How to use it
- Count the questions in your draft by type: single choice, multiple choice, rows in grids, open-ended and numeric.
- Enter the counts; the calculator adds a fixed allowance for the introduction.
- Enter the number of completed interviews you need.
- Enter the expected incidence, the share of contacted people who qualify, and the expected response rate.
- Read the estimated duration and the number of invitations required.
Formula and logic
Each question type is given an average answering time:
| Question type | Seconds |
|---|---|
| Introduction (once) | 30 |
| Single choice | 7.5 |
| Multiple choice | 10 |
| Each row of a grid or matrix | 5 |
| Open-ended | 30 |
| Numeric entry | 10 |
Duration in seconds = 30 + 7.5 × single + 10 × multiple + 5 × grid rows + 30 × open + 10 × numeric.
For fieldwork, the calculator works out how many people must be invited:
invitations = target completes / (incidence × response rate)
Example: 10 single-choice questions, 4 multiple-choice questions, 2 grids of 6 rows each, 2 open questions and 2 numeric questions add up to 285 seconds, or just under 5 minutes. For 400 completes with an assumed incidence of 30% and a response rate of 10%, you would need about 13,334 invitations. How long fieldwork takes then depends on how many invitations your sample source can send per day.
How to read the result
The timings are averages for self-completion online surveys. Interviewer-administered surveys by telephone or face to face run longer, because every question and answer option is read aloud. Older respondents, complex topics and B2B questions that need a moment's thought also take more time, so add a margin for those audiences.
Duration matters because data quality falls as questionnaires grow. Long grids invite straightlining, late questions get hurried answers and more respondents drop out before the end. The most effective fix is to remove every question that does not feed the decision, and to break large grids into shorter blocks.
The invitation estimate is only as good as the incidence rate and response rate you enter. Use figures from earlier waves, a pilot test or desk research where you can. To decide how many completes you need in the first place, use the sample size calculator.
Frequently asked questions
How long should an online survey be?
As short as the decision allows. There is no fixed limit, but every extra minute increases dropout and careless answering, especially on mobile phones. Write down which decision each question informs; any question without an answer to that test is a candidate for removal. If the estimate still looks long, consider splitting topics across two shorter surveys.
What is the incidence rate?
It is the share of people contacted who meet the criteria for the survey, for example people who bought running shoes in the past year. A low incidence means many people must be screened out for every qualifying respondent, which raises the number of invitations and lengthens fieldwork. Desk research or a short screening wave can give a first estimate.
Why are grid rows counted separately?
Because a grid with ten statements is really ten questions presented on one screen. Respondents read and answer each row, and long grids are where fatigue shows first. Counting rows rather than grids gives a much more realistic duration and makes the cost of adding statements visible while you are still drafting.
Does the calculator work for telephone and face-to-face surveys?
It gives a lower bound. Spoken interviews take longer because the interviewer reads each question and answer list, and respondents often comment before answering. For those modes, treat the result as a minimum and confirm the real duration in a pilot. For help planning fieldwork across modes, see our survey research service or request a proposal.
Related reports
Surveys
Survey research measures how many people in a defined population hold a view, behave in a certain way or would choose…
Read →B2B research
B2B market research studies how organisations, rather than individual consumers, choose, buy and use products and…
Read →Consumer research
Consumer research is the systematic study of how people buy and use products in a category: who buys, how often, where,…
Read →