Consulting and recruiting
How to value a market research firm
By SourceX Editorial · Reviewed by Noah Loul ·
Short answer
A market research firm is usually valued on adjusted earnings times a multiple, and the multiple moves with the quality of those earnings: recurring trackers, a diversified client base, proprietary panels and norms, documented methods and a team that stays. Proprietary data adds value only when the firm can show it holds the rights to use it.
Key takeaways
- Buyers price the quality of earnings, so recurring tracker revenue usually counts for more than one-off ad hoc projects.
- Client concentration is a common discount; know your top clients' share and contract terms before buyers ask.
- Proprietary panels and norms databases add value only when consent, client contracts and platform terms support their use.
- In an acquisition, databases can be recognized as intangible assets separate from goodwill, so documentation matters.
- Buyers now ask which deliverables AI can produce and which data the firm holds that AI cannot generate.
How are market research firms usually valued?#
Market research firms are usually valued on normalized earnings, typically EBITDA for larger firms or seller's discretionary earnings for owner-run ones, multiplied by a figure that reflects risk and growth. Buyers cross-check against comparable deals and sometimes a discounted cash flow model.
Public data on multiples for research agencies specifically is thin, and published ranges for broader marketing services often mix very different businesses. Treat any headline multiple with caution and ask an advisor who has closed research agency transactions for comparable deals.
The earnings figure is adjusted before any multiple is applied: owner pay is set to market, one-time costs come out, and revenue recognition is checked so that a tracker billed in advance is not counted twice.
Which value drivers move the multiple?#
The value drivers that move the multiple are the ones that make future earnings more predictable or harder to copy. The table shows how buyers typically read each driver and what evidence they will request in diligence.
| Driver | Raises value | Lowers value | Evidence buyers request |
|---|---|---|---|
| Revenue mix | Multi-year trackers and retainers | Mostly one-off ad hoc projects | Revenue by type and client over several years |
| Client concentration | No single client dominates | One or two clients carry the business | Top-client share, contract terms, renewal history |
| Proprietary panel | Engaged, profiled, consented members | Aging panel with thin consent records | Recruitment sources, notices, activity data |
| Norms and benchmarks | Category norms clients pay to compare against | Benchmarks built on client data without permission | Database contents, source studies, rights basis |
| Methods and IP | Documented, named methods and tools | Methods that live in one founder's head | Method documents, IP ownership, contractor assignments |
| Team | Second-line leaders who stay | Dependence on the owner | Org chart, tenure, retention agreements |
| AI exposure | AI used inside workflows with clear controls | Revenue tied to deliverables clients can now draft with AI | Tool inventory, vendor terms, margin trend by service |
| Data rights | Clear records of who owns what | Unclear rights to reuse respondent or client data | Consent archive, client MSAs, platform terms |
How proprietary data assets show up in a deal#
Proprietary data assets show up in a deal as identifiable intangibles. Under US acquisition accounting, an intangible asset is recognized separately from goodwill if it arises from contractual or legal rights or is separable, and the standard's examples list databases among technology-based intangibles and customer lists among customer-related ones.
That only helps a seller whose data can be described and defended. A norms database with a documented list of source studies, a rights basis and an update history can be valued as an asset. The same database built on client-funded studies with no reuse clause becomes a diligence risk instead.
Panels work the same way. Buyers look at how members were recruited, what notices they accepted, how active they are, and whether consent supports the uses the buyer has in mind, including any AI use.
What buyers ask about AI in diligence#
Buyers ask about AI in diligence from two angles: how exposed current revenue is to AI-produced work, and whether the firm holds data and methods AI cannot replicate. Clear written answers shorten diligence; vague ones tend to become price adjustments, escrows or specific indemnities in the purchase agreement.
- Which services could a client now produce with AI tools, and how much revenue do they represent?
- Which AI tools does the firm use, on which data, under which vendor terms?
- Do respondent notices and client contracts allow any planned AI use of past studies?
- Has the firm licensed data to anyone, on what terms, and is any license exclusive?
- Which records, such as norms, item banks, questionnaire libraries and project reviews, are proprietary and documented?
