In our guide to research participant incentives, we walked through a six-group qualitative job that lost £750 without anyone making a bad decision. Every over-recruit was sensible. Every payment was correct. The money disappeared because the quote never allowed for what the field team was always going to do.
That is the pattern behind most fieldwork margin problems. By the time a job is in field, the price is fixed and the only lever left is effort. The real decisions were made days earlier, usually by one experienced person with a spreadsheet, a gut feel for incidence and a client waiting for a number.
This guide moves upstream from incentives to the quote itself. It is written for field directors, operations leads and the people who actually build fieldwork costings at UK research and fieldwork agencies. It covers what really drives the cost of a job, the maths of incidence and strike rates, the 2026 labour floors that old rate cards may now sit beneath, the quality costs that are easy to forget, and how to stop quoting from memory and start quoting from your own data.
"A fieldwork quote is a forecast. The agencies with healthy margins are not the ones that guess best. They are the ones that check every forecast against what actually happened."
The anatomy of a fieldwork quote
Whether the job is qualitative recruitment, face-to-face interviewing or telephone fieldwork, the cost is built from the same handful of drivers. The difference between a profitable quote and a painful one is usually not the rate card. It is whether each driver was estimated, or assumed.
| Cost driver | What it really depends on | How it goes wrong |
|---|---|---|
| Incidence rate | The share of people screened who qualify | Estimated optimistically to win the job, then reality halves it |
| Strike rate | How many qualifiers agree, are available and actually attend | Treated as a constant when it varies hugely by audience, day and venue |
| Length of interview or session | Questionnaire or discussion guide length, plus overruns | Questionnaire grows after sign-off; price does not |
| Quota complexity | Interlocking quotas, hard cells, exclusions | Last few hard cells consume a disproportionate share of hours |
| Over-recruitment and floaters | Audience reliability and the client’s tolerance for empty seats | Recruited and paid, but never quoted |
| Incentives | Audience, time, method and fees | Quoted for seats, paid for people |
| Labour cost | Pay model, wage floors, holiday pay, employment status | Rate cards set years ago, not updated for new wage floors |
| Quality control | Validation, back-checks, listening-in, re-recruits | Done properly, but absorbed rather than priced |
| Field management | Briefing, daily reporting, client changes | Senior time spread thinly and never recorded against the job |
| Pass-through costs | Venues, viewing facilities, travel, hosting, translation | Quoted at last year’s prices, or missing entirely |
Two of these drivers, incidence and strike rate, do more damage than all the others combined, because cost does not move in a straight line with them. That is where we start.
Incidence and strike rate: the maths that decides the job
Incidence rate is the share of people screened who qualify for the study. Kantar describes it as both a cost and a feasibility indicator, and draws the distinction every field team knows: the predicted incidence you quote on is not the actual incidence you discover in field.
Strike rate (some agencies call it conversion or cooperation) is what happens next: of the people who qualify, how many agree to take part, are free on the right date, and turn up. For quantitative fieldwork the equivalent is the completion rate once someone has started.
Put them together and the number of people you need to contact is:
Contacts needed = Recruits required ÷ (Incidence rate × Strike rate)
Recruiter hours = Contacts needed ÷ Contacts per hour
The important word is divide. Because incidence sits underneath the line, cost does not rise steadily as incidence falls. It accelerates.
A worked example
Take a qualitative job needing 40 recruits (four groups of eight, plus two over-recruits per group). Assume a 40% strike rate, 12 contacts per recruiter hour and an illustrative fully loaded recruiter cost of £20 per hour. Only the incidence changes.
| Incidence | Contacts needed | Recruiter hours | Recruitment labour cost |
|---|---|---|---|
| 20% | 500 | 42 | £833 |
| 15% | 667 | 56 | £1,111 |
| 10% | 1,000 | 83 | £1,667 |
| 5% | 2,000 | 167 | £3,333 |
Moving from 20% to 15% incidence adds about a third to the recruitment cost. Moving from 10% to 5% doubles it. The same five-point error costs four times as much at the bottom of the table as at the top.
