Marketing research interpretation workshop / 5 MIN READ
Put the tested attribute ranges beside conjoint importance scores
Explain how the feature levels used in a conjoint study shape the relative importance reported afterward.
AI-assisted research and analysis for the SNS Creative Fieldbook
Sources checked 2026-11-07
An importance score describes the influence an attribute can express across the levels included in that study. It is not a permanent customer priority detached from the alternatives the researcher chose.
What you will leave with: An importance-chart review with the original levels, interpretation boundary and a decision-relevant scenario comparison.
Use this when
- A conjoint report says price, brand or a feature accounts for a certain share of importance.
- Teams want to compare attribute-importance percentages from two differently designed studies.
This is not a full conjoint-estimation tutorial, a causal sales forecast or a claim that all software calculates importance identically.
Recover the attributes and their exact levels
Place the study’s original level definitions beside the chart. Show the cheapest and most expensive tested prices, the service times offered and the product configurations actually presented. Record whether the descriptions were realistic and understood. Without this context, a large importance percentage can be mistaken for an enduring statement about what customers value most.
Sawtooth Software explains that importance scores summarize influence over the included range of levels and that changing that range can change the score. Its documented approach expresses attribute ranges proportionally. This supports reading the chart within its design, not treating the normalized percentages as universal customer constants.
Review the design choices behind the apparent priority
Ask why each range was selected. A narrow price variation and a dramatic service-level difference create a different choice task from a large price spread and nearly identical service. Both may be legitimate if they match different business questions. The problem occurs when the resulting importance chart is reported without its decision context.
Check that the tested levels cover the feasible alternatives the business actually needs to compare. Do not widen a range after seeing results merely to make a favored attribute appear important. If the original study missed a consequential range, document that limitation and ask the researcher what additional work is needed. The report should preserve the study that happened, not a reconstructed version that tells a better story.
Compare studies only with a defensible bridge
When a later wave produces a different ranking, inspect the populations, attributes, ranges, wording and analysis method before calling it a preference shift. A new feature or expanded range changes the relative frame. Even an unchanged attribute can receive a different normalized share when another attribute changes.
Ask the analyst which comparisons remain valid and whether a common-range scenario, bridge study or separate presentation is appropriate. Do not manufacture comparability by simply rescaling charts to the same axis. Preserve genuine changes in customer response while distinguishing them from changes in what the study allowed people to trade off.
Move from global ranking to the actual decision
Identify the concrete alternatives under consideration: a specific service upgrade at a defined cost, a limited price increase or a change in package contents. Ask the analyst to evaluate those plausible scenarios using the model within its supported range. A high overall importance score does not tell the team which small operational change is worthwhile.
Combine the scenario with feasibility and economics outside the preference model. Record the tested tradeoff, uncertainty and reasons for the decision. Keep the importance chart as a summary of the research design rather than a spending formula. Future teams should be able to recover why one attribute looked prominent and which practical question the study was intended to answer.
Make the reasoning concrete.
A fictional service study tests prices within a narrow band but compares same-day support with a week-long wait. Support receives the largest importance score, and the team claims customers barely care about price.
Generalize the ranking
Opening: Present the result as a permanent priority for the whole market.
Hypothesis: The strategy sounds decisive.
Risk: The narrow price range limits what the study can say about wider price differences.
Interpret the tested tradeoff
Opening: Show both ranges and evaluate the feasible price-and-support options the business faces.
Hypothesis: The result answers a bounded decision.
Risk: A materially wider price question may require new evidence.
The team reports the importance within the tested ranges and uses a relevant scenario comparison. It avoids saying that price is unimportant in every buying situation.
Take the questions into your next brief.
Use these prompts in your own notes. Nothing is submitted from this page.
Levels
What exact alternatives did respondents evaluate?
Range rationale
Why were these minimum and maximum levels selected?
Comparison
What changed between studies besides customer responses?
Scenario
Which feasible tradeoff must the business decide now?
Boundary
Which conclusion would go beyond the tested range?
Watch for the wrong conclusion.
- Reading normalized importance as a fixed trait of the customer.
- Comparing studies with changed ranges as a simple preference trend.
- Choosing research levels to force a preferred attribute to lead.
- Using importance percentages as an automatic resource-allocation formula.
Your next action
Add the tested levels to one importance chart and replace its universal headline with a conclusion tied to those ranges.
Sources & scope
Workflows, worksheets and hypothetical examples are original editorial analysis. Platform documentation supports only identified technical principles. No measured business result or named human reviewer is claimed.
- Sawtooth Software: Importances →Primary research-software provider methodology guidance · Supports: The dependence of conjoint attribute importance on the included level ranges and the documented proportional range calculation. Interpretation workflow and examples are original editorial analysis.