How to Conduct Target Audience Research Before Positioning Work

On this page

• Why Broad Audience Labels Are a Positioning Liability
• Step-by-Step ICP Refinement Using Quantifiable Filters
• A Named Framework: The ICP Definition Canvas
• Mapping ICP Attributes to Signals and Sources
• Validating Your Audience Assumptions
• Balancing Segmentation Granularity
• How Precise Target Audience Definition Improves Positioning
• Common Pitfalls to Avoid
• Conclusion: Validate Before You Position

If you’re a founder or growth leader in B2B SaaS, you know positioning can make or break your product’s success. Here’s the hard truth: most positioning fails because it starts with a vague or incorrect target audience. Labels like “SMB” or “mid-market” are lazy shorthand that kill clarity and waste marketing dollars. Before writing a single positioning statement, define and validate your target audience precisely. Here’s how.

Positioning is a decision about whose problem you solve best and why you win against the alternatives. That decision needs an input: a sharp, evidence-backed picture of the people who will buy. If you skip the research and start at the messaging stage, you are writing claims about a customer you have only imagined. The market corrects you later, slowly and expensively, through low reply rates, long sales cycles, and churn you cannot explain. This article walks through how to do the research first, in an order that holds up under pressure.

Why Broad Audience Labels Are a Positioning Liability

Calling your target market “SMB” is like saying your ideal customer is “someone who owns a car.” It’s too broad to guide messaging or sales strategy. SMBs vary wildly in industry, budget, team size, and buying behavior. Trying to appeal to everyone in that bucket means resonating with no one. This leads to diluted messaging, confused sales teams, and poor campaign ROI.

The deeper problem is that size bands are accounting categories, not buying categories. “Mid-market” is a revenue range a finance team uses to slice a TAM (total addressable market, the full revenue opportunity if everyone bought). It tells you nothing about why someone wakes up wanting your product. Two companies can both sit in the same revenue band and have nothing in common: one is a 200-person agency drowning in manual reporting, the other a 200-person logistics firm that has never felt the pain you solve. A label that groups them together will mislead every downstream decision.

Broad labels also hide the buying committee. In B2B, you rarely sell to a “company.” You sell to a champion who feels the pain, an economic buyer who controls the budget, and a set of blockers who can veto. A label like “enterprise” flattens all three into one faceless segment. Good audience research keeps them separate, because each one needs a different message and a different proof point.

Step-by-Step ICP Refinement Using Quantifiable Filters

Slice your broad audience into measurable segments using filters like:

• Industry vertical: Tech, healthcare, finance, manufacturing, etc.
• Sales team size: 1-5 reps, 6-15 reps, 15+ reps
• Marketing budget: <$50k, $50k-$200k, $200k+ annually
• Average deal size: <$5k, $5k-$20k, $20k+

These filters help create an Ideal Customer Profile (ICP) specific enough to target but broad enough to scale. For example, “Tech startups with 6-15 sales reps and marketing budgets between $50k-$200k” is far more actionable than “SMB.”

But firmographics alone are not enough. A complete ICP works across three layers: firmographic (who the company is), behavioral (what they do), and contextual (what is true right now that makes them ready to buy). The contextual layer is where most teams under-invest, and it is usually the strongest predictor of a closed deal. A company that just hired its first head of revenue operations is in-market for tooling in a way an identical company without that hire is not.

The table below shows how the three layers differ, and why each one matters for positioning.

ICP layer	What it describes	Example criteria	Why it shapes positioning
Firmographic	Stable facts about the company	Vertical, headcount, revenue band, region, funding stage	Sets the addressable boundary and the reference customers you can name
Behavioral	What the company already does	Tech stack in use, hiring patterns, content they publish, tools they renew	Reveals readiness and the “before” state your message must mirror
Contextual	What changed recently	New funding, leadership hire, regulation, reorg, public failure	Supplies the trigger event your campaign timing and hook depend on

Work from firmographic to contextual, not the other way round. Firmographics bound the universe; behavior narrows it to companies that could plausibly care; context tells you which of those will care this quarter. Positioning written against all three reads as if you understand the customer’s situation, because you do.

