CatalystMR
Methodology Paper
B2B Research · Feasibility & Incidence

B2B Sample Incidence

A Methodology for Modelling Feasibility by Title and Industry
● 2026 Edition

A five-step workflow for estimating who will qualify for a B2B study — and how long it will take to reach them — before field begins. Built on treating incidence as a modelled range to be validated, not a published benchmark to be looked up, and grounded in the AAPOR, ESOMAR, and ISO 20252 standards buyers increasingly expect.

Published byCatalystMR Research Team
SeriesMethodology Papers
Reading time~17 minutes
Edition2026
Read the companion Insights article → ⬇  Download PDF
APA
CatalystMR Research Team. (2026). B2B Sample Incidence — A Methodology for Modelling Feasibility by Title and Industry. CatalystMR Methodology Papers. https://www.catalystmr.com/insights/methodology-papers/b2b-sample-incidence/
BibTeX
@techreport{catalystmr_b2b_sample_incidence,
  author={{CatalystMR Research Team}},
  title={B2B Sample Incidence — A Methodology for Modelling Feasibility by Title and Industry},
  institution={CatalystMR}, year={2026}, type={Methodology Paper},
  url={https://www.catalystmr.com/insights/methodology-papers/b2b-sample-incidence/}
}
RIS
TY  - RPRT
AU  - CatalystMR Research Team
TI  - B2B Sample Incidence — A Methodology for Modelling Feasibility by Title and Industry
PB  - CatalystMR
PY  - 2026
UR  - https://www.catalystmr.com/insights/methodology-papers/b2b-sample-incidence/
ER  -
Abstract

Incidence rate — the share of an accessible target audience that qualifies for a study — is the single variable that most often decides whether a B2B project comes in on time, on budget, and on specification. Yet it is routinely treated as a fixed number to be looked up, when in reality it shifts with seniority, function, company size, industry, geography, and decision authority, and again with every change to the screener.

Rather than offer benchmark tables, this paper gives a repeatable five-step workflow for modelling B2B incidence honestly — define the criteria, model the range, translate it into a fielding plan, validate with soft-launch, and revise — and then maps the levers that move a modelled incidence up or down. It deliberately publishes no incidence figures, because the right number is the one modelled from your audience, in your market, against your finalised screener.

01 The problem

Incidence is the share of your audience that qualifies — and it decides everything downstream.

In survey research, incidence rate is the percentage of people in the accessible target population who meet a study's qualification criteria — a close cousin of the qualification and eligibility rates that the AAPOR Standard Definitions codify for the field.2 A broad B2B study may qualify managers across many industries; a narrow enterprise-software study may require only budget owners at companies above a specific revenue threshold. The difference between those two is not a detail — it is the difference between a routine fortnight in field and a project that quietly fails to launch.

Why incidence is the variable that governs the others

Incidence is not just one input among many; it sets the ceiling on what is feasible. It drives screening effort, the incentives required, field time, the source mix, the total sample you must touch, and ultimately cost. A study's design can be flawless, but if its real incidence is a fraction of what was assumed, every other plan built on top of it gives way.

The relationship in one line
High incidence lowers cost and field time; low incidence raises screening effort, incentive needs, and feasibility risk. The narrower the industry, title, and decision role, the more the project's success depends on modelling incidence before field — not discovering it during.

The trap: treating a benchmark as a lookup

The most common — and most expensive — mistake is to treat incidence as a fixed figure to be retrieved from a table and dropped into a plan. Published "benchmarks" feel authoritative, but a benchmark drawn from a different audience, market, or screener tells you very little about your study. Incidence is a property of a specific question asked of a specific population at a specific moment; it is modelled, not looked up. The rest of this paper sets out how to model it — as a five-step workflow.

Principle

If a feasibility number arrives without the assumptions behind it — the audience, the market, the screener it was modelled against — it is not yet a feasibility estimate. It is a guess wearing the costume of one.

02 The method

Model incidence as a workflow, not a lookup.

Because incidence is a property of a specific study rather than a fixed fact, it has to be built — through a repeatable sequence that ends in a measurement, not an assumption. The five steps below are the spine of this paper; Sections 03–05 take them in turn, and Section 06 maps the levers that move the number up or down.

Step 1
1

Define the criteria

Capture title, function, firmographics, and behaviour as explicit, separate variables.

Step 2
2

Model the range

Estimate a banded incidence — upper and lower — tied to the finalised screener.

Step 3
3

Translate to a plan

Convert the range into field time, source mix, sample volume, and cost.

Step 4
4

Soft-launch

Release a small fraction to measure the real qualifying rate before committing.