Does data licensing revenue help or hurt valuation?#
Data licensing revenue can help valuation when it is recurring, documented and free of restrictions that bind a buyer. A non-exclusive, time-limited license with a clear rights record reads as a new revenue line; an exclusive license that blocks the acquirer's own plans reads as a constraint.
Buyers will also check that licensed data was licensed, not sold outright, so the firm still owns what it was paid for, and that each license survives a change of control without surprises. Keep the agreements, the rights review and the approvals in the data room.
How the revenue is recognized also matters to a buyer's model. Upfront fees, staged payments and usage-based terms can be treated differently under revenue recognition rules, so have your accountant review each license before it appears in the earnings a buyer will multiply.
Illustrative: an owner prepares a research firm for sale#
Illustrative: the fictional owner of a 75-person research firm plans a sale to a larger insights group. Most revenue comes from three category trackers plus ad hoc work for a handful of large clients.
Before engaging bankers, the owner commissions a quality-of-earnings review, documents the firm's norms database with a list of source studies and the client permission behind each, and has every freelance analyst sign an IP assignment. A review of panel notices shows older members never agreed to model building, so the firm records that limit rather than leaving a buyer to find it.
The owner also inventories internal records, such as questionnaire libraries, fieldwork logs and project reviews, and runs a metadata-only licensing fit check. The result is not a number but a cleaner story: what the firm owns, what it may use, and what a buyer is getting.
Steps to take before you talk to buyers#
Each step below removes a question a buyer would otherwise answer with a discount, and most of them also make the firm easier to run in the meantime.
- Normalize earnings and separate tracker, retainer and ad hoc revenue for several years.
- Calculate client concentration and gather the contracts behind your largest clients.
- Document panels, norms and item banks with source studies, consent basis and update history.
- Collect IP assignments from employees and freelancers who built methods and tools.
- Inventory AI tools, the data each one touches and the vendor terms behind them.
- Record any data licensing, with terms, exclusivity and approvals.
How SourceX frames research data value#
SourceX frames research data with the SourceX Enterprise Data Value Framework, a SourceX methodology rather than an industry standard. Uniqueness, domain expertise, human-generated signal, scale, recency, data cleanliness, rights and AI utility increase value; exclusivity increases price; reproducibility reduces value; and preparation cost and privacy burden reduce net value. It produces qualitative ratings, not prices or valuation figures.
If a firm does license records, each package runs through the SourceX five-step transaction, from Supply through Delivery, and the SourceX Evidence Packet documents where the data came from, the rights behind the license, the permitted use, how privacy was handled and who authorized release. That file is the kind of record a buyer's diligence team asks for.
Frequently asked questions
Is there a standard EBITDA multiple for market research firms?
No reliable public figure exists for research agencies specifically. Multiples depend on size, growth, revenue mix, concentration and buyer type, and published ranges for broader marketing services mix different businesses. Ask an advisor for recent comparable research agency transactions rather than relying on a headline range.
Does a proprietary panel always add value?
No. A panel adds value when members are engaged, profiled and recruited under notices that support the buyer's intended use. An inactive or poorly documented panel can cost more to fix than it is worth, and consent gaps may rule out AI uses a buyer had planned.
Should I license data before selling the firm?
Only if the license is clean: non-exclusive or narrowly exclusive, time-limited where sensible, and documented with a rights review. A well-documented license can show that the data has value; a rushed or exclusive one can narrow the buyer pool. Discuss timing with your deal advisor first.
What will buyers want in the data room about data?
Expect requests for respondent notices by period, panel terms, client MSAs and amendments, platform and vendor terms, any data licenses, a data inventory with systems and years of history, and records of privacy rights requests and incidents.
How does AI exposure affect quality of earnings?
Buyers look at whether revenue depends on work clients can now do themselves with AI, such as simple reporting or quick concept screens. Showing that the firm has already repriced or repackaged that work, with stable margins, reduces the risk a buyer will price in.
Sources
- Under ASC 805, an intangible asset acquired in a business combination is recognized separately from goodwill if it arises from contractual or legal rights or is separable, and the illustrative examples list databases among technology-based intangible assets and customer lists among customer-related intangible assets. Source
Related resources
See if your company qualifies
A short company assessment. No data uploads are needed.