If you’re responsible for making sure commission payouts are accurate—and actually match what sales reps earned—you know how quickly things get messy. Even with “automation” tools, errors slip through. People get cranky about missing money, finance gets stuck chasing down discrepancies, and trust in the numbers tanks.
This guide is for anyone who has to audit and reconcile commission payments, specifically using Spiff. Maybe you’re in finance, revenue ops, or sales admin. Maybe you inherited a Spiff setup and need to figure out what’s actually going on. Either way, I’ll walk you through a clear process to keep audits tight and headaches to a minimum.
Let’s get into it.
Step 1: Know What You’re Auditing (Don’t Just Click “Export”)
Before you even open Spiff, get clear on what’s in scope:
- Which commission plans? Are you looking at a single team, or is this a company-wide review?
- What time period? Some teams audit monthly, others only at quarter-close. Know your window.
- Payouts or accruals? Are you reconciling what’s been paid, or what’s been earned but not yet paid?
Pro tip: Write this down before you start. It sounds basic, but if you don’t, you’ll end up re-running reports or missing something obvious.
Step 2: Pull the Core Data (Spiff and Source of Truth)
You need two sets of data:
- What Spiff Thinks Happened
- In Spiff, pull the commission statement exports for your scope—pick “Detail” if you want line-by-line, or “Summary” for higher level.
-
Export as CSV, not PDF. You need data you can work with, not screenshots.
-
What Actually Happened
- Grab your “source of truth” data. Usually, this is your CRM (like Salesforce), ERP, or the payroll system that sends money out.
- Make sure to pull matching periods and fields: deal IDs, rep names/emails, amounts, and dates.
Don’t skip this: If your source of truth is a mess, start there. No tool can fix garbage data—Spiff included.
Step 3: Map the Data (So You’re Comparing Apples to Apples)
This is where most audits go sideways. Spiff might call something “Rep Name,” but your payroll system uses “Employee Email.” Align the fields before you start reconciling.
- Create a mapping table: Just a simple spreadsheet that says “Spiff Field” = “Payroll Field.”
- Normalize IDs: If you don’t have a shared unique identifier (like a Salesforce Opportunity ID), you’ll have to match on names and dates. Prepare for manual work if that’s the case.
What to ignore: Don’t waste time on columns you don’t care about (like “Manager Name” if it’s not tied to payout). Focus only on fields that affect money.
Step 4: Reconcile the Numbers (Find Gaps, Fast)
Now, the meat of the audit. There are fancy tools and scripts for this, but you can get 90% there with a spreadsheet and some patience.
- Join the datasets: Use VLOOKUP, XLOOKUP, or a pivot table to match Spiff’s numbers to your source of truth, deal by deal.
- Flag discrepancies: Highlight where amounts don’t match, or where one side has a deal the other doesn’t.
- Check for common issues:
- Missing deals (exists in one system but not the other)
- Calculation errors (rates or splits applied wrong)
- Duplicates (watch out for deals paid twice)
- Timing issues (paid in wrong period)
Honest take: Spiff is pretty good at handling straightforward plans. The more custom logic you have (tiered rates, clawbacks, special deals), the more likely you’ll find edge-case errors. Don’t assume “automation” means “error-free.”
Step 5: Dig Into Discrepancies (Don’t Assume Malice)
You’ll always find mismatches. The key is to triage:
- Is it a config issue? (e.g., Spiff plan logic doesn’t match the comp doc)
- Is it a data issue? (e.g., wrong close date in CRM, missing rep assignment)
- Is it a timing issue? (deal booked late, payout period off by a month)
- Is it a human error? (manual adjustments, overrides, or bad data uploads)
How to fix:
- If it’s a config issue, you’ll need a Spiff admin to update the plan logic.
- Data issues get fixed at the source—CRM or payroll.
- Human errors are the hardest. Document them, fix them, and (ideally) change the process so they don’t happen again.
What doesn’t work:
- Blaming Spiff for everything. No tool is magic. Garbage in, garbage out.
Step 6: Document Everything (But Keep It Simple)
Nobody wants to write a novel about their audit, but you need a paper trail.
- Keep a change log: What did you find? What did you fix? Who approved it?
- Summarize the issues: One-liners are fine—“Q2, Rep X, missing deal, fixed in CRM.”
- Save exports: Keep copies of the original and reconciled files.
If someone asks “why did Rep Y get a correction?” you should be able to answer in 30 seconds, not after a 2-hour data dive.
Step 7: Communicate and Close the Loop
Once you’ve reconciled and fixed errors, don’t just file the audit away.
- Share the findings: Depending on your org, this might be a quick summary to finance, sales ops, or leadership.
- Tell reps about corrections: If someone’s payout changed, let them know proactively. It builds trust.
- Update your process: Did you find recurring issues? Fix the root cause—better data flows, clearer comp docs, tighter plan logic in Spiff.
What to ignore:
- Don’t waste cycles on “zero-impact” issues (like spelling differences in names that don’t affect payouts). Focus on what hits the bank account.
Pro Tips for Smoother Audits
- Automate what you can, but don’t trust black boxes. Spiff’s automations are great, but always sanity-check the math.
- Lock down data sources. The fewer hands in the pot, the easier it is to trust your numbers.
- Schedule regular reviews. Auditing once a year is a recipe for disaster. Monthly or quarterly is best.
- Learn Spiff’s quirks. Every tool has them. Save time by knowing what reports actually show vs. what’s just for show.
Keep It Simple, Iterate Often
Commission audits will never be “set it and forget it.” But you don’t need to drown in endless spreadsheets or chase your tail with every payout. Start with clear scope, map your data, focus only on what matters, and keep your documentation simple. If you find a mess, don’t panic—fix what you can, improve your process, and do it a little better next time.
You’ll never catch every edge case, but with a solid, repeatable process, you can keep things tight and keep your sanity. And that’s what actually matters.