Step 5
5

Revise

Re-model against measured data; trigger backups or a method change as needed.

Steps 4 → 5 loop until measured incidence and the plan agree

Why a workflow beats a benchmark

A benchmark answers "what is incidence for this kind of audience?" — a question your study has already outgrown the moment it is specified. The workflow answers the only question that matters: "what is incidence for this audience, against this screener, in this market — and how confident are we?" Each step narrows that uncertainty deliberately, and the final two steps replace estimate with evidence.

It is a loop, not a line

The workflow is drawn left to right, but Steps 4 and 5 feed back into the plan. Soft-launch almost always teaches you something the model could not — a qualifying rate slightly off, a cell filling slowly — and the discipline is to revise the plan against that evidence before full commitment, not to push on and hope the average recovers.

03 Steps 1 & 2

Define every criterion — then model a range, never a point.

The first two steps decide the quality of everything after them. Step 1 makes the target explicit; Step 2 turns it into a banded estimate. Incidence moves on three axes — title, firmographics, and behaviour — and a target's true incidence is the product of all three narrowing at once, not any single one.

Axis 1

Title & function

Seniority (Manager, Director, VP, C-Suite) and the functional lens — finance, IT, security, marketing, HR, operations.

Axis 2

Firmographics

Employee count, revenue band, geography, and industry vertical. The same title means different things across these.

Axis 3

Behaviour

Software usage, purchasing stage, installed base, recent purchase, and purchase involvement or authority.

Why the axes multiply, not add

Each criterion you add narrows the qualifying population, and they compound. A "Director of IT" audience may be readily feasible in online panel; CISOs at hospitals using a specific cybersecurity platform stack four constraints — seniority, function, industry, and a behavioural installed-base requirement — and may require telephone (CATI) outreach and partner sourcing to reach at all. This is why a single headline figure for "B2B decision-makers" is meaningless for planning: the moment you specify the study, you are no longer sampling that population.

Performance analytics and graphs on a screen
Fig. 01 — Each added criterion — title, firmographic, behavioural — narrows the qualifying base, and the constraints compound · Photo: Luke Chesser / Unsplash

Step 2: the screener is part of the incidence

The same audience can yield very different incidence depending on how the qualifying question is worded. "Do you influence IT purchasing?" and "Are you the final decision-maker for IT security tooling?" describe overlapping but very different populations. Until the screener is locked, any incidence figure is provisional by definition — which is why Step 2 produces a range with its assumptions attached, re-modelled once the screener is final. The ESOMAR framework explicitly asks providers what they do to put upper and lower boundaries around feasibility.1

BroadManagers, many industries
Higher incidence
SpecificDirectors, one vertical
Moderate
NarrowC-level + behavioural criteria
Low incidence

Illustrative only — relative widths show how added criteria compress the qualifying base. CatalystMR publishes no fixed incidence percentages; the figure for any study is modelled from that study's own audience, market, and finalised screener.

04 Step 3

A modelled range becomes field time, source mix, and cost.

A modelled incidence is only useful when it is translated into a fielding plan. The banded estimate from Step 2 is used to project field time, source mix, sample volume, and cost before launch — and to decide, for low-incidence audiences, whether online panel alone can carry the study or whether it needs reinforcement.1

What the range lets you project

  • Field time — how long it takes to reach the required completes at the modelled qualifying rate.
  • Sample volume — how many people must be touched to yield the completes you need.
  • Source mix — whether one panel suffices, or partner sourcing and CATI are required.
  • Cost — the screening effort and incentives that a given incidence implies.

When low incidence triggers a method change

For low-incidence B2B audiences, online panel alone may not reach the depth a study needs. This is the point at which a feasibility model should recommend telephone (CATI) interviewing, screen-sharing CATI for visual tasks, mixed-mode recruitment, or partner sourcing — not as upsells, but as the practical means of filling cells that online cannot.

Buyer's question

Ask directly: "At this incidence, can online panel alone deliver — and if not, what's the plan?" A credible answer names the method change before launch, not after the field stalls.

Qualifying depth, not headline panel size

A large B2B panel can still be thin for a narrow, low-incidence target. What matters for your study is not a total count of business professionals but the verified qualifying depth in your specific audience. Headline panel numbers tell you almost nothing about whether a narrow, behaviourally-defined cell can actually be filled — which is why incidence must be modelled against your criteria, not inferred from a provider's size.

05 Steps 4 & 5

Turn the estimate into a measurement before you commit the budget.

A modelled incidence is a hypothesis. Soft-launch is the experiment that tests it, and revision is what you do with the result. Releasing a small fraction of sample first converts an assumption into a measurement — so that if real incidence tracks the model, the full launch proceeds with confidence, and if it does not, you revise before the budget and timeline are committed rather than after the field is contaminated by the scramble to hit a number.

