At Kard, we consider higher information results in higher rewards — and that begins by understanding what individuals really purchase.
By categorizing transactions at scale, we’re in a position to assist manufacturers goal the appropriate clients, issuers improve card utilization, and shoppers get rewarded in ways in which really feel private.
Traditionally, categorizing transaction information was messy and guide. However with a brand new Databricks-powered method, Kard is now in a position to classify billions of transactions shortly, precisely, and flexibly, laying the inspiration for personalised rewards that drive loyalty and long-term worth.
What Kard does
Kard drives loyalty for each cardholder and shopper by a rewards market.
Our platform offers manufacturers like Dell, CVS, Allbirds, and Spherical Desk Pizza entry to tens of hundreds of thousands of shoppers by delivering money again affords by issuer and fintech banking apps, rewards packages, and EBT platforms. Seeing a ten% or 15% money again supply nudges clients towards a purchase order (usually one which’s larger so as worth).
And on Kard’s pay-for-performance mannequin, manufacturers solely pay when a purchase order happens, guaranteeing ample attain with out the excessive prices or dangers of conventional media shopping for.
Money again rewards profit the issuers and fintechs, too. By providing rewards that customers care about, they improve engagement and utilization amongst their cardholders.
However what makes Kard significantly particular is the category-level insights it captures, offering perception with out exposing any PII.
Why category-level insights matter for rewards
Realizing what customers spend their cash on helps manufacturers (and banks and fintechs) perceive their buyer bases in a richer manner. In combination, the spend patterns Kard collects:
- Gas smarter advertising campaigns — you’ll be able to determine high-intent segments based mostly on habits. For instance, if a big share of customers commonly use rideshare providers late at night time, banks and types can goal them with weekend-specific cashback affords.
- Inform product design by revealing unmet wants. If information reveals that youthful customers are shifting spend from grocery shops to meals supply apps, a fintech may prioritize rewards tied to convenience-driven classes.
- Encourage new partnerships by surfacing frequent service provider overlaps throughout person cohorts. As an illustration, if frequent vacationers persistently guide the identical chain of resorts and rental automotive companies, there’s a robust case for negotiating co-branded rewards or unique perks with these companions.
Categorical patterns get much more highly effective whenever you zoom in on the person.
As an illustration, maybe a particular person spends essentially the most on sports activities playing. A generic retail supply may go unnoticed, however a promo for a betting app might drive instantaneous engagement.
Say a unique person has decreased spend on groceries however elevated their use of meals supply apps over the past 90 days. That alerts shifting habits — and a possibility to reward comfort over price.
Lastly, one other person flies usually, however all the time with the identical airline. That loyalty will be bolstered with focused rewards, and even upsold to that airline’s premium tier. Different airline manufacturers could not even wish to goal that particular person. Or they may solely floor the very best money again affords to enhance their odds of stealing the client away from their most well-liked airline.
With out dependable transaction classes, although, none of those personalization eventualities are potential.
How rewards platforms traditionally labeled transactions
Categorization is the important thing to unlocking high-ROI go-to-market methods for our manufacturers and issuers, but it surely’s more durable than it sounds.
First, you’ve bought to label all of the transactions. Historically, there’ve been two methods to perform this:
- Have analysts evaluation every transaction, line by line, tagging each based on a predefined taxonomy. As you may guess, this technique is tedious, error-prone, and extremely arduous to scale.
- Let customers categorize their very own transactions. Whereas this method leaves much less work for analysts, it additionally riddles the information with inconsistencies. One person may label Domino’s as “quick meals,” one other may name it “pizza,” and a 3rd may tag it “consolation meals,” making it extraordinarily tough to attract dependable insights.
As soon as a considerable quantity of transactions are labeled, engineering groups can begin coaching machine studying fashions like LightGBM, XGBoost, or BERT to predict classes for brand spanking new, unseen transactions.
Over time, these fashions might get rid of the necessity for guide tagging. Nevertheless, they require upkeep and upgrades as companies evolve and transaction codecs change. Including new class varieties (say, for an rising business or a brand new shopper vertical) might contain retraining and even re-architecting the mannequin.
To assist our rising enterprise, we would have liked a extra streamlined, correct, and versatile method to categorizing the billions of transactions we obtain every month.
How Databricks powers a contemporary categorization method
Working with Databricks, we’ve provide you with a singular, scalable system for transaction categorization:
- Leveraging Databricks AI Features to run batch, agentic workflow that categorizes transactions based mostly upon an internally derived taxonomy.
- The outcomes are constrained with structured output performance, utilizing the json_schema response format with the enum characteristic to restrict errors.
- AI brokers course of incoming transactions towards the required taxonomy, one for every kind of categorization. In a single occasion, we will seize high-level classes like Journey, after which determine hierarchical classes like Journey → Airfare and even additional, Journey → Airfare → Regional Airline.
- Inconsistencies are handed right down to paths which are evaluated by agent judges, whichallows for re-categorization within the case of errors.
The light-weight prices of this new method have given our group extra flexibility. If a brand new line of enterprise opens up, we will alter our classes instantly — with out having to completely retrain the mannequin. In actual fact, we simply opened up some new CPG classes to assist a partnership with a preferred rewards app.
A few of our shoppers have requested that we use their very own class mapping to align with their inside methods. Now, we will simply go that different taxonomy straight to our new system and it’ll translate outputs accordingly.
“Having the ability to roll up retailers into their respective classes affords us a number of leverage with clients,” says Chris Wright, Kard employees machine studying engineer.
“For instance, we will inform retailers that customers inside their class usually discover supply varieties x, y, and z work greatest. We are able to additionally assist retailers goal a section of customers who’ve bought with them prior to now and had a current acceleration in spend inside, say, meals supply or journey share. And we will inform our clients who they’re competing with of their class and area to allow them to refine their campaigns accordingly.”
What’s subsequent for Kard and Databricks: hyper-personalization
Transaction classes could seem to be a behind-the-scenes element. However the agility we get from the Databricks AI Features-powered categorizer makes it potential for us to maneuver quick with out breaking our information basis, and believe within the scalability of the answer.
Plus, it additionally opens the door to new sorts of services for Kard clients, like:
- Personalised card affords based mostly on shifting meals or journey habits
- Stickier rewards for loyal clients of a particular service provider
- Sensible nudges based mostly on time-of-day or seasonal habits
- Service provider-funded cashback packages focused by section, not simply demographics
- Earned factors packages (for manufacturers and issuers)
By investing in smarter categorization now, we’re laying the groundwork for a very personalised rewards expertise that enhances buy frequency, will increase AOV, and sustains buyer loyalty for manufacturers and issuers alike.
Conclusion
On this weblog put up, we confirmed how Databricks AI Features are powering information enrichment for Kard’s categorization pipeline. This permits personalization at scale, and drives loyalty and worth at a fraction of the trouble it might usually take.
Fascinated about studying extra? Attain out to considered one of our specialists right this moment!
About Kard
Kard is a New York-based fintech firm based in 2015 that gives a rewards-as-a-service platform for banks, neobanks, and card issuers. Its API allows monetary establishments to shortly launch and customise cardholder rewards packages, connecting customers to hundreds of retailers and types throughout the US. Kard’s platform is designed to drive buyer loyalty and engagement by making it straightforward for cardholders to earn rewards on on a regular basis purchases. The corporate is backed by main traders and serves over 45 million cardholders by its issuer and companion community.
