Straight answer: AI tools are useful for researching what other people in your field charge, but they will consistently price you lower than you’re worth, because they’re built to give you the safe, average, scraped-from-the-internet number. Use AI to gather the data, then ignore its final recommendation and set the price yourself based on positioning, not averages.
Why I even bothered asking AI to price me
I’ve been consulting on AI and marketing for over a decade. I know what I charge. But I kept seeing small business owners on LinkedIn saying they’d asked ChatGPT “what should I charge for my services” and treated the answer like it came down from a tribunal. So I ran the experiment. I gave four different AI tools the same brief: a 25-year marketing veteran, Forbes-listed, UK-based, working with US and Israeli clients, offering full-day AI and marketing strategy consulting. I asked each one, straight, what my day rate should be.
What each tool told me
Here’s the exact spread, and it’s worth sitting with because the range tells you more than any single number does.
- ChatGPT (GPT-4o): £600 to £900 a day
- Claude: £550 to £850 a day, “with premium positioning up to £1,200”
- Gemini: £450 to £750 a day for “experienced marketing consultants”
- Perplexity, which pulled live sources: £500 to £900, citing a mix of freelancer platform averages and a UK consultancy salary survey
My actual day rate sits well above the top of every single one of those ranges. Not because I’m arrogant about my worth, but because none of those tools knew what I bring to a room, they only knew what a “marketing consultant” typically charges according to whatever data they were trained on or could scrape in the moment.
The uncomfortable bit nobody selling you AI wants to say out loud
Every one of those tools anchored to the middle of a distribution. That’s not a bug, it’s how they’re built. Large language models predict the statistically likely answer, and the statistically likely day rate for “marketing consultant” is a number pulled from job boards, freelancer marketplaces and salary surveys, most of which skew toward generalists, not specialists with a Forbes badge and fifteen years of client results behind them. If you read the work of Richard Thaler on anchoring, this is exactly the trap he described decades before AI existed: give someone a number first, even an irrelevant one, and it quietly drags their final decision toward it. AI doesn’t remove that bias. It automates it and hands it to you wearing a lab coat.
The awkward part is that a lot of small business owners want AI to hand them a low number. It’s easier to accept “the algorithm says £650” than to sit with the fear of naming £3,000 yourself and having a prospect say no. Outsourcing the number to a chatbot feels like research. Mostly it’s avoidance with better branding.
Where AI is useful in this process
I’m not saying bin the tools. I used all four again, differently, and got something worth keeping.
- Ask AI to pull comparable rates for very narrow, specific niches, not broad job titles. “AI implementation consultant for e-commerce brands turning over £2m to £10m” gives a far more honest range than “marketing consultant.”
- Ask it to list what typically justifies a premium in your field: certifications, media mentions, case studies, speaking history, book deals. Use that as a checklist against your own bio, not as a pricing formula.
- Ask it to draft three versions of your pricing page copy at three different price points, then notice which one it writes with the most confident language. That’s often a tell for which price point feels “normal” to the model, which is useful information precisely because you want to price above normal.
How I set my rate, and why it has nothing to do with an average
My rate is built on scarcity, proof and positioning, three things no AI tool can see from a prompt. Positioning is the whole game here, and it’s the same lesson you get from studying how big brands price themselves. Look at how Disney built a brand people will pay a premium for without ever competing on being the cheapest theme park, or how Coco Chanel priced scarcity into the DNA of a brand from day one. Nobody asks Chanel to justify its price against the average handbag. That’s the point.
The same logic applies to a one-person consultancy. My day rate reflects that a client isn’t buying eight hours of my time, they’re buying the shortcut past five years of trial and error, and results I can point to. AI can’t price that in because it has no visibility into your specific track record, your specific case studies, or the specific gap you fill that generalists in your field don’t.
A five-step process that works
If you want to use AI in your pricing decision without letting it lowball you, here’s the sequence I’d now recommend to any client asking me the same question I asked those four tools.
- Step 1: Ask AI for rate ranges in the narrowest version of your niche you can describe in one sentence, not the broad category.
- Step 2: Ask a second, separate question: “what makes someone in this field charge at the top of that range rather than the bottom.” Write that list down.
- Step 3: Score yourself honestly against that list. Three or more strong yeses means you should be pricing at the top of whatever range AI gave you, not the middle.
- Step 4: Add a minimum of 20 to 30 percent above that top figure to account for the fact that AI ranges are always slightly behind current market reality, because training data and scraped sources lag real pricing conversations by months, sometimes years.
- Step 5: Test the number on three real prospects before you commit it to your website. If nobody blinks, you’ve priced too low. A little wince is the sign you’ve got it roughly right.
I’ve watched clients go through this and land 40 to 60 percent above the number AI first gave them, purely by refusing to accept the average as the ceiling.
Put the pricing question on your website, not just in your prompts
One thing worth doing once you’ve settled on a rate: stop making prospects email you to find out roughly what you cost. A well-built pricing or ROI calculator on your site does two jobs at once, it qualifies leads before they waste your time, and it lets visitors self-select into the price bracket that suits them, rather than anchoring on a single number they saw on a competitor’s site. I’ve written before about how interactive calculators quietly convert more visitors than a static price list ever will, and pricing is one of the best places to use one.
The businesses that get this right tend to be the ones that treat pricing as part of brand, not part of admin. Look at how differently GoDaddy positions and prices its tiers compared with how Headspace built pricing into a calm, premium experience rather than a discount race. Neither company asked a chatbot what the “average” competitor charges and copied it. Your one-person consultancy shouldn’t either.
If you’re weighing up whether to hire help with any of this
A fair number of small business owners land on this exact question, how do I price myself or my AI-related services, right around the same time they’re deciding whether to bring in outside help rather than keep guessing alone. If that’s you, it’s worth reading through what an AI consultant typically costs before you set your own rates, because it’ll show you exactly the kind of range distortion I’ve described here, and how much of that spread comes down to specialism, proof and positioning rather than hours in the day.
Free resource: The Price Conversation Swipe File.
Frequently asked questions
Can AI tell me the right price for my services?
It can give you a rough, averaged starting range based on job titles and scraped rate data, but it can’t account for your specific proof, positioning or scarcity, which is where the real premium sits. Treat its answer as a floor, not a final figure.
Why did every AI tool give me a lower number than I expected?
Because language models predict the statistically common answer for your job title, and most public rate data skews toward generalists rather than specialists. The more narrowly and specifically you describe your niche and proof, the higher and more honest the range becomes.
Should I just pick the highest number an AI tool gives me?
No. Use it as one data point among several, add 20 to 30 percent to account for lag in the data, and test the resulting number on real prospects before it goes on your website or proposal.
Is there a better way to use AI for pricing research than just asking “what should I charge”?
Yes: ask two separate questions, one for narrow-niche rate ranges and one for what typically justifies premium pricing in that field. Score yourself against the second list before you accept any number from the first.
Related reading: I Asked AI to Audit My Own Blog. It Told Me to Delete Half of It. and I Write Client Proposals With AI in 20 Minutes. Here’s the Bit Nobody Tells You.
