Indie HQ
Indie Blueprint
Expert partnerArima
The Blueprint Series · Synthetic Data

Synthetic data, for real.

The future is synthetic. Here's a plain-English guide for independent agencies: what it is, which kind to trust with which decision, and how to put it to work.

Indie Blueprint · 2026
Built with expertise from
Arima

Arima's team supplied the framework, the validation research and the checklist at the heart of this Blueprint.

Read this firstThis Blueprint is for educational and illustrative purposes only. It is not legal, financial, technical or purchasing advice, and it is no substitute for your own due diligence. Arima, an IAN partner, sells synthetic data products and contributed the three-kind framework, the checklist and a summary of its validation study with a third-party research firm, which is an abstract of a longer paper Arima plans to release. This Blueprint is educational, and Indie Agency News does not recommend any vendor. Views belong to the people and companies named, and inclusion of any company implies no endorsement by Indie Agency News.

Section 01 · The Opener

Meet your market before it meets you.

Imagine putting a research question to an entire market without fielding a survey. That is the promise of synthetic data.

Synthetic data is changing how research gets done. Think of a synthetic population as a population simulator, or a marketing sandbox: a place to test an idea on a market before you spend money on the real one.

Arima's team uses an analogy that keeps both the promise and the limits in view.

"Think of it like a flight simulator. The early ones had trees that looked like broccoli. We're well past that, but it isn't hyper-realistic yet. A simulator lets you train faster and more often, but you still have to fly a real plane to get your license."

Ray Kong · CMO, Arima

That's the frame for this Blueprint. Use the simulator to practice and move faster, then check it against the real thing before the decisions that count.

Arima's Chief Innovation Officer, Sabaa Quao, takes the analogy a step further. Fighter pilots go back to the simulator a couple of times a week to rehearse the emergencies that are too dangerous or too expensive to stage in a real aircraft, like an engine failure. A marketing sandbox works the same way. You can run the what-ifs, including three things going wrong at once, before any of them happen for real.

The respondents are realistic stand-ins for people and markets, generated from real data sources using statistics. They come with demographics, locations and habits. They answer fast, and none of them will ever file a privacy complaint. They don't copy any living person. They reproduce the patterns.

The appeal is easy to see. Surveys are getting harder to run. Privacy rules are tighter, hard-to-reach audiences answer less often, bots pollute the results and the people who do respond are tired of being asked. Traditional research is also slow. By the time the findings are gathered, analyzed and acted on, the market may have moved.

Know which kind you're using

The words "synthetic data" are covering three very different things: personas, panels and populations. A persona works like a focus group, a panel like a research panel and a population like a census. They share a name, and they are built for very different decisions. Ask a persona to size a market and you'll get a number it was never built to give.

Our expert partner on this Blueprint is Arima, which builds synthetic populations for a living. The spine of the report is a test Arima ran with an established third-party research firm, putting a synthetic population up against a human panel and recording where it matched and where it missed.

The misses are the most useful part. They tell you where to trust synthetic data, where to check it and how to ask it better questions.

Why this matters for an indie

Used well, synthetic data puts research behind your pitches, makes you the adviser when clients get pitched it and can open a new service. Used without care, it can put a number in front of a client that the data can't back up. This guide is about getting it right.

3products sold under one name
20questions put to people and to the simulator
5places the answers split
8questions to ask of any synthetic data

The expertise behind this Blueprint comes from Arima.

Arima is one platform across the full marketing workflow: research, strategy, optimization, measurement and attribution, with no cookies, no PII and no data-science team required. Its recognitions include Digiday's Best Marketing Analytics Platform and Gartner's Coolest Innovation in Synthetic Data. Exclusively for IAN members, Arima partners with agencies on pitches to help them win business. Its team supplied the three-kind framework, the validation research and the eight-question checklist at the core of this guide.

Talk to Arima →
If you read one page

Five things you need to know.

