Dispute-tracking fintech Glimpse introduced Wednesday that it raised a $35 million Collection A led by Andreessen Horowitz, with participation from 8VC and Y Combinator.
Founders Akash Raju, Anuj Mehta, and Kushal Negi, attended Purdue collectively and had been initially constructing a startup that did Airbnb product placements. That firm launched in 2020, however by 2024, the founders pivoted to an entirely new thought: Glimpse, a platform that helps retailers automate monetary deduction processes.
It raised a $10 million spherical final yr, led by 8VC, after the enterprise pivot, which it referred to as a Collection A spherical on the time. It’s now calling this contemporary $35 million a Collection A spherical, and rebranding its earlier Collection A as a seed spherical. The corporate has raised $52 million up to now, together with funding they raised earlier than they pivoted.
“We in the end felt we lacked product-market match and determined to laborious pivot,” Raju stated of the primary, unsuccessful thought. “On this course of, we had publicity to manufacturers’ again workplaces and the chaos that was promoting in retail, in the end main us to start out Glimpse as it’s at this time.”
They met their lead a16z investor by way of a mutual founder buddy. “We constructed a powerful relationship as we scaled the enterprise. Actually excited we are able to companion with them for this subsequent stage of development,” he continued.
Deductions are the quantities a retailer subtracts from what they owe a model when settling an bill. It’s commonplace and usually works like this: A model payments the retailer, the retailer pays the model. If it pays lower than what was billed, it gives a cause, resembling if the products had been broken.
A few of the deductions are for legitimate causes, however some aren’t — these are referred to as invalid deductions, and they’re tedious to trace and handle on the again finish. “These errors are surprisingly frequent,” Raju, the corporate’s CEO, stated, including that “a model would possibly ship stock accurately however nonetheless be charged for a brief cargo.”
Techcrunch occasion
San Francisco, CA
|
October 13-15, 2026
“Groups log into a number of retailer programs, pull scattered paperwork, assessment line gadgets, reconcile towards inner information, and handle disputes end-to-end. The problem is pushed by fragmented, unstructured information and siloed workflows throughout programs and groups,” he stated of how the method often goes.
If the model doesn’t reconcile each invalid deduction, that would lead “to constant income leakage,” he stated.
Glimpse says it helps with this course of by reviewing deductions, flagging invalid ones, and submitting disputes, serving to corporations recuperate cash they could have missed or misplaced. The platform’s AI brokers log right into a retailer’s portal, discover and centralize all needed paperwork, then classify every deduction, Raju defined. From there, the AI brokers validate every change towards inner information (resembling provide chain information and promotion calendars) to find out which deductions are reputable and which aren’t.
The corporate stated it really works with greater than 200 retail manufacturers, together with Suave and its lip balm model Chapstick.
“When points are recognized, Glimpse robotically recordsdata disputes, follows by way of on the method, applies recovered money, and syncs every thing again to the model’s ERP,” Raju stated, including that the product integrates throughout a number of programs. Along with the principle enterprise useful resource planning monetary software program, it integrates with promotion calendars, and retail portals. It could truncate an extended course of all the way down to days, he stated.
Regardless of Glimpse’s automation, Raju stated his firm does have people within the loop, “primarily round making certain outcomes,” he stated, like “following up on disputes to drive decision and money restoration, in addition to high quality assurance on vital steps like classification and information extraction.”
The system will get smarter every time a deduction is processed and constantly refines its classification, validation, and backbone. “Over time, this creates a compounding information benefit, the place every new integration and buyer makes the system smarter and more practical throughout the whole community,” he stated.
Others are tackling invalid deduction with software program, too, resembling Revya and Confido.
“Our imaginative and prescient is to be the AI infrastructure for CPG and retail manufacturers, and this capital helps proceed executing towards that imaginative and prescient,” he stated.
