Brand Core · Brand Perception · Module

Measuring perception

How what people think is actually established — with the method’s limits recorded alongside the result, because every method here has serious ones.

The idea

How it works

Every method used here is biased in a known direction. Surveys collect considered answers to questions people had not thought about; social sentiment collects the views of the small minority who post; interviews collect what people will say to a stranger. Using one method is choosing one bias.

The instrument has to hold still. Perception moves slowly, which means the trend is the point — and a reworded question or a changed sample makes the trend measure the instrument instead.

Working with it

In practice

  1. 01

    Use at least two methods

    One quantitative for the trend and one qualitative for what the trend means. Neither answers the other’s question.

  2. 02

    Ask openly before asking on scales

    A scale measures what you already suspected. Open questions are where the surprises are.

  3. 03

    Hold the instrument still between waves

    Wording, sample and timing all fixed. Improving the survey destroys the comparison.

  4. 04

    Record who was not asked

    Non-customers and lapsed customers are the groups most often missing and most often the ones with the finding.

One level in

The components of measuring perception

A component is something that exists afterwards which did not exist before — a deliverable or a mechanism, not an intention.

  1. The instrument

    What is asked, of whom, how often, held constant between waves.

    Learn
  2. The reading

    The results, with the sample and response rate that qualify them.

    Learn
  3. The limits

    What the method cannot see, and who was not asked.

    Learn

Perception moves slowly, so the trend is the point — and improving the survey destroys the trend.