One major culture shock I experienced when I decided to go headfirst into this founder journey was that when it came to B2B marketing, up is down and down is up. Nowhere is this more true than in how we talk about pricing and packaging with prospective customers.
I remember during an early customer discovery call, a prospect (who also happened to be a founder selling a B2B SaaS product), asked me what I would be “willing to take” as a price for my services.
Wait, what.
Not: what was he willing to pay.
Instead: what was I willing to take.
The next day, I was told to remove all pricing from my landing page entirely.
Um…excuse me? What was the point of a putting any options up there at all?
For context, I’m coming in with 15+ years of experience in consumer go-to-market research. A world that obsesses over price elasticity curves, needs-based market segmentation, precise positioning and rigorous message testing. That’s when I realized that B2B sales and marketing runs on deliberate ambiguity. Hidden pricing, gated content, jargon-heavy positioning.
Who doesn’t love a pricing page…with no prices?!
Now there were really good reasons for this: every deal is different, sales wants the chance to build value before showing a number…and of course, nobody wants competitors seeing their prices.
This approach made sense…back when you controlled the conversation with every buyer.
But now, AI agents are part of the conversation, and they are the most confused buyer in the room. Agents are out there actively trying to self-serve your product — just like the 87% of B2B buyers who want to — and they, too are hitting the same wall of opacity.
THE NUMBER
1 in 3 developer tools brands is inaccurately described by AI. Not missing — just plain wrong.
And it's not just dev tools. The most opaque B2B categories — supply chain (34.5% inaccuracy), insurance (29.6%), HR tech (22.1%) — are the ones AI gets wrong most often. Across all B2B categories, brands are misrepresented at 1.6x the rate of consumer brands.
The pattern: the less public information available about what you do, what you sell, and what it costs, the more AI fills in the blanks. And it fills them in wrong.
This isn't just showing up in our data. G2 recently started publishing estimated pricing on 9,000 B2B software profiles — because vendors won't do it themselves. G2 also happens to be one of the top sources AI systems learn from when forming opinions about B2B brands.
Everyone is talking about AEO — AI engine optimization. But you can't optimize what isn't there. Most B2B brands haven't done Step 1: existing in the information ecosystem with enough clarity for AI to have anything to work with.
THE SIGNAL
The more obvious your product is, the better AI describes it. The more abstract, the more invisible you become.
Here's the spectrum:
Food & beverage has a 97.7% accuracy rate. When what you sell is self-evident, AI gets it right.
SaaS has the worst fair representation rate of any major category: 59.6%. If you're reading this newsletter, there's a good chance you're in that number. 4 in 10 SaaS brands are either invisible to AI, inaccurately described, or both.
Professional services has a 21.5% phantom rate — the highest of any major B2B category. 1 in 5 professional services firms is essentially invisible to AI. When your value lives in relationships and expertise rather than a product page, AI has nothing to grab onto.
HERE’S WHAT TO DO INSTEAD
Stop optimizing for AI. Work on being legible first and feeding it structured facts about you, your products, your customers and yes — your pricing too. AI is already evaluating you from these perspectives whether you ask it to or not.
Here are 3 prompt templates you can use right now to have AI critique your pricing and packaging.
First, paste your pricing page into Claude/ChatGPT. Then:
Ask: 'You're a CFO evaluating tools in [your category] for a team of 50. Based on this pricing page alone, would you shortlist this product? What's unclear?'"
"Ask: 'You're a first-time buyer who's never heard of [your brand]. Based on this page, what do I do, who is it for, and what does it cost?' — whatever it can't answer is what AI can't answer about you either."
"Ask: 'Compare this pricing page to [competitor]. Which one is easier to evaluate without talking to sales?'"
The point isn't to publish your pricing tomorrow. The point is to see what AI sees — because that's what your next buyer is seeing too.
If AI can't understand what you do from what's publicly available, that's not an AI problem — that's a you problem.
See you next week!
—Apurva
PS: I asked you, dear readers, about your AI Brain Fry moment. My favorite response this week is the founder who hired an agent to manage their agents.
I call it Master of Bots - this agent tells my other agents what to do so I only have to talk to one agent instead of 5. Unfortunately I still don’t know exactly what any of them are doing but I guess if the code works it must be working because I don’t really have time in my day to babysit interns who work 24/7.
ABOUT THIS NEWSLETTER
Hi! 👋 My name is Apurva. I’m a brand strategist, data nerd, recovering big-tech marketer, accidental founder, reluctant React debugger, mom of two, and the kind of person who blows past her screen time limits doomscrolling parenting reels while stress-searching LinkedIn for anything that feels human. I started this newsletter because I think marketers are drowning and nobody's naming the why.
You're getting this because we've crossed paths recently — a project, a conversation, a demo, a conference, or you signed up to learn more about Optimly. If this isn't for you, no hard feelings — unsubscribe anytime. But if you're a marketer trying to make sense of AI right now (without going crazy), stick around.