It starts with a spreadsheet.
Not a complicated one, necessarily. Just a file that needs to be opened, cross-referenced, cleaned, and moved into the right system before anyone can do anything useful with it. Then another file. Then another. Then a transaction that doesn’t map cleanly to any jurisdiction. Then a rate that changed last quarter and nobody updated the lookup table. Then a classification flag that has to be resolved before the filing can go out.
By the time the actual tax work begins, the day is half over.
This is the quiet reality inside accounting functions at growing businesses. The work that was supposed to be compliance has become data operations — and most teams didn’t sign up for that.
When Tax Became a Data Problem
PwC research found that tax teams spend as much as 60% of their compliance time on data management. Not analysis. Not planning. Not the judgment calls that require expertise. Data gathering, cleaning, formatting, and entry.
That number is striking, but it’s not surprising to anyone who’s lived it. High-volume businesses generate thousands of transactions every month. Each one needs proper classification, jurisdiction mapping, and rate application. When you’re doing that manually — or semi-manually, with spreadsheets and human reconciliation — the math becomes brutal fast. You’re not running a tax function. You’re running a data processing operation that occasionally produces a tax output.
The consequences are real. A transposed digit triggers an audit. A missed classification creates a penalty. A filing that went out with stale rate data generates a notice that lands on someone’s desk six months later, with interest attached.
And the margin for error is shrinking. Tax authorities aren’t waiting for quarterly filings to catch up anymore. Digital reporting requirements are spreading globally, and real-time monitoring is increasingly the standard. Errors that once might have gone unnoticed now surface faster — and the consequences are more immediate.
Why Manual Processes Break at Scale
There’s a version of manual tax compliance that works. It works when transaction volume is manageable, when your business operates in a handful of jurisdictions, and when the rules don’t change too often. A lot of companies built their compliance function around that version.
Then they grew. Or expanded into new states. Or launched a SaaS product alongside their physical goods. Or started selling through a marketplace. And suddenly the system that worked fine is straining under complexity it was never designed to handle.
Traditional approaches tend to break at the same places: account mapping, rate application across multiple jurisdictions, and exception handling. A process that takes a few hours when your transaction volume is modest can take days when it isn’t. The team doesn’t grow proportionally to the volume. So what was a Tuesday afternoon task becomes a two-week sprint, and the team that should be doing analysis is stuck doing data entry instead.
This is where the gap between teams using automation and teams still relying on manual processes starts to compound. It isn’t just an efficiency gap. It’s a risk gap, an accuracy gap, and increasingly, a talent gap. The finance professionals you want doing this work didn’t get into the field to copy numbers between spreadsheets.
What Changes When AI Handles the Data Layer
AI-powered tax processing doesn’t eliminate the need for accounting expertise. What it does is remove the part of the job that accounting expertise shouldn’t be spent on.
When the data layer is automated, the workflow changes shape entirely. Systems process and classify transactions continuously rather than in batches. Rate calculations happen in real time, not at deadline. Jurisdiction mapping doesn’t require someone to manually maintain a lookup table. Exceptions get flagged rather than buried.
The practical result is that the compliance function can actually function as a compliance function. Controllers and tax leads spend their time on review, judgment, and planning — not on cleaning imports or chasing down mismatched data. Filing timelines compress. Audit exposure drops. And the team isn’t wrecked at the end of every close period.
There’s also a downstream effect that’s easy to underestimate: when the data is clean and the outputs are reliable, client and leadership conversations change. Instead of explaining what happened and why a flag appeared, you’re reporting on outcomes with confidence.
The Window Is Narrowing
Tax automation isn’t a future-state initiative anymore. The regulatory environment is pushing the timeline. As more jurisdictions move toward real-time digital reporting, the lag that manual processes introduce becomes a liability rather than just an inconvenience.
The businesses pulling ahead right now aren’t the ones with the most sophisticated tax teams. They’re the ones that removed the friction from the data layer so their teams could actually do the work they were hired to do. That shift is what turns tax compliance from a source of organizational anxiety into something closer to a system that just runs.
The question for most accounting functions isn’t whether AI will change how this work gets done. It already is. The question is whether your team is spending its energy on the part of the job that actually requires human judgment — or whether it’s still stuck at the spreadsheet.
How Zamp Approaches This
Zamp is a fully managed sales tax compliance platform built to take the data and execution burden off accounting teams entirely. We handle classification, jurisdiction mapping, rate calculation, filing, and notice resolution — end to end, under your brand if you’re a firm, or directly for your business if you’re managing compliance in-house.
The model is simple: your team owns the strategy and the client relationships. We own the execution. The data layer doesn’t become your problem, because it never lands on your desk in the first place.
If your team is spending more time on the mechanics of compliance than on the substance of it, we’d like to talk.
Brandon Roth
