Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Glossary term

Survey Wave

One round of fieldwork in a repeated survey or tracker, drawn, worded and coded the same way as earlier rounds so the results can be compared over time.
AnalyticsconceptUpdated

Definition

A survey wave is one round of fieldwork in a repeated survey or tracker: a sample drawn, invitations sent, answers collected over a fieldwork period, open answers coded and the result read and stored. The word marks the survey as part of a series. The guide to research design and the brief sets out the choice behind it: a one-off study answers the question now, while repeated waves reveal direction but cost more and need an instrument that stays stable.

What makes a round a wave rather than a new survey is that it is built to be compared. Its population, draw, wording, coding and reading rules are inherited from the waves before it, and its result is read against theirs.

What stays fixed between waves, and what breaks the series

The customer survey wave SOP takes a single wave from the date it falls due to the record that is filed, and most of its steps exist to keep the wave identical to the last one where it matters:

  • The draw. The same population, cleaning rules, contact rule and strata, repeated from the written record of the last draw. Drawing a wave by another method measures something else, so it is annotated rather than compared silently.
  • The questionnaire. Wording, scales, question order and the invitation stay as frozen. A deliberate change starts a new series with its own baseline.
  • The coding. The last wave's codeframe is applied again, with new codes added and dated rather than old ones redefined.
  • The reading. Each wave is judged by rules agreed in advance, such as limits on a control chart, so that ordinary noise is not explained as a change.
  • The timing. Relationship waves avoid the weeks of a price rise or an outage, unless the reaction to it is what is being measured; a wave that would fall in one is moved and the move noted.

The survey launch checklist is the gate before each wave, run again every time rather than once for the series. On a tracker, it treats the freeze as most important once the baseline exists: any later change needs a reason in the log and the wave from which it applies.

In practice: baseline waves, trackers and the research rhythm

Repeated surveys appear in more places than the research team's calendar. In the customer research programme playbook, conversations with customers run on a standing rhythm and a survey wave is added only when a decision depends on how widely something holds; its questions are built from what the conversations turned up, and the draw is written down for the next wave to repeat.

Measurement plans use a first wave as a baseline. The guide to PR outputs, outcomes and impact asks for the current position on each measure before any target is set, which for awareness means at least one survey wave. Brand trackers work the same way, with the same questions put to a fresh sample of category buyers in every round.

Illustrative example — not a real result

A climbing centre's satisfaction drop that was a different wave

Wynnaor Climbing, an indoor climbing centre with about 1,900 members, surveys a random 500 of them each quarter and reports the share who say they are satisfied. Four waves had come in between 77% and 80%. The latest showed 64%, and the owners asked what had gone wrong on the walls. The survey owner checked the draw before any explanation: a new assistant had sent the survey to the 430 members who had booked a session that month, taken from the booking system, instead of a random draw from the full membership list. That is a different population, so the result could not sit on the same line. The wave was annotated and kept off the trend, and the next one, drawn by the recorded rule, came back at 78%.

A Brand Tracker is a series of waves with its instrument held still, so that movement between rounds can be read as the market's rather than the survey's. The Sample Size for each wave, and for each stratum within it, starts from the business's own earlier waves, which show how many answers each group actually returns. A Codeframe is frozen with a date so that later waves can apply it and theme counts stay comparable. A Soft Launch releases the survey to a small share of the sample first and pauses to check the data; a fault it finds is fixed before the full send, not left for the next wave.

Common confusions

  • Any batch of survey sends is a wave. In research a wave is one round of a repeated survey, not a batch of invitations within one round.
  • A small improvement to a question is harmless. Any change to wording, scale, order or invitation creates a new measurement; the series restarts, on purpose and with its reason recorded.
  • Every wave needs a fresh sample design. The draw is repeated from the record. Redesigning it each time compares different groups of people.
  • A movement between waves is a change. It may be noise. The rules agreed before the wave decide whether it counts as a signal, before anyone offers a reason.
  • One wave shows a trend. A single wave answers the question for now. Direction needs repeated waves run the same way.

More terms to look up

Where this term is used

Related