A practical refinement loop looks like this. Pull your last 12 to 24 months of closed-won and closed-lost deals. For each, record the firmographic, behavioral, and contextual attributes you can find. Then look for the attributes that separate won from lost. The ones that separate are your real ICP filters. The ones that appear equally in both are noise you should drop, no matter how intuitive they feel.

A Named Framework: The ICP Definition Canvas

To make this repeatable, use a single-page structure I call the ICP Definition Canvas. It forces every assumption into the open and ties each one to a piece of evidence, so positioning is built on what you can defend rather than what you hope.

The canvas has six blocks, filled in order:

1. Trigger. The event that puts an account in-market. State it as something observable from the outside (a new hire, a funding round, a tool migration), not an internal feeling.
2. Buying committee. Name the champion, the economic buyer, and the likely blockers. List the title, the metric each one owns, and what each fears.
3. Job to be done. The progress the champion is trying to make. Write it as a sentence in their words, not yours.
4. Current alternative. What they use today, including spreadsheets, a competitor, or doing nothing. Positioning is always relative to this.
5. Disqualifiers. The attributes that make an account a bad fit even when it looks right on paper. This block is as important as the inclusion criteria and is usually skipped.
6. Evidence. For every claim above, the source that backs it: an interview quote, a CRM pattern, a usage stat. Any block without evidence is flagged as an assumption to validate.

Filling the canvas takes a focused afternoon for a first draft and a few weeks of interviews to validate. The discipline is the point: by the time the canvas is full and every block has evidence, you have written most of your positioning’s raw material without realizing it.

Mapping ICP Attributes to Signals and Sources

A canvas is only useful if you can actually populate it. The hard part is moving from an attribute you want to know to a signal you can observe to a source you can pull it from. The matrix below maps common ICP attributes to a concrete signal and where to collect it. Use it as a checklist when you build your research plan.

ICP attribute	Signal to collect	Where to source it	Confidence
In-market trigger fired	New RevOps or growth hire in last 90 days	Job boards, LinkedIn, hiring pages	High
Champion feels the pain	Pain described unprompted in interview	Customer interviews, sales call recordings	High
Behavioral readiness	Specific tool present in stack	BuiltWith/Wappalyzer, integration directories	Medium
Budget authority	Stated budget range and approver	Discovery calls, win/loss interviews	Medium
Deal-size fit	Average contract value by segment	CRM closed-won records	High
Disqualifier present	Mandatory feature you lack	Lost-deal notes, support tickets	Medium
Expansion potential	Multiple teams with the same pain	Product usage by team, account mapping	Medium

Two rules keep this honest. First, prefer observed behavior over stated intent. What a prospect says in a survey is weaker evidence than what they actually did in their account or their hiring. Second, label your confidence. A high-confidence signal from CRM data can anchor positioning; a medium-confidence inference from a third-party tool should be treated as a hypothesis until an interview confirms it.

Validating Your Audience Assumptions

Don’t rely on gut feel or internal opinions. Validate your ICP with real data and customer insights:

• Surveys: Targeted surveys to customers or prospects confirm pain points, budgets, and buying processes. Keep questions quantitative and focused.
• Interviews: Structured interviews uncover motivations and objections numbers alone won’t reveal.
• Data analysis: Use CRM data, product usage stats, and sales records to identify patterns and validate segment characteristics.

External validation reduces bias and ensures your positioning targets real, addressable markets.

The order of these methods matters. Start qualitative, then go quantitative. Interviews tell you what to measure; surveys and data tell you how common it is. Run a survey first and you risk measuring the wrong things with great precision. A reasonable validation sequence for a single ICP hypothesis looks like this:

1. 8 to 12 customer interviews, split across recent wins and recent losses. This range is large enough to hear a pattern repeat and small enough to run in two to three weeks. You are listening for the trigger, the job to be done, and the words people use for the pain.
2. 5 to 8 lost-deal or churned-account interviews. These are uncomfortable to run and the most valuable. They surface disqualifiers and the real alternative, which your happy customers cannot see.
3. A survey of 100 or more in-segment respondents, but only after interviews have told you what to ask. Use it to size how widespread the pain and the trigger are, not to discover them.
4. A CRM and usage data pull to confirm the segment is real in your own numbers: deal size, win rate, sales-cycle length, and expansion by segment.

