternatives. Dunford's point, which took me years to properly absorb, is that in B2B your main rival is usually a spreadsheet, a contractor, an intern, or a decision to do nothing this quarter. When you lose to a competitor you at least know why. When you lose to nothing, you rarely find out.
Share of qualified B2B pipeline that ends in no decision rather than a loss to a named competitor. Across most reported benchmarks, no-decision outcomes beat any single rival by two to three times.
That number is the single best positioning diagnostic you have, and almost nobody uses it that way. A high no-decision rate is not a sales execution problem. It is buyers who could not build a good enough internal story for why this, why you, why now. That is positioning.
The reason it stays high is that buying groups keep growing. Gartner's tracking has the typical committee moving from roughly 5 people a decade ago to 6 to 10 today, and 11 to 20 on larger enterprise deals. Your champion has to repeat your positioning to eight people who never met your rep, in a meeting you are not in, using words they remember. If your positioning takes a paragraph to explain, it does not survive that meeting. We wrote about the mechanics of this in the buying committee mapping guide.
Your CRM already knows, but it is not recording it
Here is the awkward part. The data that would prove or disprove your positioning is theoretically in your CRM, and in most B2B companies it is worthless.
Those figures come from win/loss research firms like Clozd and match what I find in nearly every audit I run. The lost-reason picklist is either empty or full of "price" and "timing", which are the two answers a buyer gives when they do not want to have a longer conversation with the rep who just lost.
"Price" almost never means price. It means the buyer could not connect what you do to a number big enough to justify the cost. That is a value articulation failure, which is positioning. "Timing" usually means no decision with better manners.
So the first job is not a positioning workshop. It is making your CRM capable of answering the question. If your lost-reason data is a mess, start with CRM data quality before you build anything on top of it.
The four fields that make positioning measurable
In HubSpot or Salesforce, on the deal object, I set these up on every engagement. If you are still choosing a system, the CRM build matters less than whether these four fields get filled in honestly.
Primary alternative. Not competitor. A picklist with entries like: internal build, spreadsheet or manual process, existing vendor renewal, agency or contractor, do nothing this cycle, and then your actual named competitors. Required on every deal that reaches your first qualified stage. This one field changes more positioning arguments than anything else, because most teams discover that their named competitors show up in under a third of deals.
Loss category. Six options, no more. No decision or no budget released. Lost to named competitor. Lost to internal build. Value not accepted. Requirement gap. Process or relationship failure. Six is deliberate. Twelve options produce noise because reps pick from the top.
Value claim tested. A multi-select of the two or three claims your positioning makes, filled in at proposal stage. Which one did this deal actually run on? If one claim is on 80 percent of won deals and 20 percent of lost ones, that claim is doing the work. If a claim never gets selected, it is decoration.
Champion's words. A free-text field, one sentence, captured after discovery. How did the buyer describe the problem in their own language? This is the highest-value field on the list and the one reps fight hardest. Make it required to move to proposal and it gets filled.
None of this is exotic. It is thirty minutes of CRM configuration and a conversation with your sales leader about why it matters. The cost is roughly ninety seconds per deal. The output is the only positioning evidence base you will ever have that is not somebody's memory.
Positioning is a hypothesis about why buyers pick you. Hypotheses get tested against data, not against a room.
Every positioning statement makes a falsifiable claim: that a specific segment weighing specific alternatives will choose you for a specific reason. All three parts are visible in closed-won and closed-lost data if you record them. Most teams argue about wording instead.
Testing the positioning you already have
Assume you have some positioning, even if it was never written down. Before you rewrite it, find out what is actually happening. This takes about two weeks with existing data and gives you a much better starting point than a blank page.
Pull your last 40 closed-won deals. Segment them by industry, company size, and buying trigger. What you are looking for is concentration. Most companies find that 60 to 70 percent of revenue comes from a segment nobody deliberately targeted. That segment is your positioning answer, and it has been hiding in the data for two years. This overlaps heavily with proper ICP work, which is the same exercise viewed from a different angle.
Pull your last 40 closed-lost deals and split them by loss category. If no decision is over 35 percent, your problem is the "why now" and "why change" part of positioning, not competitive differentiation. Teams get this backwards constantly. They build battlecards against a competitor who appears in 12 percent of deals, while 44 percent of pipeline dies against the status quo with no plan at all.
Run 10 buyer interviews. Five wins, five losses, and do not let the rep who owned the deal run them. Twenty to thirty minutes each. The one question that matters: what were you going to do instead, and what made you decide against that? Buyer-reported reasons and CRM-recorded reasons agree about 15 percent of the time, so the interviews are not optional decoration. Our win/loss analysis guide covers how to run these without turning them into a sales call.
Compare sales cycle length by segment. Positioning that lands shortens cycles because the buyer needs less internal convincing. If one segment closes in 40 days and another takes 110, the difference is rarely product fit. It is usually that the second segment does not have a ready-made internal story for you, so your champion is building one from scratch.
