What is LOI, and how does length of interview shape a quantitative study?
Length of interview looks like a descriptive statistic. It is actually a parameter you set before fielding, and it goes on to decide what your study costs, how fast it fields, and which respondents get thrown out.
In brief
LOI, or length of interview, is the average time a respondent takes to complete a survey from first question to submission. In a quantitative study it functions as a fielding parameter rather than a description: it determines the incentive and therefore the cost per complete, it affects how quickly the study fields, and it sets the benchmark against which speeders are identified and removed. Most online consumer studies run a 10 to 12 minute LOI.
What length of interview means in market research
LOI is the average clock time from a respondent's first question to submission, measured in minutes. It covers only respondents who completed. Screen-outs and drop-offs are excluded, which matters because a study with an aggressive screener can show a comfortable LOI while most people who encountered it never finished.
Two distinctions worth keeping straight. LOI is not question count, because question types differ in effort and a twelve-question study can run longer than a thirty-question one. And LOI is not fieldwork duration, which is how many days the study takes to reach its target sample. The two are related, since longer LOI slows fielding, but they measure different things.
The term carries over from telephone and face-to-face interviewing, where an interviewer's time was the cost being managed. In online research the interviewer is gone but the metric survived, because respondent time is still the thing being purchased.
Estimated LOI vs actual LOI
Every study has two LOI figures and the gap between them causes most LOI-related problems.
Estimated LOI is what you declare before fielding. It sets the incentive, prices the sample, and gets communicated to respondents. Actual LOI is what the completed interviews measured.
When the estimate is too low, respondents were paid for a shorter survey than they took, drop-off rises partway through, and the panel relationship degrades across future studies. One documented case involved a 65-minute estimated LOI where the survey in field ran well past 80 minutes, and participation and conversion rates collapsed as a result.
When the estimate is too high, you overpaid for every complete, and, more damagingly, a large share of attentive respondents look like speeders against a benchmark that was never realistic.
Neither error announces itself. Both show up as a quiet cost, one in money and one in discarded data.
How LOI affects fieldwork cost and feasibility
LOI is a primary driver of cost per complete because it determines what a respondent has to be compensated for. A 20-minute study does not cost twice a 10-minute study, it usually costs more than twice, because willingness to participate falls faster than length rises.
The feasibility effect compounds it. Pushing past 15 minutes raises drop rates within the survey, which in turn increases sample usage and cost and extends project timelines. Every respondent who abandons at minute twelve consumed panel capacity and returned nothing, so you pay for reach you cannot use.
Where LOI interacts with incidence rate, the effect multiplies. A long survey targeting a low-incidence cohort is the most expensive combination in commissioned research, because you are screening a large number of people to find each qualifier and then losing a meaningful share of them to length. Check that combination before quoting a timeline.
How LOI is used to detect speeders
Automated quality scoring uses estimated LOI as the reference point for identifying respondents who could not plausibly have read the questions. The common industry threshold flags anyone completing in 30% or less of the estimated LOI as a potential speeder.
That makes the estimate consequential in a way most people do not anticipate. Declare 15 minutes for a survey that genuinely takes 6, and the threshold sits at 4.5 minutes, which a fast, attentive respondent can easily beat. Those respondents get penalised or disqualified for answering efficiently. Declare 6 minutes for a survey that takes 15 and the threshold drops so low that no one triggers it, which means genuine speeders pass through into your data.
Since every disqualification removes a respondent from a cohort base, speeder detection and sample planning are the same decision. Set the LOI honestly, then size the sample above target so the removals do not push a cut below its planned base.
How LOI affects drop-off and data quality
Long surveys damage data in two ways, and only one of them is visible.
The visible one is drop-off. Respondents abandon partway through, and because the people who abandon are disproportionately those with less time or less interest in the category, what remains is not a random subset of your audience.
The invisible one is worse. Respondents who stay start conserving effort. Open-text answers shorten, so the last open question in a long survey returns single words while the first returned paragraphs. Grid responses flatten towards a single column, and a flat grid reads as a considered attitude rather than as fatigue. Nothing in the dataset marks which responses came from the second half.
This is why LOI belongs to design rather than to fieldwork. By the time you are looking at actual LOI, the damage is already in the data.
Typical LOI by study type
| Study type | Typical LOI | Notes |
|---|---|---|
| Pulse or single-issue check | 2 to 3 minutes | Highest completion rates, lowest cost per complete |
| Price sensitivity | 5 to 8 minutes | Short by necessity, since the exercise is cognitively heavy |
| Packaging or name test | 6 to 8 minutes | Visual stimulus adds time per concept |
| Concept test | 8 to 10 minutes | Rises quickly with the number of concepts shown |
| Brand health or tracker | 10 to 12 minutes | Kept short deliberately, since length compounds across waves |
| Category deep dive | 12 to 15 minutes | The practical ceiling for general consumer audiences |
| Specialist or B2B audiences | 15 to 25 minutes | Tolerable only where the audience has professional reason to engage |
Mode matters as much as study type. Mobile respondents complete more slowly than desktop respondents on identical questionnaires, and audiences answering in a second language slower still. A 10-minute LOI estimated on desktop testing will overrun on a majority-mobile sample.
How to estimate LOI accurately
Build the estimate from question types rather than question count. Single-choice questions cost around 10 seconds, grids around 10 seconds per row, ranking 30 to 45 seconds, and open text 45 to 60 seconds. Sum those, then add the screener and any loading time between pages.
Account for piping, since a piped question is one question in the builder and up to six in the interview. Estimate on average selections from the source question, not on the minimum.
Then validate against a pilot. Field the questionnaire to a small group and measure both how long it actually took and whether respondents could still answer well by the end. Pilot timing is the only estimate worth declaring, because it is the only one measured on the actual instrument with real respondents on real devices.
Finally, check actual against estimated after fielding and carry the correction into the next study. Most research teams have a consistent directional bias in their estimates, and the only way to find yours is to compare.
LOI checklist
Estimate built from question types, not question count. Piping accounted for on average selections. Mobile and language effects factored in. Estimate validated against pilot timing before declaring it. Speeder threshold understood as a consequence of the estimate. Sample sized above target to absorb disqualifications. Actual LOI compared against estimate after fielding.
Why Flickly?
Flickly lets you set an expected length of interview, then measures actual completion time on every response and flags speeders against that benchmark automatically.
Frequently asked questions
Run your next study with verified Indian consumers
Design the study, reach the right respondents, and get decision-ready insights in 72 hours.