You know to stop when interviews start repeating and no new disqualifiers appear. That saturation point, not a fixed number, is the real signal that your ICP is validated. If you reach 12 interviews and still hear contradictory triggers, you have more than one ICP and should split them.

Screen interview participants tightly. A loose screen produces interviews with people who are not in your ICP, which feels like progress but pollutes your findings. Recruit against the firmographic and contextual filters from your canvas: the right title, the trigger fired, the alternative in use. Reject anyone who does not match, even when they are easy to schedule.

Balancing Segmentation Granularity

Too broad means generic messaging; too narrow risks costly over-fragmentation. Aim for segments that are:

• Large enough to justify dedicated marketing and sales efforts
• Homogeneous enough to share common pain points and buying triggers
• Distinct enough to require tailored messaging

This balance keeps your positioning relevant and scalable.

A useful test: a segment is worth treating separately only when it needs a different message, not just a different logo on the case study. If two segments respond to the same pain, the same trigger, and the same proof, they are one segment for positioning purposes, however different they look in a spreadsheet. Splitting them only multiplies the assets you maintain without improving conversion.

For an early-stage company, start with one ICP and one segment. Resist the urge to name three. You do not yet have the evidence or the resources to position credibly to more than one audience, and a single sharp position beats three blurry ones. Add segments only when your data shows a second group buying for a clearly different reason.

How Precise Target Audience Definition Improves Positioning

A clearly defined and validated target audience sharpens your positioning. Messaging hits the right pain points, sales teams speak the customer’s language, and marketing campaigns convert better. Companies that refined their ICP often see significant improvements in lead quality and faster sales cycles.

Concretely, a validated ICP feeds four positioning decisions directly. It names your competitive alternative (block four of the canvas), which is the thing your positioning differentiates against. It supplies the trigger that sets your campaign timing and your outreach hook. It defines the buying committee, so you write a different proof point for the champion than for the economic buyer. And it gives you the customer’s own language for the pain, which is the single most reliable source of copy that converts. None of these come from a size band. All of them come from research done before the positioning work begins.

Common Pitfalls to Avoid

• Assuming you know your audience: Internal biases lead to misaligned positioning.
• Skipping validation: Without customer input, your ICP is guesswork.
• Using vague labels: Broad terms like “SMB” or “enterprise” don’t translate into actionable strategy.

Two more traps are worth naming. Survivorship bias is the first: if you only interview happy customers, you learn why people who already love you love you, and nothing about why everyone else passed. The lost-deal interviews in the validation sequence exist to counter this. Confirmation in disguise is the second: writing survey questions that lead the respondent to the answer you want. Ask “what would make this useful?” not “would feature X be useful?” Leading questions return data that feels validating and means nothing.

A final, quieter pitfall is treating the ICP as finished. Markets move. Triggers that were rare become common, alternatives improve, a new entrant resets expectations. Revisit the canvas at least once a year, and immediately after any sharp change in win rate or sales-cycle length, since those numbers are your early warning that the audience has shifted under your positioning.

Conclusion: Validate Before You Position

Positioning without validated target audience research is a costly gamble. Refine your ICP with quantifiable filters, validate with real customer data, and balance segmentation wisely. This groundwork pays off in clearer messaging, stronger sales alignment, and better market traction.

Your next step is small and concrete: pull your last 20 closed deals, won and lost, and fill in the trigger and current-alternative blocks of the ICP Definition Canvas for each one. You will likely find a pattern within an hour, and a list of the interviews you need to run to confirm it. Do that before you touch the positioning statement.

If you need help refining your ICP or conducting customer research, get in touch. We help B2B SaaS founders and growth teams get their positioning right from day one—no guesswork, just results.

Contact us to validate your target audience before you reposition.