Making Each Search Rewarding: How Ibotta Reworked Provide Discovery With Databricks

At Ibotta, our mission is to Make Each Buy Rewarding. Serving to our customers (whom we name Savers) discover and activate related affords by our direct-to-consumer (D2C) app, browser extension, and web site is a important a part of this mission. Our D2C platform helps hundreds of thousands of customers earn cashback from their on a regular basis purchases—whether or not they’re unlocking grocery offers, incomes bonus rewards, or planning their subsequent journey. By way of the Ibotta Efficiency Community (IPN), we additionally energy white-label cashback applications for a number of the greatest names in retail, together with Walmart and Greenback Common, serving to over 2,600 manufacturers attain greater than 200 million customers with digital affords throughout associate ecosystems.

Behind the scenes, our Knowledge and Machine Studying groups energy important experiences like fraud detection, provide suggestion engines, and search relevance to make the Saver journey customized and safe. As we proceed to scale, we want data-driven, clever techniques that assist each interplay at each touchpoint.

Throughout D2C and the IPN, search performs a pivotal position in engagement and must hold tempo with our enterprise scale, evolving provide content material, and altering Saver expectations.

On this publish we’ll stroll by how we considerably refined our D2C search expertise: from an bold hackathon venture to a sturdy manufacturing characteristic now benefiting hundreds of thousands of Savers.

We believed our search may higher sustain with our Savers

Person search conduct has developed from easy key phrases to incorporating pure language, misspellings, and conversational phrases. Fashionable search techniques should bridge the hole between what customers sort and what they really imply, deciphering context and relationships to ship related outcomes even when question phrases don’t precisely match the content material.

At Ibotta, our authentic homegrown search system, at instances, struggled to maintain tempo with the evolving expectations of our Savers and we acknowledged a possibility to refine it.

The important thing areas for alternative we noticed included:

• Enhancing semantic relevance: Specializing in understanding Saver intent over actual key phrase matches to attach them with the correct affords.
• Enhancing understanding: Deciphering the complete nuance and context of consumer queries to supply extra complete and really related outcomes.
• Rising flexibility: Extra quickly integrating new provide sorts and adapting to altering Saver search patterns to maintain our discovery expertise rewarding.
• Boosting discoverability: We wished extra strong instruments to make sure particular kinds of affords or key promotions have been persistently seen throughout a big selection of related search queries.
• Accelerating iteration and optimization: Enabling quicker, impactful enhancements to the search expertise by real-time changes and efficiency tuning.

We believed the system may higher hold tempo with altering provide content material, search behaviors, and evolving Saver expectations. We noticed alternatives to extend the worth for each our Savers and our model companions.

From hackathon to manufacturing: reimagining search with Databricks

Addressing the constraints of our legacy search system required a targeted effort. This initiative gained important momentum throughout an inner hackathon the place a cross-functional group, together with members from Knowledge, Engineering, Advertising and marketing Analytics, and Machine Studying, got here along with the concept to construct a contemporary, different search system utilizing Databricks Vector Search, which some members had discovered about on the Databricks Knowledge + AI Summit.

In simply three days, our group developed a working proof-of-concept that delivered semantically related search outcomes. Right here’s how we did it:

1. Collected provide content material from a number of sources in our Databricks catalog
2. Created a Vector Search endpoint and index with the Python SDK
3. Used pay-per-token embedding endpoints with 4 completely different fashions (BGE massive, GTE massive, GTE small, a multilingual open-source mannequin, and a Spanish-language-specific mannequin)
4. Linked all the pieces to our web site for a reside demo

The hackathon venture gained first place, generated sturdy inner buy-in and momentum to transition the prototype right into a manufacturing system. Over the course of some months, and with shut collaboration from the Databricks group, we reworked our prototype into a sturdy full-fledged manufacturing search system.