Check where deals die in the funnel. Losses concentrated at first meeting mean the market-facing positioning is attracting the wrong people. Losses at proposal mean the value claim is not surviving contact with finance. Losses after verbal commitment mean the champion could not carry the story to the committee. Three different failures, three different fixes, one useless conversation about "our messaging" if you do not separate them.
The 90-day loop
Once you have a hypothesis worth testing, run it as an experiment rather than a launch. This is the sequence I use with clients, and it fits inside one quarter.
Notice that the website rewrite is last. Every failed positioning project I have seen did it first. The homepage is the most expensive place to test a message and the slowest to give you a signal, because traffic is noisy, intent is mixed, and it takes a quarter before anything shows up in pipeline. Sales calls give you a read in three weeks and the feedback is a human telling you they do not get it.
If your team runs conversation intelligence, this gets much easier. Tag calls where a specific value claim was made, then look at stage progression by tag. I have built this in n8n against Gong and Fathom transcripts for a few clients: pull the transcript, run a classification prompt for which claim appeared, write the result back to the deal record. It removes the reporting burden from reps, which is usually the reason the data goes stale by week five.
Where this goes wrong
Positioning by committee. If six people have veto power over the wording, you end up with the intersection of six opinions, which is always the most generic version. Pick one owner. Everyone else gets input, not a vote.
Confusing positioning with taglines. Positioning determines who you sell to, which alternatives you fight, and what you claim. The tagline is a downstream artifact. Rewriting the tagline while the segment and alternative stay undefined changes nothing, which is exactly what happened to the CMO in my opening.
Positioning for the market you want. The move-upmarket positioning that a Series A team writes because the board wants enterprise logos is a reliable way to lose the SMB deals that were paying the bills, without winning any enterprise ones. Position where you win now. Move deliberately, one segment at a time, with the data to justify it.
Never revisiting it. Positioning has a shelf life of roughly 18 months at an early-stage company, shorter if a well-funded competitor enters your space. The four CRM fields make the review cheap: pull the report, see whether the alternative mix has shifted, decide if anything needs to change. That is a 90-minute exercise once a quarter, not another offsite.
Marketing owning it alone. Positioning that sales does not use is not positioning, it is copy. The person who has to say the words in a live call needs to be in the room while it gets written. Every time. This is the same sales and marketing alignment problem in a different costume.
What actually changed for that Series B company
We instrumented the four fields in January. By the end of Q1 the data said something nobody in the offsite had suggested: their named competitors showed up in 22 percent of deals. Internal build showed up in 41 percent. They had spent two years and a lot of content budget positioning against three vendors who were barely in the room, while the actual alternative was an engineering team saying "we could do this ourselves in a quarter."
The fix was not a new category. It was a rebuild of the top of the deck around build-versus-buy total cost, a two-page teardown of what the internal build actually costs at month 18, and a discovery question that surfaced the build option in the first call instead of month three.
No-decision rate went from 46 percent to 31 percent over two quarters. Win rate moved from 20 to 26 percent. Nobody rewrote the homepage until the following September, by which point they knew exactly what it should say.
That is the whole argument. The positioning work was fine. The missing piece was a way to find out if it was true.
Not sure what your pipeline says about your positioning?
We run a 30-minute audit on your CRM data: no-decision rate, alternative mix, and where deals actually die. You will leave knowing whether you have a positioning problem or an execution one.
Book an audit →FAQ
What is the difference between positioning and messaging?
Positioning is the strategic decision: which segment you serve, which alternatives you beat, and what value your difference creates for that segment. Messaging is how you say it in a specific channel to a specific person. Positioning changes maybe once every 18 months. Messaging gets rewritten constantly. If you are arguing about word choice, you are doing messaging. If you are arguing about who the customer is, you are doing positioning.
How do I know if my positioning is the problem or my sales execution is?
Look at where deals die. High drop-off between first meeting and second means the market-facing positioning is attracting people who were never a fit, which is a positioning problem. High no-decision at late stage means buyers understood you but could not build an internal case, which is also positioning, specifically the value and urgency part. Deals that die because of slow follow-up, missed next steps, or single-threading are execution. The win rate breakdown by stage tells you which one you have in about an hour.
Can a small team do this without hiring a product marketer?
Yes. The CRM instrumentation is a half-day of setup. The 40-deal analysis takes a day if your data is clean and three days if it is not. The 10 buyer interviews take a week. What a good product marketer adds is speed and the ability to write the statement well, but the evidence gathering is operations work that a founder or a RevOps person can do. I would rather see a founder run 10 buyer interviews personally than read a consultant's deck about it.
How many segments should we position for?
At Series A, one. At Series B, two or three at most, each with its own statement and its own primary alternative. The failure mode is positioning for five segments with one message, which produces language so general it describes half your market. If your homepage headline would work for a competitor with a different product, you have over-generalised.
Should the positioning statement be public?
The full statement is an internal document. It names alternatives explicitly, includes the segments you are choosing not to serve, and reads like strategy rather than copy. What goes public is the output: the headline, the way reps open a call, the first slide of the deck. Publishing the internal version tends to invite competitors to argue with it, and it is written for your team anyway.