That has a practical consequence: the lower the incidence, the more your quote depends on evidence rather than instinct. A seasoned field director can usually estimate a general consumer audience within a few points. For niche B2B, rare conditions or tightly interlocked quotas, a few points is the whole margin.
Strike rate is not a constant
Most costing spreadsheets hold strike rate as a single number. In reality it moves with the audience, the day of the week, the time of year, the venue location, the incentive level and whether sessions are online or in person. An evening group in a city centre in December does not convert like a Tuesday afternoon online depth. If your quoting model cannot distinguish between them, it is averaging away exactly the risk you need to price.
Hard cells and the last 10%
Quota jobs have their own trap. The first 80% of a sample often fills at close to the quoted rate. The final cells, the older men in rural postcodes, the decision-makers in a specific sector, can take as many hours as everything before them. A quote that applies one blended incidence to the whole job will look fine on day one and wrong by day five. Where quotas are tight, cost the hard cells separately.
Labour costs: check your rate card against the 2026 floors
Many fieldwork agencies pay interviewers and recruiters per completed interview or per recruit rather than per hour. That model is popular for good reasons: it is simple, it rewards productivity and it moves some incidence risk away from the agency. But it carries a compliance risk that is easy to miss in a quote.
The National Living Wage
From 1 April 2026, the National Living Wage rose to £12.71 per hour for workers aged 21 and over, with lower rates for younger workers and apprentices. If any of your field team are workers or employees rather than genuinely self-employed, their pay must meet the relevant minimum for the time they work.
Per-interview and per-recruit pay
Paying per output is allowed for workers, but the rules are specific. Under the rated output work rules, an employer can pay a “fair” piece rate instead of paying for every hour, but only where the contract sets no working hours, the employer does not determine or control the hours worked, a mean hourly output rate has been established and the worker has been given the required notice. HMRC’s guidance on calculating fair piece rates sets the piece rate as the minimum wage divided by the average hourly output rate, multiplied by 1.2, and it must be recalculated whenever minimum wage rates change.
Here is why that matters for quoting. Your average hourly output rate is driven by incidence and strike rate. When a job comes in at half the expected incidence, recruits per hour fall, and a flat per-recruit fee that looked generous at the quote can fall below the floor in practice. The incidence risk you thought you had passed to the field team can come back as a compliance risk.
Holiday pay for irregular hours
For irregular-hours and part-year workers, the 2024 reforms allow holiday to be accrued at 12.07% of hours worked and permit rolled-up holiday pay, itemised separately on payslips, for leave years starting on or after 1 April 2024. If your field labour is costed at the base rate only, every hour in your quote is understated.
Employment status
Whether a freelance interviewer or recruiter is self-employed, a worker or an employee depends on the facts of the relationship, not the label in the contract, and it determines which of these rules apply. It is a question for an employment lawyer or adviser, not a blog post. The operational point is narrower: your quoting model should use the true loaded cost of the labour you actually use, including wage floors and holiday pay where they apply, and it should be updated every April rather than whenever someone remembers.
Quality costs: price what the standard requires
Good agencies do the quality work regardless. The question is whether they charge for it.
The Interviewer Quality Control Scheme standard sets a minimum of 10% validation across methodologies, with 5% for telephone-centre validation during fieldwork. That validation involves people re-contacting participants, listening to interviews and checking records, and every one of those activities takes time that belongs on the job.
Add the quality work that is not in any standard but happens on every job: re-screening participants on the day, replacing no-shows, re-recruiting when a client changes the spec, and checking for professional respondents (we covered the payment-side fraud checks in the incentives guide). If those hours are not in the quote, they come out of margin.
A simple discipline helps: express quality as a line in the costing model, calculated from the volume of interviews or recruits, rather than as a vague contingency. It is then visible, defensible to the client, and measurable after the job.
Over-recruits, incentives and the seats-versus-people gap
The £750 example from our incentives guide came from one gap: the quote counted seats, but the field team paid people. Over-recruits and floaters were recruited, briefed and paid in full, but never priced.
