Your First 30/60/90 Days as a New Controller
You walked in on Monday and the close was already 12 days long. There are two open audit findings from last year that nobody has remediated. The CFO told the board, on a slide you weren't there to write, that the team will hit a 5-day close by end of Q1. And there's a 47-tab Excel reconciliation in your shared drive that the prior Controller built, that nobody else fully understands, and that the auditors flag every year.
Welcome to the job.
The Controllers I've watched succeed in their first 90 days do one thing differently from the ones who blow up: they refuse to commit to a close-speed number in week one. The ones who promise a 5-day close before they've sat through a single month-end almost always miss it, and the miss is what defines them for the next year. Diagnose first. Fix second. Commit third. In that order.
Here's the plan.
Why the first 90 days actually matter
Close speed is the single most visible Controller metric to the CFO and the audit committee. Faster close means faster reporting, fewer surprises, more time for analysis. Every CFO has been pitched on it. Every board has heard the slide. So when a new Controller starts, the clock is already running, and it's running on a number you didn't set.
Move too fast and you miss something material. Material misses become 10-K disclosures, restatements, or audit committee meetings you don't want to be in. Move too slow and you become the bottleneck the CFO already warned the CEO about, and now you're the topic of a 1-on-1 in month four.
The trick is the first 90 days have to produce three things, in order: a credible diagnosis, three small visible wins, and a written plan the CFO has signed off on. Get all three and you've earned a year of runway. Skip any one and the team's confidence in you will quietly erode.
Days 1-30: diagnose, don't fix
The single biggest mistake new Controllers make in month one is issuing directives. You don't know enough yet. The team you inherited has knowledge you don't have, the close has dependencies you can't see, and the audit findings have history you weren't there for. So month one is observation and documentation. Nothing else.
Week 1: read everything
Pull the prior year's audit report and the management response letter. Read every open finding and the proposed remediation. Read the SOX risk and control matrix if you have one, even if it's three years stale. Read the last six months of close calendars and look at the variance between planned and actual close days. If the planned close is day 8 and the actual is day 12, that gap of four days is your first clue.
Pull the close binder, if there is one. If there isn't one, that's also a clue.
Week 2: audit the close calendar day-by-day
Sit down with whoever owns the close calendar (usually the Sr Accounting Manager or Asst Controller) and walk through it task by task. For each task, write down: who does it, when it starts, when it ends, what it depends on, and what depends on it. You're looking for three things:
- Sequential tasks that could run in parallel. Most stale close calendars have at least two of these. Bank rec waiting on AP cutoff is the classic one.
- Hand-offs that slip. Somebody finishes their piece on day 4 but the next person doesn't start until day 6. The two-day gap is a calendar problem, not a work problem.
- Tasks that block multiple downstream tasks. These are your real bottlenecks. The revenue cutoff is almost always one of them.
Week 3: map material accounts
Decide your materiality threshold and write it down. The standard B2B SaaS rule of thumb is 5% of revenue or 5% of pre-tax income, whichever is lower, but check whatever your audit firm uses. Then list every account that breaches that threshold. These are the accounts that will fail an audit if you get them wrong. Everything else is housekeeping.
You'll find the list is shorter than you expected. Probably 15 to 25 accounts in a $50M-$200M ARR SaaS company. Those are the accounts that get your real attention. Anything below the threshold gets a process, not a deep review.
Week 4: sit through one full month-end as an observer
Take notes. Don't issue directives. Don't suggest fixes. Don't rebuild anything. Just watch.
Watch where the team waits. Watch which reconciliations get done at 11pm on day 5 because they were left until last. Watch how revenue cutoff gets settled. Watch what the prior Controller's 47-tab Excel file actually does during the close (this is when you find out it's pulling deferred revenue from three different systems and applying a manual override that nobody documented).
By the end of week 4 you should be able to name your top 3 close blockers. In B2B SaaS they're almost always: (1) revenue cutoff and the deferred revenue waterfall, (2) manual reconciliations that should be automated, and (3) intercompany or multi-entity eliminations. If yours are different, that's interesting and worth pausing on.
The line to use when the CFO asks for a faster commitment in week 3:
"I'll have a credible plan by day 90. I'd rather give you a number I can hit than one that sounds good in week 3 and slips in month 5."
Most CFOs respect that. The ones who don't are telling you something useful about how the next year is going to go.
Days 31-60: pick three fights, win three fights
Month two you ship things. Three things, specifically. Not five, not seven. Three. Pick fights you can actually win in 30 days, because a visible win in month two is the foundation of the credibility you'll spend in month three.
Fight 1: cut one day off the close
One. Day. Not five. The easiest day to cut is almost always a calendar problem, not a work problem. Look for the sequential-tasks-that-could-run-in-parallel pattern from week 2. The classic move: AP cutoff and bank rec running serially when they have no actual dependency. Run them in parallel and you save a day.
If you can't find a parallelization win, look for a task that's scheduled for day 4 but has all its inputs ready by day 2. Move it earlier. If a controller's review is happening on day 7 because that's when it's always happened, but the schedule it's reviewing is locked on day 5, you just found two days.
Don't try to compress the whole close. Cut one day, prove it sticks for two months, then cut the next one.
