Skip to content
Get started

Synthetic personas that answer in your buyers' words, built from your own sales calls

Synthetic personas are AI stand-ins for your buyers. We build them from your own recorded sales calls and win/loss notes, so you can test a message, a pricing page or an objection in days and check each answer against real buyers before you ship.

Alpha: we take a few teams at a time. Three taps and your site, then we reply within two business days.

One plan, one scoreboard

  • SEO
  • AEO
  • Digital PR
  • Content
  • Analytics

Your scoreboardOne team

Example
  • Search
  • AI answers
  • Press
One team runs the modules you pick, and every result reports to the same scoreboard.

Clients

Their stories ran in

Outlet coverage from our client reporting. Logos belong to their owners; no endorsement implied.

In short

Synthetic personas are AI stand-ins for your buyers. Green Flag Digital builds them from your own recorded sales calls and win/loss notes, so each one answers in the words real buyers used and every answer traces back to the calls behind it. Use the panel to test messaging, pricing pages and objections in days, then check the answers against real calls before anyone acts.

Why synthetic personas matter now

Aaru, a synthetic research startup, raised a Series A at a 1 billion dollar headline valuation in December 2025 (reference 1). Electric Twin raised about 14 million dollars in February 2026 to build synthetic audiences from real survey data (reference 2). The Drum has asked whether fake buyers can show B2B marketers what real ones want (reference 3).

The launch date is set, the homepage needs new words, and a round of customer interviews takes a quarter, because few buyers are ready to talk. At any one time only 5 percent of your market, your audience's in market, like they're in that window when they're looking to make a change. Joe Robison's rule on messaging: I should have only three big messages and themes and repeat it 1000 different times. I think agencies have to stand out by really having like a strong, clear personality, and drawing people to that clear personality. The same is true of a product, and the hard part is choosing the three. When Google says make good content, that's not super actionable, but what is actionable is doing fundamental marketing strategy.

Your buyers already rehearse with AI. One buyer told us he "used every AI tool and then asked like 90 different questions" before he found us. You can rehearse your pitch the same way.

What we do: Map, Build and Run

GFD is the AI firm for the marketing department: we Map, Build and Run.

Map: collect the calls and sort the buyers

We gather recorded calls, win/loss notes and any survey answers you own. One founder told us, "We got all this data from that survey and we didn't really do much outreach with it." We group buyers by role, stage, trigger and objection, because they want different things. There are some clients that want boring guaranteed success. There's some clients that want to see compounding traffic numbers. There are some clients that want to prove it out as a channel at all.

Build: a panel that talks like your buyers

Each persona answers in its source buyers' words and links to the passages it drew on. It's just super useful to be on the other end and think like they do, and then that can help us tailor stuff. The test bench holds your homepage draft, positioning lines, pricing page and objection list. I like two as a number, because one is best but you want to see a little bit of variance.

Run: re-test each launch and keep it fresh

New calls go in, stale claims come out, and the bench re-runs before each launch. Joe tests new ideas the same way, small and early, where it has to be social, first for validation and then do the content on the back end. It's not the things that you do in one day in a flash of brilliance. It's the things that you return to over and over again, because they're super valuable, and they are really the second external part of your memory.

What's included

What you get

  • A map of buyer types, each tied to its calls
  • One synthetic persona per type, with a trace on every answer
  • A test bench for messages, pricing page and objections
  • A short report: which message wins, which objections recur, which words buyers use

How it works

How a panel gets built

  1. Map. We pull the questions each buyer type asked on calls. List out 5-10 questions, answer them fully.
  2. Build. We run your messages through the panel. Share new formats and test them. Spray and test what works! Test headlines and hooks.
  3. Check. We compare the top answers with the real calls. Always double check this stuff. It's super annoying.
  4. Report. The winning message, the recurring objections, the buyer's own words.

The limits

What a synthetic panel can't tell you

Synthetic answers are a fast first filter. They never replace talking to customers.

A panel only echoes what your buyers have said. It cannot predict a new market or measure how strongly people feel. If you're doing true scientific stuff, you usually want a higher correlation, like a 0.7 and up. But any study around human behavior will tend to be closer to 0.5. So it's kind of a good reminder that we're not gonna find a lot of really high correlation. Treat a verdict as a lead, not a fact.

We check every result against the real calls before anyone acts. When the stakes are high, talk to a buyer. In person is the new luxury. I want to be in society and get a ton of reps in and learn it all, soak it all up, feel all the human emotions to get and feel perspectives.

Not all that works, and I need to communicate this upfront. We do not promise a tested message will convert. We show you the evidence behind every answer.

How we built this on ourselves

We keep a voice-of-customer bank of real buyer lines from our own sales calls, and we write every page on this site from those words.