  1. "Synthetic data" is three different things.Personas, panels and populations all go by the same two words. Before you use any of it, know which one you have.
  2. Match the tool to the money.Personas and panels are for exploring ideas and getting a direction. The more that rides on the answer, the closer to a full population you want to be.
  3. It has been tested against real people, and it mostly held up.Arima and a third-party research firm put the same 20 questions to a human panel and a synthetic population. The firm's own statistics team signed off on most of the comparison.
  4. The simulation drifts in five predictable places.The future, slicing after the fact, fine-grained scales, "don't know" and any language that isn't English. Know them and you can write around them.
  5. How you ask matters as much as what you use.All five weak spots are about how a question is asked. Write questions the simulator handles well, and check the rest against something real.
Section 02 · The Test

20 questions. Two crowds. One of them simulated.

The first thing anyone asks about synthetic data is the right thing to ask: how do you know it's accurate?

Arima's team hears it more than any other question, and the honest answer starts with an awkward admission. No complete outside dataset exists to check a synthetic population against. Nobody is holding the answer key.

So Arima runs two checks. The first is housekeeping. The synthetic population is continuously rebalanced against the source data it was built from. If the source says one share of men wear glasses and a different share of women do, the synthetic version has to land on those same shares every time.

The second is a stress test, and it's the heart of this section. Arima went to an established third-party research firm and set up a side-by-side.

The humans
A human research panel

A large, representative online panel of real people, surveyed the usual way.

20
identical
questions
The simulator
A synthetic population

Built by Arima and asked the same questionnaire through an AI polling tool.

Three ways to keep score

The two sets of answers were compared on three levels. The formal names are in the small type. The plain versions are worth memorizing.

Same headline?Marginal equivalence
Do the top-line results from both methods line up?
Same story underneath?Segment equivalence
Do the subgroups show the same patterns and behaviors?
Same call?Decision equivalence
Would either set of data lead you to the same strategic recommendation?

The third one is the keeper. You're rarely paying for a number. You're paying for a decision, and two datasets can sit a few points apart and still send you to the same place.

The verdict, from the human side of the table

The company that built the synthetic population is not the right judge of the result. The read that matters came from the statistics team at the firm running the real panel, and its conclusion was that the two lined up in most instances.

"Most instances" is the honest part of that verdict. A claim of a perfect match would deserve less of your trust. Section 04 covers the rest: the five conditions where the two crowds stopped agreeing.

What we don't have yet

The scores. This section draws on an abstract of a longer paper Arima plans to release. The question-by-question results aren't in the abstract, and the list of 20 questions wasn't included with it. When the full study is published, both belong on this page.

Why this test is worth copying

Strip away the vocabulary and the method is simple enough for any agency to borrow. Take questions you can put to real people. Put the same questions to the synthetic version. Compare the headline, the subgroups and the decision. Then write down where they split.

You don't need a giant research panel to try a small version. If you already hold survey results for a client, you have a benchmark. Run the same questions through the synthetic tool and see how close it lands on a subject you know cold.

What you can use now

  • Use all three checks. Same headline, same story underneath, same call. A comparison that stops at the first has done a third of the work.
  • Treat "validated" as the opening of a conversation. The follow-up is always the same: where didn't it match?
  • Ask who did the checking. An outside research firm's statistician carries more weight than a self-assessment.
Section 03 · The Field Guide

Three products hiding behind one word.

Before you use synthetic data, know which of three kinds you're working with.

Arima's team explains each one by its cousin in old-fashioned research. A persona works like a focus group, a panel like a research panel and a population like a census. They are not interchangeable, and the trouble starts when one is asked to do another's job.

Like a focus groupPersona
Good for

Exploring what customers want, how they behave and what they prefer. Ideas, messaging and early concepts.

Not for

Forecasting or sizing a market. It isn't statistically representative.

Brilliant in a brainstorm. Not built for a spreadsheet.
Like a research panelPanel
Good for

Segment-level analysis and directional testing. Comparing audiences and choosing targets, with bigger samples than personas.