From proof of idea to manufacturing

Shifting the hackathon proof-of-concept to a production-ready system required cautious iteration and testing. This section was important not just for technical integration and efficiency tuning, but additionally for evaluating whether or not our anticipated system enhancements would translate into constructive modifications in Saver conduct and engagement. Given search’s important position and deep integration throughout inner techniques, we opted for the next method: we modified a key inner service that referred to as our authentic search system, changing these calls with requests directed to the Databricks Vector Search endpoint, whereas constructing in strong, swish fallbacks to the legacy system.

Most of our early work targeted on understanding:

Within the first month, we ran a check with a small share of our Savers which didn’t obtain the engagement outcomes we had hoped for. Engagement decreased, significantly amongst our most energetic Savers, indicated by a drop in clicks, unlocks (when Savers specific curiosity in a proposal), and activations.

Nevertheless, the Vector Search answer provided important advantages together with:

• Quicker response instances
• A less complicated psychological mannequin
• Larger flexibility in how we listed information
• New skills to regulate thresholds and alter embedding textual content

Happy with the system’s underlying technical efficiency, we noticed its higher flexibility as the important thing benefit wanted to iteratively enhance search end result high quality and overcome the disappointing engagement outcomes.

Constructing a semantic analysis framework

Following our preliminary check outcomes, relying solely on A/B testing for search iterations was clearly inefficient and impractical. The variety of variables influencing search high quality was immense—together with embedding fashions, textual content combos, hybrid search settings, Approximate Nearest Neighbors (ANN) thresholds, reranking choices, and lots of extra.

To navigate this complexity and speed up our progress, we determined to determine a sturdy analysis framework. This framework wanted to be uniquely tailor-made to our particular enterprise wants and able to predicting real-world consumer engagement from offline efficiency metrics.

Our framework was designed round an artificial analysis atmosphere that tracked over 50 on-line and offline metrics. Offline, we monitored commonplace info retrieval metrics like Imply Reciprocal Rank (MRR) and precision@okay to measure relevance. Crucially, this was paired with on-line real-world engagement indicators resembling provide unlocks and click-through charges. A key determination was implementing an LLM-as-a-judge. This allowed us to label information and assign high quality scores to each on-line query-result pairs and offline outputs. This method proved to be important for fast iteration based mostly on dependable metrics and amassing the labeled information obligatory for future mannequin fine-tuning.

Alongside the way in which, we leaned into a number of components of the Databricks Knowledge Intelligence Platform, together with:

• Mosaic AI Vector Search: Used to energy high-precision, semantically wealthy search outcomes for analysis checks.
• MLflow patterns and LLM-as-a-judge: Offered the patterns to judge mannequin outputs and implement our information labeling course of.
• Mannequin Serving Endpoints: Environment friendly deployment of fashions immediately from our catalog.
• AI Gateway: To safe and govern our entry to 3rd get together fashions by way of API.
• Unity Catalog: Ensured the group, administration, and governance of all datasets used throughout the analysis framework.

This strong framework dramatically elevated our iteration velocity and confidence. We performed over 30 distinct iterations, systematically testing main variable modifications in our Vector Search answer, together with:

• Totally different embedding fashions (foundational, open-weights, and third get together by way of API)
• Varied textual content combos to feed into the fashions
• Totally different question modes (ANN vs Hybrid)
• Testing completely different columns for hybrid textual content search
• Adjusting thresholds for vector similarity
• Experimenting with separate indexes for various provide sorts

The analysis framework reworked our improvement course of, permitting us to make data-driven choices quickly and validate potential enhancements with excessive confidence earlier than exposing them to customers.

The seek for one of the best off-the-shelf mannequin

Following the preliminary broad check that confirmed disappointing engagement outcomes, we shifted our focus to exploring the efficiency of particular fashions recognized as promising throughout our offline analysis. We chosen two third-party embedding fashions for manufacturing testing, accessed securely by AI Gateway. We performed short-term, iterative checks in manufacturing (lasting just a few days) with these fashions.

Happy with the preliminary outcomes, we proceeded to run an extended, extra complete manufacturing check evaluating our main third-party mannequin and its optimized configuration in opposition to the legacy system. This check yielded blended outcomes. Whereas we noticed total enhancements in engagement metrics and efficiently eradicated the damaging impacts seen beforehand, these positive aspects have been modest—principally single-digit share will increase. These incremental advantages weren’t compelling sufficient to completely justify an entire alternative of our present search expertise.

