CompanyErdunCo-founderMay 9, 20264 min read
We get asked at least once a week why ClaimIt exists when Paribus already did this. The short answer is that Paribus doesn't, anymore. But the longer answer is interesting, because what happened to the early services in this space is exactly the problem we're trying to solve.
Paribus launched in 2014, in the early consumer-fintech wave. The pitch was simple: connect your Gmail, and when prices drop on your purchases at supported retailers, we'll file the claim for you. The product worked well enough to be acquired by Capital One in 2016.
After the acquisition, Paribus was folded into what became Capital One Shopping. Over the following years, the automated-claim-filing capability quietly receded. Capital One Shopping still exists as a browser extension and price-comparison tool, but the original Paribus magic (connect Gmail, refunds appear) is no longer the product.
Earny launched around the same time, with similar mechanics, and went through similar transitions. Other smaller services have come and gone. The pattern is consistent: launch with a strong demo, get traction, run into the long tail of retailer-specific integration work, and either pivot away from the hardest half or quietly stop fulfilling the promise.
Filing a claim that gets approved at Best Buy is a different problem than filing one at Marriott, which is a different problem than filing one at Delta. Each retailer has its own form, its own preferred channel, its own quirks, its own response patterns. There are no clean APIs. The work scales linearly with the number of retailers, and most of the work is unglamorous infrastructure that nobody writes blog posts about.
Pre-LLM, the only way to handle this was to maintain hand-coded integrations per retailer. As a business that gets pressure on margins, you eventually start cutting the most expensive retailers, then the next most expensive, then you find yourself shipping a service that does price comparison and not the original claim filing at all. That's roughly what happened across the space.
We're betting that an LLM-based agent can do the retailer-specific work that previously required hand-coded integrations and human operations. Not perfectly. We're early. But well enough that the marginal cost of adding a retailer drops by an order of magnitude. That's the bet under all of ClaimIt.
We're not building a faster Paribus. We're building what becomes possible because the cheapest way to handle 26 retailers is no longer a 26-person ops team. If we're right, this category has another decade of growth in it. If we're wrong, at least we'll have shipped something that worked for a while, which is more than the alternatives have lately.
Erdun
Co-founder
Most major retailers and airlines refund you when prices drop after purchase, but only if you ask. ClaimIt does the asking, automatically, across 26 platforms.
Most major retailers will refund you when prices drop after you buy. Almost nobody collects. Here's why, and why we think that's about to change.
A monolithic agent looked great in demos and failed in production. Here's how we got to a four-agent pipeline that actually ships claims.