Portfolio Hypnosis: why enterprise software companies keep hiring the wrong messaging agency

Why do enterprise software companies keep hiring the wrong messaging agency?

Three enterprise software companies walked me through their agency shortlists this quarter. Different categories, different revenue bands, one identical shape: the finalist was the firm with the best-looking case studies. Eleven months and a serious spend later, two of the three still couldn't get their message through the second call with a buying committee.

You choose a messaging agency for enterprise software by testing it against the shape of your deal, not the beauty of its work. Enterprise deals get decided by a group, in rooms you're not invited to, by people who'll never read your homepage. The right firm builds a message that survives being repeated by someone else. A portfolio can't tell you whether a firm can do that. Only a live test can.

Here's what makes this specific to enterprise software and not just "hiring an agency." When someone buys a $200-a-seat tool, the message has one job: convince the person reading it. When someone buys your platform, the message has a second job that matters more. It has to be repeatable by a champion, in their words, to a committee that includes at least one person actively hunting for a reason to say no.

Most of the messaging work that looks gorgeous in a portfolio was built for job one. You're buying for job two.

What is Portfolio Hypnosis?

Portfolio Hypnosis is the failure mode where a buying decision gets made on the quality of the work shown instead of the fit of the work needed. It's the reason smart CEOs hire firms that have never once handled a procurement review, a security questionnaire, and an eleven-month cycle at the same time.

It works because portfolios are designed to be admired, and admiration feels like evidence. You look at a beautiful brand system built for a design tool with 40,000 self-serve users, and your brain quietly converts "this is good" into "this will work for us." Those are unrelated claims.

This is a different question from which KIND of firm to hire. That decision comes first, and we worked it through in Brand strategy firm vs messaging consultancy vs positioning consultant vs fractional CMO: what does each deliver?. Portfolio Hypnosis hits later, once you've picked a category of firm and you're choosing between three of them. It's also narrower than the pattern where the agency was never going to work at all, which has its own diagnosis in Why does every marketing agency we hire fail?. Portfolio Hypnosis is more expensive than either, because the firm you hired is genuinely good. It's just good at a different sport.

Why is this worse now than it was two years ago?

Because polish is free now, and it wasn't before.

AI brought the cost of producing a beautiful deliverable to roughly zero. Every agency's case-study page now looks like the top of the market. The visual craft, the tight copy, the confident brand system, all of it is available to a two-person shop and a fifty-person agency at the same quality bar. Portfolio quality stopped being a signal at exactly the moment buyers started leaning on it hardest.

Meanwhile the thing you actually need got harder. Gartner's B2B buying research found that customers spend only about 17% of the total purchase journey meeting with potential suppliers, and that slice splits again across every vendor on the shortlist. Your message spends most of its working life travelling without you. It gets forwarded, paraphrased, half-remembered, and now summarized by an AI assistant before a human ever reads a word you wrote.

A firm that's never built for that condition hands you language that works beautifully when you're in the room to defend it. That's the trap. It tests well in the presentation and dies in the wild.

How do you test an agency before you sign?

Stop reviewing work they've already done. Make them do work in front of you. Three tests, all of which fit inside a single 60-minute finalist call, and none of which are easy to fake.

1. 1The committee test. Hand them a real deal you lost in the last six months. Give them your product, the committee makeup, and the objection that killed it. Ask them to write the three sentences your champion would have repeated to the CFO. Judge one thing only: would a finance person say those sentences out loud without feeling stupid? Most firms hand back something elegant and unrepeatable.
2. 2The subtraction test. Ask what you should stop claiming. Not what to add, what to give up. A firm that only adds is decorating. Real positioning costs you something: a market you'll stop chasing, a feature you'll stop leading with, a buyer you'll stop pretending to serve. If nobody on their side will take something away from you in a sales call, they won't do it after the contract either.
3. 3The unfamiliar-category test. Give them twenty minutes on the genuinely technical part of your product, then ask them to explain it back to a non-technical buyer. You're not testing whether they understood it. You're testing whether they can hold a complex thing and make it simple without making it wrong. Every enterprise software company has been burned by a firm that simplified until the claim stopped being true.

Test one is the one that predicts the most, because it's the actual job. The champion-enablement problem underneath it gets its own treatment in How do you equip a champion to sell you to the buying committee?. If you want the broader pre-contract question set, separate from these live tests, it's in What questions should you ask a messaging or positioning consultancy before you sign?.

What do we see across 200+ B2B companies?

The firms that work for enterprise software share a profile that has almost nothing to do with how their work looks. Across 200-plus B2B companies, the same seven filters keep separating the engagements that hold from the ones that quietly get shelved.

That last filter is the newest, and it's the one most shortlists ignore completely. Buyers now research through AI assistants that summarize you before they ever meet you, which changes what "good messaging" even means. That shift is worked through in What changes about B2B positioning when AI is doing the buyer research?.

Gartner also puts the typical B2B buying group at six to ten decision makers for a complex solution. Read that against filter two. A firm that never asks who else is in the room is building a message for one person in a purchase that involves nine.

How does this play out in practice?

Here's a composite, drawn from several engagements rather than one client, because the pattern repeats almost exactly.

A cybersecurity software company in the $30M range hired a well-known brand firm. Beautiful work. New identity, a sharp new line, a homepage that genuinely looked like the category leader. Their VP Sales, the champion character we see in nearly every one of these, flagged the problem within eight weeks: reps couldn't use any of it on a call. The new line landed with CISOs and evaporated the moment a CFO joined the second meeting.

Nothing was wrong with the work. It got built for the person who visits the website. Their deal was decided by a group of seven, and the two people who controlled the budget never visited the website at all.

The rebuild took a quarter. Same company, same product, same category claim. The difference was that the message got built backwards from the CFO's objection instead of forward from the homepage. Cycle time moved from the high 200s of days into the 170s over the following two quarters, and the champion stopped needing to be on every call.

They didn't need a better agency. They needed one that asked, in the first meeting, who else is in the room.

What does this mean for you?

If a message has to survive being repeated by someone who doesn't work for you, the thing you're buying isn't creative work. It's a documented narrative your whole company can run on. That's what a Magnetic Messaging Framework (MMF) is for: the strategic narrative system built around category design, villain framing, an old-way / new-way contrast, and a promised-land outcome, written down so your reps, your champion, your website, and the AI engines briefing your buyers all say the same specific thing.

That's the work PitchKitchen does for founder-led B2B companies in the $5M-$75M range, and it's why the 90-Day Magnetic Messaging Sprint ends in a documented framework rather than a folder of deliverables. A campaign expires. A framework is what your champion still has in their head nine months later, in the room you're not in. It's the same argument I made in Story Craft for Disruptors, pointed at a purchase most CEOs make exactly once. What that costs, by engagement type, is broken down in How much does strategic messaging consulting cost? A 2026 breakdown by engagement type.

1. 1Run the committee test on your current shortlist. One lost deal, three sentences a CFO would repeat. Sixty minutes tells you more than six portfolios.
2. 2Score your own homepage before you brief anyone. The Brand Signal Score, PitchKitchen's free homepage messaging diagnostic at pitchkitchen.com/brand-signal-score, grades narrative clarity, trust signal, AI signal, and conversion signal. Walking into an agency conversation with a diagnosis changes who's leading it.
3. 3Write down who else is in the room. Every title, every objection, every person who can kill the deal. If your agency's discovery doesn't produce that list in week one, you already know how this ends.

The firms on your shortlist can all make you sound better. Almost none of them can make you the obvious answer to a room of eight people who've never met you. That's the only question worth asking in a finalist call. This is just truth.