Extra troubling, nonetheless, was the perception gained from our granular evaluation: whereas efficiency considerably improved for sure search queries, others noticed worse outcomes in comparison with our legacy answer. This inconsistency offered a big architectural dilemma. We confronted the unappealing selection of implementing a posh traffic-splitting system to route queries based mostly on predicted efficiency—an method that may require sustaining two distinct search experiences and introduce a brand new, advanced layer of rule-based routing administration—or accepting the constraints.

This was a important juncture. Whereas we had seen sufficient promise to maintain going, we wanted extra important enhancements to justify totally changing our homegrown search system. This led us to start fine-tuning.

Advantageous-tuning: customizing mannequin conduct

Whereas the third-party embedding fashions explored beforehand confirmed technical promise and modest enhancements in engagement, additionally they offered important limitations that have been unacceptable for a long-term answer at Ibotta. These included:

1. Incapability to coach embedding fashions on our proprietary provide catalog
2. Problem evolving fashions alongside enterprise and content material modifications
3. Uncertainty relating to long-term API availability from exterior suppliers
4. The necessity to set up and handle new exterior enterprise relationships
5. Community calls to those suppliers weren’t as performant as self-hosted fashions

The clear path ahead was to fine-tune a mannequin particularly tailor-made to Ibotta’s information and the wants of our Savers. This was made potential because of the hundreds of thousands of labeled search interactions we had gathered from actual customers by way of our LLM-as-a-judge course of inside our customized analysis framework. This high-quality manufacturing information grew to become our coaching gold.

We then launched into a methodical fine-tuning course of, leveraging our offline analysis framework extensively.

Key parts have been:

• Infrastructure: We used AI Runtime with A10s in a serverless atmosphere, and Databricks ML Runtime for stylish hyperparameter sweeping.
• Mannequin choice: We chosen a BGE household mannequin over GTE, which demonstrated stronger efficiency in our offline evaluations and proved extra environment friendly to coach.
• Dataset engineering: We constructed quite a few coaching datasets, together with producing artificial coaching information, finally deciding on:
  • One constructive end result (a verified good match from actual searches)
  • ~10 damaging examples per constructive, combining:
    • 3-4 “onerous negatives” (LLM labeled, human-verified inappropriate matches)
    • “In-batch negatives” (sampling of outcomes from unrelated search phrases)
• Hyperparameter optimization: We systematically swept issues like studying charge, batch measurement, period, and damaging sampling methods to search out optimum configurations.

After quite a few iterations and evaluations throughout the framework, our top-performing fine-tuned mannequin beat our greatest third-party baseline by 20% in artificial analysis. These compelling offline outcomes offered the arrogance wanted to speed up our subsequent manufacturing check.

Search that drives outcomes—and income

The technical rigor and iterative course of paid off. We engineered a search answer particularly optimized for Ibotta’s distinctive provide catalog and consumer conduct patterns, delivering outcomes that exceeded our expectations and provided the pliability wanted to evolve alongside our enterprise. Based mostly on these sturdy outcomes, we accelerated migration onto Databricks Vector Search as the inspiration for our manufacturing search system.

In our last manufacturing check, utilizing our personal fine-tuned embedding mannequin, we noticed the next enhancements:

• 14.8% extra provide unlocks in search.
  This measures customers choosing affords from search outcomes, indicating improved end result high quality and relevance. Extra unlocks are a number one indicator of downstream redemptions and income.
• 6% enhance in engaged customers.
  This exhibits a higher share of customers discovering worth and taking significant motion throughout the search expertise, contributing to improved conversion, retention and lifelong worth.
• 15% enhance in engagement on bonuses.
  This displays improved surfacing of high-value, brand-sponsored content material, translating immediately to raised efficiency and ROI for our model and retail companions.
• 72.6% lower in searches with zero outcomes.
  The numerous discount means fewer irritating experiences and a serious enchancment in semantic search protection.
• 60.9% fewer customers encountering searches returning no outcomes.
  This highlights the breadth of influence, displaying that a big portion of our consumer base is now persistently discovering outcomes, bettering the expertise throughout the board.