Watch for

Very small or niche audiences. The makeup of the panel also has to be managed over time.

Points you in a direction. Doesn't measure the distance.
Like a censusPopulation
Good for

Market sizing, predictive modeling, location intelligence, business cases and simulation. Broad, granular by geography and privacy-safe.

Watch for

It takes sophisticated modeling across many variables. Niche behaviors are only as good as the source data.

Built for the decisions with a budget attached.

Start with the decision, then pick the data

The sorting rule is the size of the bet. If you're exploring an idea, testing a concept or looking for a rough steer, personas and panels are fast and cheap, and they help you sharpen a strategy before real money moves.

When the decision carries financial or strategic weight, the bar goes up. A population represents the full market instead of a slice of it. That supports the heavier work: statistically sound analysis, fine geographic detail and simulations that smaller synthetic datasets can't carry.

The more money riding on the answer, the closer to a full population you want to be.

The rule of thumb

The cheat sheet

If the job is to…Reach for
Find fresh angles, explore motivations or draft messagingPersona
Get a quick read on a concept before you commit resourcesPersona or panel
Compare audience segments or choose a targetPanel
Size a market or forecast demandPopulation
Build a propensity model that predicts who is likely to actPopulation
Judge a media investment or build a business casePopulation
Plan a retail network or read a hyper-local marketPopulation
Reach a niche audience too small or too costly to surveyPopulation

Where it shows up in agency work

The use cases run across the whole workflow, from strategy to execution and from understanding an audience to measuring what worked. In practice, that means a team can do four jobs in one place:

  1. Build audience personas from a library of customer attributes, choosing the variables that fit the audience and the research question.
  2. Estimate how a media mix will reach the segments you've defined.
  3. Test messaging concepts against synthetic respondents before anything goes to market.
  4. Calibrate location-based activation for a campaign.

Two further uses are aimed at bigger bets. One is rehearsal: test what might happen if millions of dollars move into a channel before a dollar moves. The other is a sandbox, a safe place to watch how an autonomous AI campaign behaves before it meets a real consumer.

What you can use now

  • Write the decision at the top of the brief. Then name whether it calls for a persona, a panel or a population.
  • Sort every "synthetic" tool already in your stack into persona, panel or population. If you can't tell which it is, ask whoever supplies it.
Section 04 · The Watch-Outs

Five places the simulation drifted.

Knowing where a tool is weak is as useful as knowing where it's strong. Arima's side-by-side turned up five conditions where the synthetic answers pulled away from the human ones.

Arima shared the misses along with the matches, which is what makes the test so useful. Each one comes with a workaround.

The rewrite, at a glance

Instead ofAsk it this way
"How will your finances look in six months?""How are your finances today compared with six months ago?"
Asking everyone, then filtering by genderAsking about men and about women directly
A 10-point scaleA 5-point scale

The uncomfortable truth

Go back to the admission at the top of the test: nobody is holding the answer key. Arima says plainly that no comprehensive outside dataset exists to benchmark a synthetic population against. Validation is a comparison between two imperfect instruments, a model and a survey.

Assume that's true of any synthetic data you use until someone shows you otherwise. The useful question shifts from "is it accurate?" to "accurate compared with what, on which kinds of questions, and checked by whom?"

Five known weak spots are worth more than a match rate. They're a map of where to be careful.

The takeaway

What you can use now

  • Rewrite your questions for the simulator. Past and present over future. Cuts built in. Coarse scales. English first.
  • Keep a human check on the big calls. Where a result will move real budget and falls in one of the five weak spots, confirm it another way.
Section 05 · The Method

Putting it to work, step by step.

The three kinds and the five weak spots do most of the work. Here's how they fit together on a real brief.

