You opened your system this morning and found four earnings codes that all do the same thing, a deduction nobody remembers creating, and a report that breaks every time someone runs it. None of this happened overnight. It built up over years of quick fixes, staff turnover, “just add a new code for now,” and open enrollments where the goal was to survive, not to keep things tidy. That slow build-up has a name: configuration drift. And if you’re staring at it right now, you are not behind, and you are not the only one.
Here’s the reassuring part. Drift is normal, it’s fixable, and you do not have to rip the whole thing apart to clean it up. A good ADP Workforce Now cleanup is methodical, not dramatic. You inventory what you have, you decide what stays, you retire the rest in a safe order, and you write down what you did so the next person inherits clarity instead of a mystery. We’ve helped a lot of teams walk this exact path, and the system on the other side is calmer, faster to run, and far easier to trust.
This guide walks through where drift usually hides, how to take inventory without breaking anything, and how to prune safely while you document as you go. Read it as a map, not a checklist you have to finish in one sitting.
Drift doesn’t spread evenly. It pools in a handful of predictable places, and knowing where to look saves you hours of poking around.
The biggest offender is usually earnings and deduction codes. Over time you end up with three codes for the same bonus, a “temporary” code from a project two years ago, and deductions that point to the wrong tax treatment because someone copied an old one and forgot to change a field. These codes touch every paycheck, so errors here are expensive and easy to miss.
Next come inactive items hiding in plain sight. Terminated departments, closed locations, job titles nobody uses, and pay groups that served a one-time need. They clutter dropdowns, slow down the people doing data entry, and make reports harder to read.
Workflows and approvals drift too. A manager who left three years ago is still a required approver on time-off requests, so things stall. An onboarding step routes to a role that no longer exists. New hires fall through because the path they were supposed to follow quietly broke.
Security roles are the quiet risk. As people change jobs, permissions pile up instead of getting reset. Someone in a frontline role can suddenly see compensation data they shouldn’t. Drift here isn’t just messy; it’s a compliance exposure.
Finally, custom reports. You probably have a folder full of reports with names like “Final v2 USE THIS ONE.” Half pull from fields that changed. A few duplicate each other. Nobody’s sure which is the source of truth.
The common thread across all of these is that nobody set out to make a mess. Each addition made sense at the time. Drift is what happens when a hundred reasonable decisions, made under deadline pressure over many years, never get a second look. That’s why the fix isn’t blame; it’s a fresh, deliberate pass.
The single biggest mistake we see is deleting before documenting. You spot a code that “looks useless,” remove it, and discover it was feeding a year-end report. Don’t do that. Build the inventory first.
Pull a full list of your active and inactive codes, your departments and locations, your job and pay structures, your workflows, your security role assignments, and your saved reports. Put each category on its own tab in a simple spreadsheet. For every item, capture four things: what it is, where it’s used, when it was last touched, and who owns the answer. That last column matters more than the others. Most cleanup decisions come down to “ask the person who’d know,” and writing down who that is turns guesswork into a quick conversation.
You don’t need fancy tooling for this. A clean spreadsheet and a few afternoons gets you a complete picture. The inventory is the foundation for every decision that follows, and it doubles as documentation once you’re done.
One more tip: timestamp the inventory and note who built it. Six months from now, when someone questions a decision, you’ll want to know the picture you were working from. The inventory isn’t just a to-do list; it’s the record of what the system looked like before you started, which is exactly what you’ll wish you had if anything needs to be traced back later.
With the inventory in front of you, sort every item into one of three buckets: keep, retire, or investigate.
Keep is anything actively used and correctly configured. Leave it alone. Cleanup isn’t about churn for its own sake.
Retire is anything clearly dead. A location you closed. A code that hasn’t touched a paycheck in three years. A duplicate that does exactly what another code does. These are your safe wins, and you’ll have more of them than you expect.
Investigate is the gray zone, and it’s where you spend most of your thinking time. A code used once last quarter. A workflow that might still matter to one department. A report someone swears they need but can’t explain. Resist the urge to guess. Tag it, note the open question, and route it to the owner from your inventory. One short email usually settles it.
A good rule of thumb: if an item hasn’t been used in 12 to 18 months and nobody can name a reason to keep it, it’s a retire candidate. But always confirm before acting on anything tied to taxes, year-end, or reporting.
Pruning is where caution pays off. Work in a sequence that limits risk.
