Skip to main content

Wisdom Rules Examples

Example rules to shape Wisdom's responses to your workflow.

Team Enterpret avatar
Written by Team Enterpret
Updated this week

Rules help you customise Wisdom to work exactly the way you need it to. By creating rules, you can ensure Wisdom consistently follows your preferences and provides answers tailored to your workflow.


Shape how Wisdom responds

Always provide verbatim quotes

Back up insights with actual customer language.

Whenever I ask a question I always want the answer to include multiple actual verbatim quotes my users


Planning mode

Get Wisdom to ask clarifying questions before running queries.

Always ask follow up questions for best clarification of the questions so that you can provide the best answer


Rank where possible

Get ranked lists with numbers, not just themes.

e.g., top 10 features ranked by number of feedback records within a given timeframe


Bugs vs. features

Separate what's broken from what's missing.

Where relevant, always provide an assessment of bugs vs. feature requests to instruct what needs fixing vs. what needs building


Add business data to every answer

Financial Data Inclusion

Surface revenue data alongside feedback to prioritise high-value customers.

When metadata is available on feedback, user or account data, always provide ARR, Contract Value, Spend, Revenue growth, pipeline efficiency, customer value, retention, growth rate, MRR, deal size, win rate in addition to insights when available


Geographic Segmentation

Understand how customer expectations differ by region.

Always consider the location country and compare the requested metadata with available geographic data. Provide me how customer feedback expectations differ depending on geography.


Competitor Analysis

See where you stand against competitors on every feature request.

Always compare feature requests to our 2 top competitors in the market & show me where our company stands.


Reduce support costs

Cost Per Contact

Attach a dollar amount to every support interaction.

When analysing Phone Calls, Human Agent Calls, Care Center Calls, always state the full cost of mutually exclusive and collectively exhaustive, non-duplicated cost per interaction at #$$$ per call.


Self-Service Recommendation

Turn support complaints into self-service opportunities.

When generating insights, always provide self-service recommendations from complaints or help requests, always identify the user's underlying goal, propose the most relevant help-center or in-product guidance first, and ensure the recommendation reduces future support volume by addressing the root cause, not just the symptom.

Did this answer your question?