These founders turned a lockdown problem into an AI startup

Trackfundx, the Nigerian fintech founded by Precious Arikeri and Wahab Balogun, pivoted from an open‑banking expense tracker to a global AI receipt‑reading tool after mounting API fees, bank outages and a cross‑border move threatened its survival.

The Kenya angle

Kenyan SMEs can consider adopting a receipt‑scanning AI similar to Trackfundx’s new product to automate expense capture without depending on open‑banking APIs. A practical step is to pilot a low‑cost AI receipt‑reader, integrate it with existing accounting software, and train staff to capture digital copies of purchase receipts. This can reduce manual entry, lessen reliance on bank feeds, and provide real‑time insight into spending patterns.

When COVID‑19 lockdowns began in 2020, daily commutes stopped and social outings disappeared, but many people, including Arikeri, found their spending rise instead of fall. She was buying more online, topping up data, and transferring money between accounts, yet had no simple way to track where the money went. Existing personal‑finance apps in Nigeria either required tedious manual entry or were too rigid for her spending habits, prompting her to look for a better solution.

Arikeri brought the problem to Wahab Balogun, a software engineer she had worked with at Vascon Solutions. In June 2022 they launched Trackfundx, a personal‑finance app that aggregated transactions from multiple bank accounts into a single dashboard. The premise was simple: many users keep money across several banks but lack a consolidated view of their finances. By connecting accounts, Trackfundx let users pull transactions into the app and see overall spending.

However, the founders quickly realized that aggregating raw transaction logs was not enough. “If I have to come here and write, ‘transfer money for this, transfer money for that,’ it’s really not very efficient, and also time consuming,” Arikeri said. Most of her spending occurred through bank transfers or card payments, so they aimed to fetch data automatically rather than rely on user‑entered entry.

To achieve this, they pursued two parallel tracks. They hired a machine‑learning engineer to train an in‑house categorisation model using volunteered bank statements, while Balogun integrated Mono, an open‑banking API, to fetch user transactions automatically. An early version of the product used Mono as a free offering to collect feedback, and later they experimented with weekly and monthly paid subscriptions.

As the user base grew, the challenge shifted from data collection to data interpretation. Transaction logs showed when money left an account but not what was actually purchased. When Mono launched its own categorisation feature, it still fell short of Trackfundx’s needs because the founders wanted customisable categories that users could tailor. “That category didn’t really work with us, because we also wanted to be able to customise ours. We wanted users to be able to customise theirs as well,” Arikeri explained. Consequently, they continued building their own machine‑learning categorisation system, a decision that set the stage for the later pivot.

The pivot came when mounting API fees, frequent bank outages and the logistical strain of moving operations across borders made the open‑banking model unsustainable. Rather than abandon the venture, Arikeri and Balogun discarded the open‑banking approach and rebuilt the product around AI that reads receipts. This new tool now serves roughly 10,000 users across 100 countries.

Talk to us on WhatsApp at 0711 344 702, we work with Kenyan businesses on exactly these challenges.