The situation:

When you're in the business of accounts receivable,precision matters. Every client interaction carries weight. A missed question,an unanswered email, an automated response that simply doesn't land — thesearen't just inconveniences. They erode trust. And when your customer base isgrowing faster than your support team can scale, the pressure compoundsquickly.

That was the position Payt found itself in.

Payt provides digital accounts receivable services tobusinesses across the Netherlands and beyond. As their client base expanded, sodid the volume of inbound questions. The maths was straightforward, and thechallenge was real: more customers meant more questions, and more questionsmeant a growing gap between what their support operation could handle and whattheir customers actually needed.

"We knew we wanted to serve customers digitally,"says one member of Payt's operations team. "But wanting that and actuallyachieving it are two different things. The question was always: is ourknowledge base good enough to let automation do its job?"

Payt had invested in a chatbot to handle routine customerqueries. On paper, the logic was sound. In practice, the results were patchy.The chatbot could handle questions it had been trained on. But customerquestions don't always follow a script. New scenarios emerged. Edge casesappeared. When the knowledge base had gaps, the chatbot fell short — andcustomers either received incomplete answers or had to wait for a human to stepin.

‍

The solution:

"What Comprendly does is compare what customers areactually asking against what the knowledge base contains," the teamexplains. "It sounds simple, but the impact is significant. You're notguessing where the gaps are. You're seeing them clearly, question byquestion."

Comprendly analysed each relevant customer question andchecked, systematically, whether the chatbot could answer it automatically. Forevery question that fell through — every instance where the knowledge base cameup short — the content was updated by the team. Not flagged for a future reviewcycle. Not added to a backlog. Updated.

The result was a knowledge base that got progressivelystronger. Each gap identified became a question answered. Each update increasedthe proportion of queries that digital assistants could handle without humanintervention.

That distinction matters. Payt wasn't looking for a replacement to their existing tools. They were looking for something that wouldmake those tools perform as intended. Comprendly provided continuous coverageanalysis — not a one-time audit, but an ongoing process that runs in thebackground, surfacing new gaps as customer questions evolve.

For a company in growth mode, that continuity of knowledge isvaluable. Product features change. Support topics that didn't exist lastquarter become common this quarter. A knowledge base that only gets reviewedperiodically will always be catching up. One that's being improved in real timestays current.

‍

The results:

For Payt, the outcome is cleaner than it is dramatic: morequestions answered correctly, fewer escalations, and a support operation thatscales with the business rather than behind it.

‍