Custom Match Questions
Navigate to the 'Custom Match Questions' section and click on '+ Add Custom Question'. You can add up to 7 answer options per question. All answer options must be pre-filled – respondents cannot add free text.
📝 The number of custom questions you have access to depends on your pricing plan.
Additionally, you can further define how you'd like to set up your question via the Answer Settings.
Allow for multiple selections: Choose this setting if you want participants to be able to choose more than one answer option.
These answers represent a scale (ordered): Turn these on if the list is ordered by most similar to least similar. For example, an Executive is more similar to a VP than they are to a Director. Or, 1 year of experience is more similar to 2 years of experience than 5 years, so we want to match the 1 ↔️ 2 and the Executive ↔️ VP.
Match like-for-unlike: By default, Orbiit will match for like ↔️ like, if you would like to match like ↔️ unlike (e.g. mentor to mentee or technical to non-technical), select 'Match like-for-unlike'.
Do not use this question for matching: Use this setting to collect data on your members that you do not necessarily want to match on (i.e. last round of fundraising, likelihood to renew, product preferences).
Best Practices
1. Define Clear Objectives for Matching
Clarify Matching Goals: Determine what the matching aims to achieve—whether it's skill-based networking, shared interests, career stages, or mentorship opportunities. Defining this objective helps ensure your questions are purposeful.
Identify Key Matching Criteria: Based on your goal, identify the primary factors that should drive matching. For instance, professional interests, project preferences, or personal growth goals may be important depending on the community’s focus.
2. Keep Questions Relevant and Focused
Focus on Match-Relevant Questions: Avoid general questions that do not contribute directly to matching. Every question should provide information that enhances the AI’s ability to create meaningful matches.
Limit the Number of Questions: Typically, 3-4 questions per matching cycle are sufficient. This avoids overwhelming participants and helps maintain high response rates and relevance.
3. Balance Between Specificity and Flexibility
Provide Clear Answer Choices: Ambiguity in response options can lead to mismatches. Offer clear, distinct options that allow members to express specific preferences without over-complicating the response choices.
Avoid Too Many Choices Per Question: Too many options can dilute commonalities and weaken the AI’s ability to identify matches. Limit choices to maintain focused matching, ideally with no more than 4-6 options per question.
4. Use Inclusive and Accessible Language
Avoid Jargon and Technical Terms: Use clear, community-friendly language to ensure all members, regardless of background, can easily understand and respond accurately.
Focus on Universal Relevance: Questions should appeal broadly to the community and avoid niche topics that may not resonate with the larger group unless that is a defined matching objective.
5. Rotate Topics Regularly
Keep Content Fresh: Rotate question topics monthly or quarterly to align with community interests and seasonal themes. This keeps members engaged and provides varied data for AI matching.
Highlight Emerging Interests: If the community evolves in its interests (e.g., trending topics, skills, or project types), adapt questions to reflect these new areas.
6. Regularly Review and Refine Questions
Monitor Matching Outcomes: Track the effectiveness of matching and gather feedback from community members to understand if connections are meeting their expectations.
Adapt Based on Feedback and Data: Use feedback to refine questions over time, adjusting language, adding relevant options, or retiring outdated topics to keep the matching process aligned with the community’s evolving needs.