The fix sits in the quote, not in the field:
- Quote recruits, not seats. If you will recruit ten to seat eight, cost the recruitment for ten.
- Quote incentives for everyone you will pay. Show over-recruitment as an explicit, agreed line so it can be rebilled.
- Include payment fees. Gift card, payout and transfer costs belong in the incentive line or the handling fee.
- Agree the treatment of no-shows and sent-home participants before field opens, not when the invoice is disputed.
If you are also deciding how incentives should appear on the client invoice, the disbursement versus recharge section of the incentives guide covers the VAT question.
Protecting the quote: assumptions, triggers and change control
No quote survives contact with the field perfectly. The agencies that protect their margin make that explicit rather than hoping for the best.
State your assumptions
Every fieldwork quote should list the assumptions it was built on: incidence, strike rate, length of interview, quota structure, number of over-recruits, incentive levels and dates. This is not small print. It is the basis for every conversation that follows.
Agree re-quote triggers
Tie those assumptions to clear triggers. If actual incidence falls materially below the quoted figure, if the questionnaire grows beyond the agreed length, or if quotas change after field opens, the job is re-costed. Agree the threshold and the mechanism up front, so the conversation in week two is about applying a rule rather than asking for a favour.
Measure incidence early
Actual incidence becomes visible within the first day or two of screening. If your team can see screened, qualified and recruited counts against the quote in real time, you can raise a re-quote while the client still has options. If those numbers live in recruiters’ individual spreadsheets and are only totalled at the end, you find out when it is too late to do anything except absorb the cost.
"The cheapest point to fix a bad incidence estimate is day two of field. The most expensive is the day you invoice."
From gut feel to evidence: quote from your own job history
This is where the biggest improvement lies, and where most agencies are furthest behind.
Every job you run produces exactly the data you need to quote the next one accurately: actual incidence by audience, strike rate by method and venue, hours per recruit, validation time, over-recruit rates, incentive spend and final margin. In most agencies that data is scattered across recruiter trackers, the field system, timesheets, the incentive platform and Xero, so it is never brought back to the quoting stage.
When it is joined up, three things change:
- Quotes are built from evidence. “Last time we recruited this audience, actual incidence was 11% and strike rate 35%” is a far stronger starting point than “feels like about 15%.”
- Knowledge stops walking out of the door. The estimating instinct of your most experienced field director becomes a shared, improving dataset.
- You learn which work actually makes money. Some audiences, clients and methods will consistently overrun. You can price them correctly, or decide not to chase them.
Five quote-versus-actual measures worth tracking
| Measure | How to calculate it | What it tells you |
|---|---|---|
| Incidence variance | Actual incidence minus quoted incidence, by audience type | Whether your estimates are systematically optimistic |
| Hours variance | Actual field hours divided by quoted field hours | Which job types overrun, and by how much |
| Over-recruit rate | People recruited divided by seats or completes required | Whether your standard allowance matches field reality |
| Re-quote rate | Jobs re-quoted divided by jobs that breached a trigger | Whether your change control is used, or just written down |
| Margin variance | Actual job margin minus quoted margin | The single number that shows whether your quoting works |
The quality of that evidence depends on the database it comes from. Our guide to managing a research participant database covers how to keep profile, participation and screening data accurate and compliant.
None of these needs sophisticated analytics. They need the quote and the actuals to live against the same job record. That is the same connection we argued for in our piece on research agency utilisation rates: when time, cost and the original plan sit in different places, the business cannot see where its margin is going.
What a good quoting system looks like
Whether you extend an existing platform, buy a tool or build something around your operation, these are the capabilities that make a measurable difference:
- A costing model, not a blank spreadsheet. Recruits, incidence, strike rate, contacts per hour, length of interview and quota cells drive the labour estimate automatically.
- Historical benchmarks at the point of quoting. Actual incidence and strike rates from similar past jobs shown alongside the estimate being entered.
- Rate cards with effective dates. Wage floors, holiday pay and supplier prices updated once, centrally, and applied to every new quote.
- Explicit lines for quality, over-recruitment and incentives. Nothing important hidden in a contingency percentage.