ConsiderationWithout soft-launchWith soft-launch
Incidence assumptionDiscovered wrong in full fieldMeasured on a small release first
Quotas & targetingAdjusted mid-field under pressureTuned before full commitment
Backup audiencesImprovised when the cell stallsNamed and ready in advance
Budget exposureCommitted before evidenceReleased against measured incidence

Step 5: plan the backups before you need them

The time to decide how you will respond to a low qualifying rate is before launch, not during it. A sound plan names its fallbacks in advance: relax the seniority floor, broaden the industry or behavioural criteria, add a telephone channel, or bring in partner sourcing — each chosen so the research objective survives the adjustment. Backups improvised mid-field, under deadline pressure, are how a study quietly drifts off its specification.

What soft-launch measures

  • Real qualifying rate — the actual incidence against the finalised screener, not the modelled estimate.
  • Drop-off points — where qualified respondents abandon the instrument.
  • Quota balance — whether cells fill at the modelled pace, or some lag.
  • Trigger to revise — the evidence to retune targeting or add a method before full commitment.
The discipline

Soft-launch is not a delay; it is the cheapest insurance in the project. A small measured release almost always costs less than a full field built on an incidence assumption that turns out to be wrong.

06 The levers

Every lever that moves the number — and which way it moves it.

Incidence is not a fact you discover; it is the net result of choices you make. Each lever below is a decision the study controls, and moving it tightens or loosens the qualifying base in a predictable direction. The settings are deliberately directional, not numeric — the magnitude is what Step 2 models and Step 4 measures for your own audience and screener.

◀ Tighten · incidence fallsThe leverLoosen · incidence rises ▶
Final decision-maker for one named tool
Screener wordingdefines who qualifies
Involved in the category decision
C-suite only
Seniority floortitle & level
Manager level and above
A single named vertical
Industry scopefirmographic
Any adjacent sector
One enterprise revenue band
Company sizefirmographic
Companies of any size
Uses one named platform
Installed basebehavioural
Buys the category, any vendor
Bought in the last quarter
Recency windowbehavioural
Has ever evaluated
No lever has a “correct” setting — only a trade-off between how tightly the study is targeted and how feasible it is to field. Modelling incidence is really deciding where each lever sits; soft-launch is checking the setting was right. Slide one toward the left and the number falls; slide it right and it rises — which is why a benchmark borrowed from a study whose levers sat elsewhere tells you almost nothing about yours.
Conclusion

The right incidence number is the one you modelled, not the one you looked up.

B2B incidence rewards planning at the front of the process and punishes assumptions at the back. Run the five steps every time: define the criteria across title, firmographics, and behaviour; model a banded range tied to the finalised screener; translate it into field time, source mix, and cost; validate it with soft-launch; and revise against the evidence before committing. Each step is independently sensible, and together they convert a hard-to-plan audience into a fielding plan you can stand behind. The AAPOR, ESOMAR, and ISO 20252 standards exist precisely so buyers can ask for this rigour in consistent terms — and recognise it when they see it.1,2,3

§ References
Standards are cited as the authoritative source for the definitional and quality-management frameworks referenced; the AAPOR Standard Definitions are cited for terminology only, not for any rate figure. Methodological principles stated without a citation (e.g. the relationship between incidence, field time, and cost, or that a narrower target lowers incidence) are definitional to survey sampling. This paper publishes no incidence percentages or benchmark rates by title or industry; any figure specific to a study should be modelled from that study's own audience, market, and finalised screener.
§ About CatalystMR

CatalystMR

CatalystMR is a global market-research panel and fieldwork partner specialising in hard-to-reach B2B, executive, and niche audiences. We model feasibility before field and pair verified online sample with live telephone (CATI) capability for the low-incidence targets that online alone cannot reach — under one screener, one QC standard, and one point of contact.

We publish our own responses to ESOMAR's 37 Questions and treat incidence as a modelled range to be validated by soft-launch, not a benchmark to be promised.

Compliance posture: our methodology is aligned to the ESOMAR Code and Guidelines and the ISO 20252 framework, and we are certified under the EU–U.S., UK, and Swiss Data Privacy Frameworks, with personal data siloed from response data.

IncidenceFeasibilityB2B SampleSoft-LaunchAAPORESOMAR 37
Tell us your audience, completes, incidence assumptions, and methodology needs, and we'll return a modelled feasibility range and a practical fielding plan — typically within 24 hours — not a hopeful number.
Request Feasibility →