Free Survey Drop-off Estimator

Find out how long your survey takes and where people quit, before you send it. Paste your questions one per line, pick your audience, and get an instant estimated completion time, an estimated completion-rate range, and the exact questions most likely to lose respondents. The estimate is based on published survey-length research, and everything is calculated locally in your browser, so your questions are never sent to a server. Free, no sign-up, download as Word or PDF.

Your survey

Paste straight from your draft. Blank lines are ignored, and nothing you paste leaves this page.

Who is answering

An incentive buys you more of the respondent's patience. It does not make a long survey short.

Estimate

Paste your questions on the left, one per line, and the estimate appears here as you type.

SURVEY DROP-OFF ESTIMATE

Survey: [Survey Name]

Audience: General public

Incentive: None

Method: reading time at 200 words per minute plus approximate published answer times per question type.

No questions yet

Paste your survey questions into the box on the left, one question per line. The estimate builds as you type.

Nothing you paste leaves your browser: the classification, the timings, and this report are all calculated locally.

Generated by BlockSurvey

How the survey drop-off estimator works

Three steps, and it runs in your browser.

1

Paste your questions

Drop your questions into the box, one per line, then pick your audience type and whether you are offering an incentive.

2

See the estimate

Each question is classified by type and timed, giving you an estimated completion time, an estimated completion-rate range, and the questions most likely to cause drop-off.

3

Download and fix

Export the estimate as a PDF or an editable Word file. Everything is calculated locally in your browser and never sent to any server.

Free, private, and built for better research

Most tools that tell you how long a survey takes want your email first, or upload your draft to someone else's server to find out. This one does neither. There is no AI model and no API call behind it: the timings come from published averages for each question type, and the arithmetic happens on your machine.

01

100% in-browser

Your questions are classified and timed locally. Nothing you paste is sent to a server, which is why you can run an unlaunched survey through it.

02

No account required

No sign-up, no email wall, no record of what you estimated.

03

Arithmetic you can check

The same questions always produce the same estimate. You can read the rules below and check the arithmetic yourself.

What the estimator checks

Five patterns that account for most abandoned surveys:

  1. The question at which cumulative time passes what your audience will sit through, which is where most drop-off begins.
  2. Clusters of three or more open-ended questions in a row.
  3. Rating grids and scale batteries stacked past the second one.
  4. Individual questions over 40 words, which get skimmed rather than read.
  5. Demographic questions in the first three positions, before anyone has a reason to care.

How the numbers are worked out

The model is deterministic and documented, so you can sanity-check it against your own completion data:

  1. Reading time is the question's word count divided by 200 words per minute.
  2. Answering time is an approximate published average per type: 45 seconds open-ended, 25 for a matrix or scale, 12 multiple choice, 10 short text, 8 for a simple closed question.
  3. Each audience gets an attention budget in minutes: 3 for the general public, 5 for customers, 7 for employees, 8 for specialists.
  4. An incentive stretches that budget by a quarter (small) or a half (meaningful).
  5. Every minute beyond the budget costs 6 percentage points of completion.
  6. Open-ended questions past the third cost 2 points each; rating grids past the second cost 1.5.
  7. The result is capped between 15% and 95% and reported as a range, because no survey completes at 100% and a single number would imply precision this model does not have.

Where these numbers come from

The timings and thresholds are drawn from published survey-length and reading-speed research, not invented for this tool. They are industry averages, so treat them as a starting point and calibrate against your own completion data:

Worked examples: 20 questions vs 8

Here is the same customer feedback survey run through the estimator twice, once as originally drafted and once after cutting everything that would not change a decision. Same audience (customers), no incentive either time. The numbers below come straight out of the tool.

Before

The 20-question draft

11:30 · 38-48% completion

Ten open-ended questions, five rating grids, and name, age, and income asked first. Cumulative time passes the 5-minute budget at question 11, so half the survey sits in territory most respondents never reach.

After

The 8-question rewrite

4:33 · 90-95% completion

Two scales kept, three of the rating grids dropped, seven open-ended questions folded into one. Job title and company size moved to the end. The survey now finishes inside the attention budget, so no question is flagged as a drop-off point.

Twelve questions removed, seven minutes saved, and roughly twice the estimated completion. Nothing about the rewrite is clever: it cuts the questions nobody had a plan for and moves demographics to where they belong.

Built for every team that sends surveys

Anyone who has watched a completion rate collapse knows length is usually the culprit, but rarely which question did the damage. This tool names it before you send. It pairs well with an adaptive AI survey platform when the answers you collect are sensitive.

Researchers

Check questionnaire burden before fieldwork, not after the first wave comes back thin.

Product teams

See what a feedback survey actually costs the user before it ships in the product.

Marketing & CX

Defend a shorter survey with a number rather than an opinion.

HR & people teams

Keep engagement surveys inside the time you promised employees they would take.

Academics & students

Estimate the completion time your ethics submission asks you to state.

Agencies

Show a client the cost of every extra question they want to add.

Shortened your survey? Now send it somewhere private.

BlockSurvey is an adaptive AI survey platform, encrypted end to end, so the responses you gather are never sold, mined, or used to train models.

More free AI tools

Explore more

Frequently asked questions

What is the survey drop-off estimator and what does it do?

It estimates how long your survey takes to finish and how many respondents are likely to abandon it before the end. Paste your questions one per line, pick your audience and incentive, and the tool classifies every question by type, adds up reading and answering time, and returns an estimated completion time, an estimated completion-rate range, and a list of the specific questions most likely to cause drop-off. It is a statistical estimator, not an AI model: the same input always produces the same result.

How accurate are the results?

This is an estimate built from published survey-length averages, not a guarantee. The timings come from average answer times for each question type, so a survey with unusually simple or unusually demanding questions will differ from the estimate. Real drop-off also depends on things this tool cannot see: how relevant the topic is to your audience, how you worded the invitation, the incentive you offer, and the channel you send it through. Treat the numbers as a fast structured starting point and validate them against your own completion data.

What happens to the data I enter?

Nothing is sent anywhere. This tool makes no API calls and does no AI processing: your questions are classified and timed locally in your browser, and neither the questions nor the report ever reach a server. That is why no sign-up is required and why you can safely paste a survey you have not launched yet.

Is it free? Do I need an account?

It is completely free and needs no account. There is no sign-up, no email wall, and no watermark. You get the full estimate on screen and can download it as an editable Word file or a PDF.

How is the completion time calculated?

Reading time plus answering time, summed across every question. Reading time is the question's word count divided by a reading speed of 200 words per minute. Answering time is an approximate published average for the question type: 45 seconds for open-ended, 25 for a matrix or scale, 12 for multiple choice, 10 for short text, and 8 for a simple closed question. The tool infers the type from wording cues in each line, so a question phrased ambiguously may be timed as the wrong type.

What causes survey drop-off?

Length is the biggest single cause, but the order and shape of your questions matter almost as much. Respondents abandon when the survey runs past the time they expected to spend, when several open-ended questions arrive in a row, when a long matrix grid appears on screen, when a single question is too long to read comfortably, and when demographic questions come first, before any reason to care has been established. This tool flags each of those patterns and tells you where in your survey it occurs.

How long should a survey be?

Under five minutes for most audiences, and under three for a general-public panel with no incentive. People answering about their own work tolerate more, and a meaningful incentive stretches that further, but neither removes length as a problem. A practical rule: decide what you will actually do with each answer, and cut every question that fails the test. Eight focused questions almost always beat twenty broad ones.
Scripts are blocked. This site won’t work properly. If you’re using Brave, click the Shields icon and turn off Block scripts. Otherwise disable your ad blocker for this site.