Ranking Studio

🌟
This is a Premium tier feature powered by Loomi AI

What is Ranking Studio

Ranking Studio allows you to tailor the core Bloomreach ranking algorithm. This feature empowers you to:

Train the ranking algorithm leveraging Learning to rank (LTR) machine learning
View how various ranking signals influence product ranking
Modify ranking signal weights
Add new signals (like margin, offline revenue, forecasts, and reviews) to the algorithm

With these, you can curate highly customized ranking algorithm(s) for various business use cases and run multiple algorithms for various subsections of your site experience.

LTR training framework overview

LTR is a machine learning approach that allows the customization and personalization of the Bloomreach ranking algorithm by integrating new signals (such as margin, offline revenue, forecasts, and reviews) or adjusting the weights of existing signals.

Why is this feature valuable

Ranking Studio offers key benefits:

Drive Business KPIs: The flexibility to customize the ranking algorithm lets you drive metrics such as average order value (AOV), conversion rates, and revenue per visitor (RPV). The ability to AB test and experiment with different signals allows for KPI optimization.
Build, Test, and Innovate: Your in-house data science or engineering teams can now fully control Bloomreach’s ranking algorithm for Search and Category pages. This hands-on approach empowers you to experiment, innovate, and tailor the discovery experience to your business needs.

Business use cases

Here are some of the many e-commerce use cases of this feature:

Enhance the ranking experience to account for higher ratings and positive reviews in search results without affecting relevance.
Prioritize sitewide revenue over page-level revenue.
Prioritize text relevance scores over performance scores to double down on relevancy.
Use a curated score that you pass as an additional signal (for example, certain customers compute forecasted sales of a product as a score) to be considered as the main factor in determining the search ranking.

Feature access

All customization capabilities will only be available to Premium tier customers.

Ranking Studio workflow

On a high level, the workflow follows the order:

1. Dashboard end

Use the dashboard interface to create algorithms, add new signals, train the algorithm, customize signal weights, and preview the algorithm results. You can also apply the algorithm to different query scopes and set up AB tests. Check the dashboard guide.

2. API end

After customizing the algorithm on the dashboard, the algorithm will be in the Available state. Use the API parameter to apply the algorithm for a site experience and AB test with an external AB testing tool. Check the API guide.

FAQs

If I change the search ranking behavior, how does it impact other Discovery features?
This table summarises the impact on other Discovery features:

Feature	Expected Impact
Content Search	No impact
Dashboard Synonyms	No impact
Facet Ranking	No impact
Lookups	No impact
Merchandising operations	Merchandising operations like boost, bury, slots, add to recall, and so on will continue to be applied as they do today. However, you might want to review the impact of the existing rules in case there’s a significant deviation in the signal weights of the curated algorithm.
Query Relaxation	No impact
Ranking Diagnostics	While you will see the new experience, there is currently no UI indication that the new algorithm is being applied.
Recommendations and Pathways	The ranking changes will be reflected in the Recommendations and Pathways results.
Redirects	No impact. Redirects will continue to function the same.
Relevance-by Segment	The ranking is optimized based on the newly added signals and the existing segment signals.
Real-Time Segments	The ranking is optimized based on the newly added signals and the existing segment signals.
SEO	No impact
Keyword Search Precision modes	No impact
SKU Select	No impact
Spell Correction	No impact

How does Ranking Studio differ from Algo Weight Customization?
- Ranking Studio lets you define the weight of each signal relative to each other, whereas Algo Weight Customization allows you only to add a boost on top of the signal weight that the algorithm has learned.
- Ranking Studio allows you to essentially create your own algorithm by customizing one or all signals that Bloomreach utilizes to power search and category ranking.
- Ranking Studio allows you to add additional signals to the mix to create your own algorithm.
Can I AB test the new ranking experience?
Yes, you can AB test the algorithm using the dashboard UI and frontend API parameter. Visit the linked guides to see the AB testing process.
Are there any changes in the Search and Category response format?
There is no change in the Search and Category API response format.
Does the feature require changes in feed/pixel?
No additional integration effort is required on your end.