  1. Name the decision. Write down what you're deciding and how much rides on it. That sets how much proof you need.
  2. Pick the kind. Persona for ideas and language. Panel for direction across segments. Population for sizing, forecasting and anything with a budget attached.
  3. Write the questions for the simulator. Past and present over future, subgroups built in, coarse scales, and extra care with "don't know" and non-English questions.
  4. Check it against something real. Include a few questions you already have human answers for. Compare the headline, the subgroups and the call.
  5. Tell the client what it is. Label synthetic results as synthetic, say whether they're directional or diagnostic, and note any weak spots that apply.

What a population makes possible

Arima's team, which builds full synthetic populations, points to four things a population can do that a handful of personas can't.

A census every time
Sampling exists because asking everyone is impractical. A synthetic population puts each question to every synthetic individual in the target audience.
Small places, small audiences
Subgroups at low levels of geography, or niche targets that would be impossible or too costly to survey.
Diagnostic, not just directional
In creative testing, scores on each dimension explain why an execution works or doesn't, so the team can fix the weak part instead of starting over.
A building block
Fused with first-party data such as loyalty, transaction or CRM records, it gives a wider view of customers and markets than that data alone.

The trade-offs are real. A population takes sophisticated modeling, niche behaviors are only as good as the source data, and for exploring ideas a persona or panel is faster and cheaper. Use the kind the decision needs.

Section 06 · The Growth Case

The research department you never had.

For a smaller shop, the case fits in a sentence: synthetic data can level the competitive playing field.

The case is practical. Synthetic data promises faster insight, lower research costs, more flexibility and more detail than traditional data. For a shop without a research team down the hall, that changes what you can bring to a pitch. You can test a message, size an audience or model a media plan before the meeting, and show your work.

Three ways it helps an indie grow

  1. Win the pitch. Walk in with a tested message and a sized audience instead of a hunch.
  2. Be the adviser in the room. Your clients are getting synthetic data pitches too. The agency that can tell a persona from a population is the one they call.
  3. Add a service. Testing and research you couldn't afford to offer before can become something you sell.

Learn it while it's cheap

It won't stay new for long. Gartner estimates that by 2028, 60% of the world's top 100 retailers will use synthetic population data to validate marketing strategies, up from fewer than 20% today (figures as cited by Arima). The more abstract the data gets, the more it rewards creativity upstream, which is where indies do their best work.

The sensible first move is modest. Dip a toe in on a small brief. Find out where synthetic data adds value in your workflow and where it doesn't, while the stakes of learning are still low and before a client asks what you think of it.

The through-line

Put the sections together and one idea runs through all of them. The technology isn't the hard part. The matching is. A persona pressed into service as a population will fail, and so will a population asked about a future it can't see. Every warning in this Blueprint is a version of the same instruction: fit the tool to the question.

Know which kind you're using. Ask where it breaks. Let the size of the decision set the standard of proof.

The one-sentence version, in three sentences
Section 07 · The Toolkit

Eight questions for any synthetic data.

Run these before synthetic data goes in front of a client, whether it's a provider's, a client's or your own. The checklist comes from Arima's team. We added the reason behind each question.

1

Is this a synthetic persona, a synthetic panel or a synthetic population?

Everything else depends on this. If nobody can answer in one word, keep asking.

2

What are the source datasets, by name, and when was each last refreshed?

Synthetic data can't be better than what it was built from. You want names and dates.

3

Are the results qualitative or quantitative, and where does it plug into our workflow?

The first half tells you what you can claim with the output. The second tells you whether you'll use it.

4

Has it been checked against a live survey? Where does it match, and where doesn't it?

A good answer covers three kinds of agreement and a list of misses.

5

Which variables come from real source data, and which are inferred by the model?

Inferred isn't bad. Unlabeled is. You need to know which numbers were measured.

6

What's the smallest geography and the smallest audience it can reliably describe?

Granularity is a population's big strength. Know where it runs out.

7

Can it be combined with our own or our client's first-party data?

This is the building-block test. It decides whether you end up with something proprietary.