Fight 2: automate one reconciliation
The bank rec is the usual first win because most ERPs already have a bank-feed connector and the rec logic is mechanical. If your bank rec is being done manually in Excel against a downloaded statement, your accounting platform almost certainly has a connector that does 80% of the work. The remaining 20% is exception handling, which is what the team should be spending their time on anyway.
If the bank rec is already automated, look at high-volume AR rec, meaning applying customer payments against open invoices. Cash application tools have gotten meaningfully better in the last two years and the ROI is fast.
Pick one. Automate it. Document the new process. Show the time savings on a slide.
Fight 3: fix one control gap
Look at the open audit findings from last year. Pick the one with the easiest remediation, not the highest risk. (You'll get to the high-risk ones in your H2 plan.) The two easiest categories of finding to close are:
- Segregation of duties. Usually solved by re-permissioning the ERP so the person who creates the journal entry isn't the person who posts it. This is mostly a config change.
- Review evidence. The journal entry was reviewed but there's no signed evidence of the review. Solve this with a workflow tool (or a checkbox in your ERP) that captures the reviewer's name and timestamp on every JE above a threshold.
Pick one. Close it. Document it for the audit lead so they can mark it remediated in their workpapers.
Start a weekly close-blocker standup
15 minutes, every Tuesday, with the close team. Three questions: what slipped last week, what's at risk this week, what do you need from me. The point is not to micromanage. It's to surface issues in days instead of at month-end. The cost of a blocker that surfaces on day 3 is half a day. The cost of the same blocker surfacing on day 10 is a missed close.
Days 61-90: own the metric, earn the mandate
Month three is when you go public. The diagnostic phase is over, you have three wins to point to, and the team has watched you operate for two months. Time to commit.
Take ownership of close-speed publicly
Put close-speed on the CFO's monthly dashboard. Not buried in a finance ops report. Put it on the dashboard the CFO actually opens. Two numbers: actual close days this month, plan for next month. That's it.
Owning the metric publicly does two things. It aligns the team around the number you care about, and it tells the CFO that you're the person they'll come to when the board asks. Both matter.
Present the 90-day report to the CFO
The 90-day report is a document, not a slide deck. Five sections, three pages.
- What I found. The diagnosis from days 1-30. Top 3 close blockers, material account list, current-state close calendar with bottlenecks marked.
- What I fixed. The three fights from days 31-60. Quantified: 1 day cut, 1 rec automated (X hours/month saved), 1 control gap closed.
- What's still broken. The 47-tab Excel file, the other open audit findings, the revenue cutoff process, intercompany. Don't soften this. The CFO needs to see you've named the real problems.
- The H2 close plan. Specific target (e.g., 12 days to 7 days by end of Q3), specific changes required (headcount, tooling, process, system), specific risks.
- What I need from you. Budget, headcount approval, audit committee air-cover for the high-risk findings, executive sponsorship for the cross-functional pieces (revenue cutoff usually needs Sales Ops cooperation).
Get explicit CFO sign-off on the H2 plan
This is the document that protects you in month six when something gets reprioritized, or in month nine when the board asks why close is still 8 days. "We agreed to 7 by end of Q3 with these three dependencies, and dependency two slipped because we deprioritized the ERP work for the financing round" is a defensible answer. "Things have been busy" is not.
Get the sign-off in writing. An email reply works. A signed memo works better.
The real-world traps
The open audit findings trap. Don't promise to close all of them in 90 days. You won't, and you'll burn credibility trying. Pick the riskiest one (or the one your audit lead is most worried about) and document a remediation plan for the rest, with target dates. Auditors care more about a credible plan than a fast close.
The 47-tab Excel reconciliation trap. Do not rebuild it in month one. Document what it does, who relies on it, what feeds it, and what would break if it disappeared. Replacement is a Q2-Q3 project. The Controllers who try to replace it in month two end up with a half-built replacement, an Excel file that's out of date, and a close that's longer than when they started.
The "fast win" trap. A process change that breaks a control is not a win. Run every change past your audit lead before you ship it. If you're cutting a review step to save time, ask whether that review step is what your auditor relies on. If it is, the time savings cost you the audit.
The "I'll fix culture later" trap. Two months in you'll notice the close team is exhausted, the Sr Manager is interviewing elsewhere, or the AP lead is doing 60-hour weeks every close. Don't wait until month four to deal with it. Visible burnout in your close team is your problem from day one. The H2 plan should include people, not just tools.
What "good" looks like at day 90
Here's the scoreboard. If you can answer yes to all five, you've had a successful first 90 days.
- The close is at least 1 day shorter than when you started, and it stuck for at least two months.
- One reconciliation that was manual is now automated, with a documented process and quantified time savings.
- One open control gap is closed and the audit lead has signed off on the remediation.
- The CFO has a written H2 plan from you with a specific close-speed target and explicit dependencies.
- The audit lead is not surprised by anything you've done.
That last one is the most underrated. The audit lead being unsurprised at day 90 is the difference between an easy Q4 audit and a hard one. New Controllers who skip the audit-lead relationship in their first 90 days find out the cost in November.
You're not done at day 90. You've earned the right to start. The next 90 days are where the real plan executes — the rev cutoff overhaul, the second day cut, the audit findings you couldn't get to in Q1. But they only happen if the first 90 went well.
Diagnose first. Fix second. Commit third. In that order.