Past user-facing positive aspects, the brand new system delivered on efficiency. We noticed 60% decrease latency to our search system, attributable to Vector Search question efficiency and the fine-tuned mannequin’s decrease overhead.

Leveraging the pliability of this new basis, we additionally constructed highly effective enhancements like Question Transformation (enriching imprecise queries) and Multi-Search (fanning out generic phrases). The mix of a extremely related core mannequin, improved system efficiency, and clever question enhancements has resulted in a search expertise that’s smarter, quicker, and finally extra rewarding

Question Transformation

One problem with embedding fashions is their restricted understanding of area of interest key phrases, resembling rising manufacturers. To deal with this we constructed a question transformation layer that dynamically enriches search phrases in-flight based mostly on predefined guidelines.

For instance, if a consumer searches for an rising yogurt model the embedding mannequin may not acknowledge, we will remodel the question so as to add “Greek yogurt” alongside the model identify earlier than sending it to Vector Search. This offers the embedding mannequin with obligatory product context whereas preserving the unique textual content for hybrid search.

This functionality additionally works hand-in-hand with our fine-tuning course of. Profitable transformations can be utilized to generate coaching information; as an illustration, together with the unique model identify as a question and the related yogurt merchandise as constructive ends in a future coaching run helps the mannequin be taught these particular associations.

Multi-Search

For broad, generic searches like “child,” Vector Search would possibly initially return a restricted variety of candidates, probably filtered down additional by focusing on and funds administration. To deal with this and enhance end result range, we constructed a multi-search functionality that followers out a single search time period into a number of associated searches.

As an alternative of simply looking for “child,” our system mechanically runs parallel searches for phrases like “child meals,” “child clothes,” “child drugs,” “child diapers,” and so forth. Due to the low latency of Vector Search, we will execute a number of searches in parallel with out rising the general response time to the consumer. This offers a much wider and extra various set of related outcomes for wide-ranging class searches.

Classes Discovered

Following the profitable last manufacturing check and the complete rollout of Databricks Vector Search to our consumer base – delivering constructive engagement outcomes, elevated flexibility, and highly effective search instruments like Question Transformation and Multi-Search – this venture journey yielded a number of priceless classes:

1. Begin with a proof of idea: The preliminary hackathon method allowed us to rapidly validate the core idea with minimal upfront funding.
2. Measure what issues to you: Our tailor-made 50-metric analysis framework was essential; it gave us confidence that enhancements noticed offline would translate into enterprise influence, enabling us to keep away from repeated reside testing till options have been really promising.
3. Do not soar straight to fine-tuning: We discovered the worth of completely evaluating off-the-shelf fashions and exhausting these choices earlier than investing within the higher effort required for fine-tuning.
4. Acquire information early: Beginning to label information from our second experiment ensured a wealthy, proprietary dataset was prepared when fine-tuning grew to become obligatory.
5. Collaboration accelerates progress: Shut partnership with Databricks engineers and researchers, sharing insights on Vector Search, embedding fashions, LLM-as-a-judge patterns, and fine-tuning approaches, considerably accelerated our progress.
6. Acknowledge cumulative influence: Every particular person optimization, even seemingly minor, contributed considerably to the general transformation of our search expertise.

What’s subsequent

With our fine-tuned embedding mannequin now reside throughout all direct-to-consumer (D2C) channels, we subsequent plan to discover scaling this answer to the Ibotta Efficiency Community (IPN). This could convey improved provide discovery to hundreds of thousands extra customers throughout our writer community. As we proceed to gather labeled information and refine our fashions by Databricks, we imagine we’re nicely positioned to evolve the search expertise alongside the wants of our companions and the expectations of their prospects.

This journey from a hackathon venture to a manufacturing system proved that reimagining a core product expertise quickly is achievable with the correct instruments and assist. Databricks was instrumental in serving to us transfer quick, fine-tune successfully, and finally, make each search extra rewarding for our Savers.