8

What can this not answer?

The most useful question on the list. A specific answer tells you exactly where to be careful.

Want a hand?

Arima can help you put this to work.

Arima, our expert partner on this Blueprint, builds synthetic populations and runs validation tests like the one in section 02. For help choosing the right kind, setting up a side-by-side test or bringing synthetic data into a client project, Indie Agency News can introduce you.

hello@indieagency.news

Section 07 · The Closer

Practical next steps.

Six moves. Pick two and start this week.

  1. Name the decision first. Before any demo, write down the decision you're trying to make and how much money rides on it. That sets how much proof you need.
  2. Audit what you already use. Label every synthetic data tool your agency or your clients use as a persona, a panel or a population.
  3. Run a small side-by-side. Take a question you already hold real survey results for. Put it to the synthetic tool. Compare the headline, the subgroups and the call you'd make.
  4. Rewrite one questionnaire for the simulator. Past and present over future, cuts built in, coarse scales, a note beside every "Don't know."
  5. Run the eight questions before client work. Use the Data Check in the companion tool and keep the result with the project file.
  6. Start small and cheap. Use a persona or a panel on a low-stakes brief this quarter. Learn where it helps before a client asks you whether you use it.
Section 08

The words, decoded.

Synthetic data
Data generated from one or more real sources using statistics. It mirrors the patterns in the source without copying any actual person.
Synthetic persona
A simulated individual you can question. Works like a focus group: good for ideas, not for numbers.
Synthetic panel
A larger group of synthetic respondents. Works like a research panel: good for comparing segments and getting a direction.
Synthetic population
A synthetic stand-in for an entire market. Works like a census: good for sizing, forecasting and fine geographic detail.
Population simulator
Another way to think of a synthetic population: a simulated market you can put questions to.
First-party data
Data a company collects itself from its own customers, such as loyalty, purchase and CRM records.
CRM
Customer relationship management. The system where a company keeps its customer records.
Census vs. sample
A census asks everyone. A sample asks some and estimates the rest.
Top-line results
The headline numbers from a survey, before any breakdown by group.
Marginal equivalence
Do the headline results from two methods match?
Segment equivalence
Do the subgroups show the same patterns in both?
Decision equivalence
Would both sets of data lead you to make the same call?
Directional vs. diagnostic
Directional tells you which option is ahead. Diagnostic tells you why.
Propensity model
A model that estimates how likely a person or group is to do something, such as buy or switch.
Location intelligence
Using geographic data to decide where to open, advertise or activate.
Hyper-local
Analysis at the level of a neighborhood or a few blocks.
Large language model
The kind of AI behind chat tools. The polling tool in this Blueprint's test used one to answer survey questions.
Sandbox
A safe test environment where nothing reaches real customers.
Global 2000
A ranking of the world's 2,000 largest public companies.
Arima

Get a second look from Arima.

Build a plan for your project in the Synthetic Data Planner, then send it to Arima's team for a second look before it reaches a client. They'll tell you whether you're tracking: the right kind for the job, the questions likely to drift and the best way to test the results. If it's headed into a pitch, they can work alongside you on it.

Build your plan →

Already have a plan or a question?

Email us with "Arima second look" in the subject, and we'll pass it to Arima's team.

hello@indieagency.news

Who made this

Expert partner: Arima. Arima's team contributed the subject-matter expertise behind this Blueprint: the three-kind framework, the validation research and the eight-question checklist.

Validation study by Arima with a third-party research firm.

Edited and produced by Indie Agency News.

About Indie Agency News

Indie Agency News is the global home for independent agencies. Our mission is to empower indies with visibility, resources and community, creating more opportunities for agencies and more choices for marketers.

Join us: indiehq.is/join · hello@indieagency.news

© 2026 Indie Agency News. The Blueprint Series. This Blueprint is for educational and illustrative purposes only and is not legal, financial, technical or purchasing advice. See the full note at the top of this Blueprint.