Comparing themes between tables

What cross-table comparison is for

When you facilitate multiple tables, the hardest part is deciding whether a point is local or shared. Cross-table comparison gives you that answer if you do it carefully.

The goal is not to force all tables into one narrative. The goal is to detect:

• patterns repeated across contexts
• differences that are meaningful, not random
• outliers that may signal risk or innovation

Start with context, not counts

Before counting themes, write down what differs between tables:

• participant roles
• table size
• prompt interpretation
• time spent on each question

Without this context, a "difference" may only reflect who was in the group, not a true disagreement in needs.

The comparison workflow I recommend

1. Extract 6-10 themes from all table summaries.
2. Define each theme in one sentence to avoid coding drift.
3. Build a table matrix: rows are tables, columns are themes.
4. Score evidence per cell:

• 0 not present
• 1 mentioned briefly
• 2 explained with concrete examples

1. Add one transcript anchor for each 2.
2. Review the matrix for shared themes, splits, and outliers.

You can do this quickly in a spreadsheet during a debrief if your theme definitions are tight.

Scenario: same topic, different table realities

You run a co-creation session on customer onboarding. One table has mostly sales leads, another has support staff, another has product managers.

In RoomRadar summaries:

• sales tables mention expectation-setting
• support tables mention repeated setup errors
• product tables mention unclear ownership in the journey

This can look like disagreement. Often it is the same system problem viewed from different operational positions.

A better interpretation than "teams disagree":

Different functions are seeing different failure points in the same onboarding chain.

That framing helps you design cross-functional fixes instead of choosing one table as "correct."

How to write a useful comparison summary

Weak summary:

Most tables talked about onboarding and some tables talked about pricing.

Useful summary:

Onboarding clarity appeared in 5 of 6 tables, with strong evidence in sales and support groups.
Pricing confusion appeared in 2 tables, both with enterprise-focused participants.

This version tells stakeholders where to act broadly and where to investigate segment-specific issues.

Common pitfalls

Pitfall 1: over-valuing majority vote

If four tables mention a minor pain point and two tables mention a severe blocker, the severe blocker may still deserve priority.

Tip: score impact separately from frequency.

Pitfall 2: inconsistent theme labels

If one analyst labels comments as "ownership" and another as "handoff," you can split one theme into two by accident.

Tip: write short theme definitions before scoring.

Pitfall 3: mixing issue consensus with solution consensus

Tables may agree on the problem but disagree on the fix.

Tip: keep separate columns for "problem" and "proposed solution."

Troubleshooting contradictory patterns

Sometimes the matrix shows conflicting signals. Use this check:

1. Re-open transcript snippets for the conflicting cells.
2. Ask whether the groups discussed the same scenario.
3. Check if one table was answering a different question.
4. Re-code with narrower theme definitions.

If conflict remains, report it directly instead of smoothing it out.

Example language:

Tables aligned on the bottleneck but disagreed on whether process or tooling is the first fix.

That sentence is honest and decision-useful.

Facilitator tips for better comparisons

• During setup, keep prompts consistent across tables so comparison is fair.
• During synthesis, assign one person to defend minority viewpoints.
• During reporting, always include one "shared" and one "divergent" finding.
• For follow-up planning, separate quick wins from unresolved disagreements.

These habits reduce rework after the workshop because your stakeholders can see what is settled and what is still open.

A practical output template

Shared themes (high confidence):
Table-specific themes (investigate):
Outliers (monitor):
Likely reasons for differences:
Decision implications:

If table summaries are still unstable, start with [How RoomRadar group summaries work](/guides/analysis/understanding-group-summaries). When contradictions persist, pair this method with [Spotting consensus and disagreement](/guides/analysis/spotting-consensus-and-disagreement).

Related guides

• [Building a workshop report](/guides/analysis/building-a-workshop-report)
• [Extracting insights from transcripts](/guides/analysis/extracting-insights-from-transcripts)
• [Identifying follow-up ideas](/guides/analysis/identifying-follow-up-ideas)
• [Measuring participation in discussions](/guides/analysis/measuring-participation-in-discussions)
• [Aligning tables on shared definitions](/guides/facilitation/aligning-tables-on-definitions)