Start with the lowest-stakes items. Inactive departments, unused job titles, and stale dropdown values rarely break anything when you clean them up. Get those done and you’ve already made the system noticeably easier to use.
Move next to deactivating rather than deleting. In most cases you want to make an item inactive, not erase it. Deactivating keeps historical records intact, which protects your reporting and your audit trail, while removing the item from active use. Hard deletion should be rare and reserved for things you’re certain carry no history.
Save earnings and deduction codes for when you’ve confirmed tax treatment and historical use. Before you retire a code, check that no current employee is assigned to it and that it isn’t referenced in any active report. When you consolidate duplicates, map the old code to the survivor and document the switch so anyone reading a historical record understands what happened.
Handle security roles as a deliberate pass. Review each role against what people in that role actually need to do, then reset assignments to match. This is the one area where it’s worth being strict, because over-permissioned access is a real risk.
Throughout, change a small batch, then run a payroll preview or a key report to confirm nothing broke. Small steps with a quick check beat a big bang every time.
The cleanup is only half the value. The other half is making sure drift doesn’t quietly return the moment you look away. That means writing things down while you work, not promising to do it later.
For every code, keep a short note on what it’s for and how its tax treatment is set. For workflows, record who approves what and why. For security roles, keep a plain-English description of what each role can see and do. Park all of it in one shared place your team can actually find.
The payoff shows up the next time someone asks “why is this set up this way?” Instead of an archaeology project, you have an answer. Documentation is what turns a one-time cleanup into a system that stays clean.
A client came to us after roughly eight years on ADP Workforce Now with no real cleanup along the way. Their earnings and deduction list had grown to around 90 active codes. Payroll took longer every cycle because the data-entry team had to scroll past dozens of options they never used, and two codes with nearly identical names kept getting picked wrong.
We started with an inventory. Of the 90 codes, about 30 hadn’t touched a paycheck in over two years, and another 10 were duplicates. That’s roughly 40 retire candidates flagged in an afternoon. We sorted them into keep, retire, and investigate, then routed the 15 or so gray-zone items to the owners we’d identified. Most came back as easy retires within a week.
We deactivated the dead codes in small batches, ran a payroll preview after each batch, and confirmed nothing shifted. We consolidated the duplicate codes down to one each and documented the mapping. By the end, the active list sat around 45 codes, payroll entry was visibly faster, and the wrong-code mistakes stopped. The whole effort took a few weeks of part-time work, and the documentation they now keep means the list shouldn’t balloon again.
Not if you deactivate instead of delete. Deactivating removes an item from active use while keeping the historical records that feed your reports and year-end filings. The risk comes from hard-deleting something with history attached, which is exactly why we confirm tax treatment and historical use before retiring any code.
It depends on how much drift has built up, but most mid-market teams can get through a meaningful cleanup in a few weeks of focused, part-time effort. The inventory is the slow part. Once you know what you have, the pruning moves quickly because you’ve already made the hard decisions on paper.
Yes. The whole approach is built to avoid disruption. You work in small batches, you check a payroll preview after each change, and you save the highest-stakes items for last. Done this way, your people keep getting paid on schedule while the system gets cleaner underneath them.
A clean system isn’t a one-time event; it’s a habit. The teams that stay clean do a few simple things. They add new codes deliberately, with a naming convention and a documented purpose, instead of spinning one up under deadline pressure. They run a light review once or twice a year to catch new drift early. They reset security roles whenever someone changes jobs, not someday. And they keep their documentation current, so the next person who asks “why is this here?” gets an answer instead of a shrug.
You don’t need to be perfect. You need a system that’s understood and a small routine that keeps it that way. That’s the difference between drift that compounds for years and drift that gets caught while it’s still small.
If the cleanup feels like more than your team can take on alongside running payroll, that’s exactly what we’re here for. Ignite HCM is staffed Only by former ADP service professionals. We’ve done these cleanups from the inside, we know where the drift hides, and we know how to prune it without breaking a single paycheck. No tickets, no hold queues — a dedicated consultant who learns your setup and stays with you.
Tired of fighting your own system? REQUEST A CONSULTATION (ignitehcm.com/solutions/optimization).
ADP and the ADP logo are registered trademarks of ADP, Inc. This article is general guidance only; confirm tax and compliance decisions with your licensed advisor or ADP representative.