We read every message sent through our contact form from late 2022 to November 2026: 259 in all, 106 of them from real buyers. "Link building" showed up in 29 of those buyer messages and "digital PR" in 27. "AEO" showed up in none. When we said "digital PR" to one buyer who had never hired a PR firm, he asked, "how does this work, do you pay them?"

The objections recur too. On our own calls, buyers asked: "What if a campaign we do, like, it just doesn't land anything?" "Three months is a while for the first link." "It's always good to have something in writing at least." "It seems like there's no guarantees, right?" "We don't know what to write, right?" "My hesitation for doing it all together is this is the first time I've talked with a PR firm." Your panel surfaces your version of this list.

Joe wrote this in his journal in 2021: I would say 50 to 70 percent of target buyers don't know about my competitors. And that's great. There's a whole wide ocean out there of people to target. So pitch clients like I'm teaching them from scratch.

Before we relaunched our website, we ran synthetic buyer walk-throughs of it to find what confused buyers, checked those flags against our sales calls and rewrote what failed. The speed at which I can research and generate ideas and publish, and then run it through a final editing curation process, I think will actually be pretty good.

How we charge

This is an alpha service. We take a few teams at a time and build each panel with them. Scope and price are set after a call. I need to create stuff that is clear and easy to buy, easy to see the value, quick and easy to scale.

Case study · moveBuddha

Cited as the ranking source when ChatGPT answers “best interstate moving companies”

One stored ChatGPT answer, pulled from Ahrefs Brand Radar on October 21, 2026, says the movers are “ranked highly by moveBuddha”.

One stored ChatGPT answer

best interstate moving companies

ChatGPT · excerpt of the answer

“… ranked highly by moveBuddha …”
Stored answer in Ahrefs Brand Radar · Oct 21, 2026
Proof from client work

Who runs the work.

Results and coverage from our client reporting. Results vary by company, budget and market.

JRJoe Robison

Joe Robison

Founder and principal, Green Flag Digital

Joe sets the strategy on every engagement and makes the final call on what counts as evidence. A principal reviews the findings before any client reads them.

Joe Robison on LinkedIn

What clients say

  • “Green Flag consistently delivered high quality PR pieces that got mentions and links from some of the largest media brands on the internet.”
    Ryan Carrigan, Owner, moveBuddha · via Clutch

Reviews from our Clutch profile. Results vary by company, budget and market.

See where you stand.

Tell us where to look. We come back with where you stand today, who gets named instead of you, and the work we'd take first, then walk you through it in a 20-minute teardown.

Questions buyers ask

What are synthetic personas?

Synthetic personas are AI stand-ins for your buyers. GFD builds ours from your recorded calls and shows the calls behind each answer.

Can a synthetic panel replace customer interviews?

No. It tells you which messages to take into real interviews. We check key answers against real calls and say what the panel cannot know.

What do you need from us?

Recorded calls, win/loss notes, the page you want to test, and a product marketer who reviews each report with us.

What does alpha mean here?

A new service built with our first clients. We take a few teams at a time and price each scope after a call.

How do we apply?

Use Apply for alpha on this page. It opens a short form, Joe reads every application, and we reply within two business days.

What do you most want from search right now?

Three taps, then your site. About a minute. Public data only, no logins.

·

Sources and method

Methodology

This page describes how Green Flag Digital builds synthetic buyer panels as of 2026-11-15. It draws on Joe Robison's journal entries, voice memos and recorded walkthroughs from 2021 to 2025, GFD's anonymised voice-of-customer bank from its own sales calls, and a count of the messages sent through GFD's contact form from November 2022 to November 2026. No buyer or prospect is named. Funding figures come from the press reports below.

References

  1. TechCrunch, AI synthetic research startup Aaru raised a Series A at a 1B headline valuation, December 5, 2025. techcrunch.com
  2. Tech.eu, Electric Twin expands AI audience platform with 14M round, February 12, 2026. tech.eu
  3. The Drum, Can fake buyers reveal what real B2B customers really want? thedrum.com
  4. Green Flag Digital voice-of-customer bank: anonymised buyer lines from GFD sales calls, and a count of 259 contact-form messages, November 2022 to November 2026.
  5. Green Flag Digital journal entries, voice memos and working sessions, 2021 to 2025.

History of this topic

  • 2021: Joe Robison notes that most target buyers do not know the firms GFD competes with, and decides to pitch as if teaching from scratch (reference 5).
  • 2025: Aaru raises a Series A at a 1 billion dollar headline valuation for synthetic research (reference 1).
  • 2026: Electric Twin raises about 14 million dollars to build synthetic audiences from real survey data (reference 2).
  • 2026: GFD counts 106 buyer messages from its contact form and finds that none use the word "AEO" (reference 4).