Before setting up an experience, it’s important to choose the appropriate logic strategy and outcome structure. This article provides strategic guidance to help you determine whether Matching or Scoring logic is right for your goals, and whether your outcomes should be fixed or dynamic. This decision impacts how personalized, scalable, and maintainable your experience will be.

Start by asking these questions:

How large is your product set?
How frequently will products be rotated in or out?
Are the recommendations fixed or dynamic?
Will the results be hosted within the quiz or on your website?
How many products do you want to recommend to each user?
How many unique outcomes are needed?

Before you begin

Before selecting a logic strategy, review your experience goals, audience needs, and catalog structure. Identify key inputs, including:

Product catalog size.
Number of questions.
Type and number of outcomes.
Personalization and update frequency.
Result location and hosting preferences.

Determine your logic strategy

Experiences by Jebbit supports multiple logic types, each suited to different use cases. Filtering logic dynamically adjusts which products display based on real-time answers mapped to product attributes. Matching logic assigns point values to answers to reach a predefined outcome. Filtering is ideal for dynamic catalogs; matching suits curated product or personality quizzes.

Logic types and outcome creation methods

Logic Type	Description
Matching	Point-based logic that connects users to a prebuilt outcome.
Filtering	Filters a product feed in real-time based on user responses.
Scoring	Assigns scores for correct answers and displays outcomes based on performance.

Ways to create outcomes	Description
In-platform	Use Matching or Scoring logic created entirely within the Jebbit interface.
Dynamic Product Feed	Use Filtering logic based on a CSV product feed with mapped attributes.

Standard outcomes

Unlike Dynamic Product Feed (which requires more lead time to configure), standard outcomes are recommended for first-time experience creation. These refer to matching and scoring outcomes below.

Matching logic

Matching logic uses a point-based system to connect users to a single, prebuilt outcome. Each answer is assigned a value for one or more outcomes, and the one with the highest cumulative score is shown. We recommend matching logic for the following:

Product matches or personality quizzes.
Small, curated product sets (< 25 items).
Static recommendations that don’t change frequently.
Simple or editorial-style logic.

Best practices:

Use import/export to streamline large sets of outcomes.
Assign different point weights based on question importance.
Use the Outcome Validator for linear experiences to validate logic.
Avoid excessive branching; only use outcome splits when outcomes are limited.
Use negative point values sparingly.

Learn how to create matching outcomes.

Scoring logic

Scoring logic is designed for knowledge-based quizzes where outcomes reflect performance. Questions have correct answers, and the total number of correct responses determines the final result. We recommend scoring logic for the following.

Trivia quizzes or assessments.
Gamified learning or educational content.
Outcomes based on user knowledge, not preferences.

Best practices:

Ensure questions have clear, definitive answers.
Tie performance brackets to specific scoring thresholds.
Keep outcome messaging aligned with performance levels.

Learn how to create scoring outcomes.

Dynamic product feed (advanced)

For large or frequently changing catalogs, use a dynamic product feed. This filters products in real time based on user answers, using metadata to match. We recommend the DPF for the following:

Catalogs with > 25 products.
Frequent updates or inventory changes.
Multi-attribute filters (e.g., size, color, skin type).
Personalized shopping experiences.

Best practices:

Use a clean, well-structured CSV feed with clear columns.
Map each quiz question to a column in the feed.
Use commas to list multiple attribute values per product.
Prioritize filter order using fallback logic.
Test using specific combinations and validate results in Excel first.

Note: Access to features like the Dynamic Product Feed requires an Enterprise-level account. Learn how to create the DPF here.

Outcome structure

Fixed vs. dynamic outcomes

Fixed outcomes: Prebuilt result pages tied to specific selections. Best for static quizzes, brand stories, or simple product matchers.
Dynamic outcomes: Filter-based results pulled from a live feed. Best for eCommerce, large catalogs, or seasonal rotations.

Single vs. multiple outcomes

Single outcome: Presents one product or result per session. Ideal for focused product recommendations or personas.
Multiple outcomes: Shows a list of relevant options. Ideal for comparisons or multiple product suggestions.

Logic type by use case

Criteria	Use Standard Outcomes	Use Dynamic Product Feed
< 20 products	✔️	✔️
> 20 products	❌	✔️
Fixed, unchanging routines	✔️	✔️
Dynamic, filtered product lists	❌	✔️
Rare product updates	✔️	❌
Frequent product updates	❌	✔️
Simple matching logic	✔️	✔️
Complex or multi-attribute logic	❌	✔️

Optimize your logic post-launch

Once your experience is live, continue to refine your logic by reviewing:

Drop-off points: Identify where users abandon the experience and consider simplifying questions or logic flow.
Popular vs. unused outcomes: Review which outcomes are most and least selected to refine product mapping or improve balance.
User behavior insights: Leverage analytics to understand how responses trend and refine logic weights or scoring accordingly.
Catalog updates: For dynamic feeds, regularly update your CSV to reflect product availability, seasonality, or trending items.