- Assumptions and triggers printed on the quote. Generated from the same data used to cost the job, so they are always consistent.
- Live field tracking against the quote. Screened, qualified, recruited and hours used, compared with plan from day one.
- Automatic quote-versus-actual reporting. Margin variance per job, and trends by audience, client and method, without anyone building a report.
- Integration with accounts. Final costs and rebills flowing to Xero or your accounting package, so margin is real rather than estimated.
Generic professional services platforms tend to handle the commercial wrapper well, such as proposals, approvals and invoicing, but rarely understand incidence, strike rates or quota cells. That gap is one of the reasons we suggested in our Accelo vs Scoro comparison for fieldwork agencies that many agencies are better served by keeping a core platform and building the fieldwork-specific layer around it.
Is your quoting costing you margin?
Send us three recent jobs: one that went well, one that overran and one you are not sure about. We will compare what was quoted with what actually happened, show you where the variance came from, and tell you honestly whether you need software, a better template or just a re-quote clause.
Book a free discovery call →Where to start
You do not need to rebuild your operation to quote better. In order of effort:
- Add an assumptions block and re-quote triggers to every quote. This costs nothing and protects you from the next low-incidence surprise.
- Update your rate card for April 2026. Check per-interview and per-recruit rates against the current wage floors and holiday pay rules, with advice where status is unclear.
- Price quality and over-recruitment explicitly. Make them lines, not absorption.
- Capture actuals against every job. Even a consistent end-of-job form recording incidence, strike rate, hours and over-recruits starts building your evidence base.
- Connect quoting, field and finance. When the volume justifies it, bring the costing model, live field data and accounts together so quote-versus-actual is automatic.
If quoting is one of several processes still held together by spreadsheets, our practical workflow for replacing spreadsheets in a research agency explains how to sequence the change without disrupting live jobs.
We build these systems in Laravel for research, fieldwork and professional services firms, and we are currently building an operations platform with a UK fieldwork agency covering quoting, recruiters, quotas, incentives and job profitability. Arc Digital is based in Leeds and works with agencies across the UK.
Frequently asked questions
What is incidence rate in fieldwork?
Incidence rate is the percentage of people screened who meet the criteria for a study. It is calculated as the number of qualified people divided by the total number screened. It is one of the biggest drivers of fieldwork cost because the number of contacts needed rises sharply as incidence falls.
How do you calculate the cost of recruitment for a qualitative job?
Start with the number of people you will actually recruit, including over-recruits. Divide by the incidence rate multiplied by the strike rate to get the contacts needed, then divide by contacts per hour to get recruiter hours. Multiply by your fully loaded hourly labour cost, then add incentives for everyone you will pay, quality control, field management and pass-through costs.
What is the difference between incidence rate and strike rate?
Incidence rate measures how many people qualify. Strike rate measures how many of those qualifiers agree to take part, are available and attend. Both reduce the yield from each contact, so both need to be estimated separately and tracked against actuals.
Can we pay freelance interviewers or recruiters per interview?
Paying per output is common, but if the people concerned are workers or employees rather than genuinely self-employed, their pay must still meet minimum wage rules, either for the hours they work or through a properly calculated fair piece rate under HMRC’s rated output work rules. Employment status depends on the facts of the relationship, so take advice from an employment specialist.
Should validation and back-checking be charged to the client?
Validation is part of delivering a quality service, so its cost belongs in the price of the job. Whether you show it as a separate line or build it into your rates is a commercial choice, but it should be calculated from the volume of work and the standard you follow, such as the IQCS minimum of 10% validation, rather than absorbed.
What should happen when actual incidence is lower than quoted?
That should be governed by the assumptions and re-quote triggers in your quote. If incidence falls materially below the quoted figure, raise it with the client early, explain the options (re-quote, relax criteria, extend timings or reduce sample) and agree a way forward before the extra hours are spent.
If you run a research or fieldwork agency and your quotes and actuals rarely meet, we would be glad to take a look. We will trace a few real jobs from quote to invoice and tell you plainly where the margin went, whether or not the answer is software.