We're pivoting. Do we announce the new story or grow into it?

What did the pivot actually change, and what didn't?

A CEO I talked with this week runs a $21M ARR software company selling to community banks. Eighteen months ago they sold compliance and exam-prep tooling. Today two thirds of their revenue is fraud-pattern monitoring, because a handful of their best accounts kept using the compliance data for something the company never built it for, and the company followed them there.

Real pivot. The board approved it. Product shipped it. The comp plan reflects it. Everybody inside the building knows exactly what business they're in.

Then he shared his screen. The homepage said compliance. The About page said compliance. Both case studies said compliance. The careers page was hiring a Compliance Solutions Consultant. The G2 category was compliance. And when he asked ChatGPT what his company does, live, in front of me, it told him: compliance software for community banks.

He laughed, and then he stopped laughing, because his head of sales had told him three weeks earlier that new prospects keep showing up with the wrong problem already in their heads.

The question he was actually asking is the one sitting on every pivot's board agenda: do we announce this properly, or do we let the work speak and grow into it? Both answers fail, and they fail for the same reason.

What is the Trailing Story?

The Trailing Story is the version of your company the market keeps repeating after you've already changed. Nobody's lying. It's the residue of every true thing you published back when it was true: old pages, old decks, old case studies, old job posts, old review-site categories, old answers now sitting inside AI models. Your direction changed. Your evidence didn't.

That's why a pivot can feel like it isn't landing when the pivot itself is fine. A company can change what it sells in a quarter and change its published record in about... never, unless somebody makes that a project with a name and a deadline.

The announcement instinct assumes the problem is news. It isn't. Your buyers didn't miss a memo. They read what's in front of them, and what's in front of them is the old company, fully documented, still ranking, still getting cited.

The grow-into-it instinct assumes the problem is time. Time makes it worse. Every month you leave the record alone, the old story picks up more repeats, more links, and more corroboration, while the distance between it and what you actually sell keeps widening.

Why is a pivot harder to communicate in 2026 than it was five years ago?

Five years ago your old story lived in places that decayed. A page slid down the rankings. A deck went in a drawer. A customer forgot. The market had a short memory and you could out-publish it.

Now your old story lives in systems that retrieve. When a buyer asks an AI engine what your company does, it doesn't reach for the freshest thing you published. It reaches for whatever it has the most repeated, most corroborated evidence about. About your company, that's still the version you spent five years documenting.

Express Writers ran a small, honest test on exactly this in 2026. They published a page carrying current company facts, waited six weeks, then asked again. All three web-enabled AI answers had the new facts. None of the five answers from models with browsing switched off did. Same six weeks, same page, two different companies depending on which mode the buyer happened to be in.

That's a tiny sample and I'm not going to inflate it into a law. But the shape matches what we see every week. An engine reading the live web can catch your pivot in weeks. An engine answering from what it already knows hands your buyer the company you used to be, in the same chat window, with exactly the same confidence. If you want the mechanics of that gap, we went deep on it in Why does AI describe our B2B company inaccurately, and how do we fix what ChatGPT says about us?.

There's a second shift underneath. AI took the cost of producing the announcement to zero. Anybody can regenerate a homepage, a launch post, and forty pages of new-direction content in an afternoon. Volume of fresh material proves nothing now, and it won't move a story that has years of corroboration behind it. What's scarce is a company that has decided precisely what it now is, clearly enough that every surface can say the same thing without another meeting.

How do you find out how far behind your story is?

Three tests. All three take under an hour, and you can run them before your next leadership meeting.

1. 1The Surface Count. Open a blank doc and list every public surface that describes what your company does. Homepage, product pages, About, pricing, case studies, the blog archive, your LinkedIn company page, your leadership team's personal profiles, job posts, review-site categories, partner directories, conference bios, that PDF someone put behind a Drive link in 2023. Mark the ones still telling the old story. Most CEOs guess five and land near forty. The number is the finding, because it tells you whether this is a copy edit or a project.
2. 2The Renewal Sentence. Message three current customers separately and ask them to write one sentence describing what your company does now. No coaching, no options, no call. If they hand back the old company, your pivot hasn't reached the people who renew, which means your expansion revenue is being quoted against the problem you used to solve. That's the expensive version, and it's the one founders find last.
3. 3The Engine Lag. Ask ChatGPT and Claude what your company does. Do it twice, once with web search on and once with it off. The gap between those two answers is your Trailing Story, measured. The off-web answer is what a buyer gets during fast, casual research, which is most of the research that happens.

What do we see across companies that just changed direction?

Across the B2B companies we score for narrative clarity, the ones who changed direction in the last two years share a shape, and it isn't the one they expect.

The new story almost always already exists somewhere in the building. Usually in the mouth of whoever owns delivery or support, not the founder, because they're the one who hears what customers say when they call. The founder describes the strategy. The delivery lead describes the purchase.

