Why You Should Think Twice Before Updating Survey Questions in Tracking Studies

This is the fourth and final blog in our mini-series on language in market research, exploring how the words we choose shape the quality, clarity, and impact of market research.

The previous articles in this series focused on the specific words we use in questionnaires to ensure that language is interpreted clearly.

This time, we’ll look at the impact of making changes – whether to question wording or to the positioning of questions. This is particularly important for tracking projects where the same questions are repeated over time.

Making changes to questionnaires

A common temptation in tracking studies is to “improve” question wording between waves. While this can sometimes be justified, it’s important to understand the implications.

For example, a question may originally be asked as:

“Overall, how satisfied are you with your account manager?”

In a later wave, this might be changed to:

“Overall, how satisfied are you with the support you receive from your account manager?”

Although this change may seem minor, the meaning has been narrowed to focus specifically on support rather than overall satisfaction. Any changes in scores may therefore be driven by the updated wording rather than a genuine shift in sentiment.

Another common example is a question such as:

“How easy is it to use our platform?”

During the presentation, it may not be clear what respondents understood by “ease of use.” You could argue that the question is too broad and open to interpretation. As a result, the decision might be made to update it to:

“How easy is it to use our platform (for example, navigating dashboards, generating reports, exporting data)?”

This version is clearer, but it asks respondents to focus on specific features of the platform. Because we’re prompting on particular aspects, responses may shift.

These updated questions may well be stronger and more informative moving forward – but any comparisons with the previous versions should be treated with caution.

Positioning of questions

It is generally recommended to ask general questions before specific ones. A good example is the placement of the classic NPS question in a customer satisfaction survey.

Opening a survey with the general:

“How likely are you to recommend brand X?”

is preferable to asking it at the end. At the start, respondents give a top‑of‑mind answer before being “primed” or influenced by questions on specific aspects of the relationship – for example, customer service, delivery, or product performance.

Moving the NPS question from the start to the end could result in different scores. Again, we wouldn’t know whether the change reflected a real shift in sentiment or simply the effect of question positioning. This makes fair comparison over time difficult.

In summary

Avoid changing wording or repositioning questions in tracking studies unless absolutely necessary. Even small adjustments can influence responses and make comparisons unreliable. If changes do need to be made, be sure to document exactly what has been updated so the analysis and interpretation can account for it later on.

To discuss how our tailored insights programs can help solve your specific business challenges, get in touch and one of the team will be happy to help.