Second pattern: the pivot gets published as a feature announcement. New capability, new module, new item in the product nav. The company changed which problem it solves and told the market it shipped a thing. Buyers read a feature and file you in the box you were already in.

Third, and this is the one that costs the most: nobody owns the back catalog. Every company we work with has a marketing lead who owns new content and literally nobody who owns the two hundred existing pages still describing the old company. During a pivot those pages aren't archive material. They're the majority of what a buyer and an engine will ever read about you. It's the same failure that makes We've outgrown our messaging. Why does it still describe our old company? such a common conversation, except a pivot compresses years of drift into a single quarter.

Worth saying plainly: the pivot itself is usually the right call. The Startup Genome Report on premature scaling, a 2012 study of 3,200 startups by Max Marmer, Bjoern Lasse Herrmann, Ron Berman, and Ertan Dogrultan, found that companies pivoting once or twice raise 2.5 times more money and see 3.6 times better user growth than companies that pivot more than twice or not at all. That research studied the decision. Nobody studied the part where you have to go tell everyone, which is why that part keeps landing on whoever has time.

What does closing the gap actually look like?

A $34M ARR telematics company, regional trucking fleets, hardware plus a dashboard. Over two years their best revenue moved to selling driver-risk data to insurance carriers. The fleets were still the customer of record. The money was in the data.

Their site said fleet tracking. Their pipeline filled with fleet managers comparing them against $12-a-truck GPS tools. Their sales cycle had gone from 54 days to 91, not because deals got harder, but because every call opened by explaining they weren't the thing the buyer had shortlisted them for.

The Surface Count came back at 47 public surfaces, 41 of them describing the old company. The Renewal Sentence came back from three customers and all three described GPS tracking, including one that was, at that moment, paying for the risk product.

The buried truth surfaced in week two, from their VP of Customer Success, not the CEO. Fleets were never buying tracking. They were buying a lower insurance renewal, and tracking was how you earned it. The carriers were buying the same thing from the other side of the table. Nobody had ever written that sentence down.

“We've always sold safer fleets. We used to sell it as hardware. Now we sell it as the data that proves it.”

... The Bridge Sentence their leadership team landed on, week three

One sentence, and it did the thing an announcement can't. It let an existing customer understand they hadn't been abandoned, and it gave a new buyer somewhere to file them other than GPS. That's what a Bridge Sentence is for: the old thing, the new thing, and the promise that didn't change.

Then they flipped the surfaces together instead of one at a time. Nine weeks, 41 pages, including the job posts and the review categories nobody wanted to own. Sales cycle back to 58 days. The thing the CEO noticed first was smaller than the number: prospects stopped asking what they do.

What should you do in the next 90 days?

The pivot decision was the hard part and you already made it. What's left is the part nobody schedules: deciding precisely what's true now, then changing the record so it says so.

That's the job a Magnetic Messaging Framework does. It's the documented version of your narrative identity, built around four anchors, and after a pivot each one has a specific job. Category design decides which box you're in now, on purpose, instead of inheriting the old company's. Villain framing names what you're against now, which changed the moment your problem changed. The old-way / new-way contrast is the argument your pivot is literally an example of. And the promised-land outcome is what your customer gets, which is usually the one thing that didn't change and the reason your existing accounts can come with you.

Here's why the document matters more after a pivot than at any other moment. Forty surfaces have to be rewritten by somebody, and most of them will be rewritten with AI. Without one source of truth, forty pages get rewritten forty slightly different ways, every one of them defensible on its own, and you've manufactured a brand new Trailing Story that shows up in eighteen months. Two companies that merge hit the identical wall, which is why How do we unify our brand messaging after an acquisition? reads like the same problem wearing a different hat.

1. 1Run the Renewal Sentence this week. Three customers, in writing, no coaching. Whatever comes back is your real starting position, not the one in your board deck.
2. 2Do the Surface Count and give it an owner with a deadline. Not the marketing lead who owns new content. Someone who owns the record.
3. 3Write the Bridge Sentence before you rewrite a single page. Old thing, new thing, promise that didn't change. If your leadership team can't agree on it in one meeting, the pivot isn't decided yet. You're still in the middle of making it.

PitchKitchen builds Magnetic Messaging Frameworks for founder-led B2B companies in the $5M-$75M range. Founded by Greg Rosner, founder of PitchKitchen and author of Story Craft for Disruptors, PitchKitchen fixes broken marketing messages and underperforming websites for CEOs whose sales are stalling because their message isn't doing the work. If you want a fast read on whether your homepage still describes the company you left, the Brand Signal Score, PitchKitchen's free homepage messaging diagnostic at pitchkitchen.com/brand-signal-score, will score it in a couple of minutes.