The 80/20 Close: Which 5 Tasks Cause Most of Your Delays

After running close for three different companies over 15 years, I've stopped believing the story teams tell themselves: that the whole process is slow. It almost never is. When I actually timed our tasks, roughly 80% of the delay came from five recurring choke points. Everything else moved fine. If you want a shorter close, you don't need to overhaul the entire process. You need to find your five and attack those.

Why the Whole Close Isn't Actually Slow

The first close I ever managed took nine business days. Leadership wanted five. My instinct was to push everyone to work faster across the board. That did nothing except burn out my team and produce more errors.

What finally worked was mapping the critical path: the specific sequence of dependent tasks that determines the total length. Most close tasks run in parallel and finish well before day-nine. Only a thin chain of dependent tasks actually drives your end date. Shave time off a task that isn't on the critical path and you save nothing.

When I logged start and finish times for every task over two consecutive closes, the pattern was obvious. Five tasks accounted for six of our nine days. The rest were noise.

The Five Tasks That Usually Cause the Damage

1. Bank and Intercompany Reconciliations Waiting on Data

Our bank rec wasn't slow because the work was hard. It was slow because it couldn't start until the last transaction posted, and we didn't know that had happened. The staff accountant would check, find something missing, wait, check again. That polling behavior cost us most of a day.

Intercompany is worse. Entity A books an amount, Entity B books something $4,200 different, and now two people are emailing screenshots trying to find the break. On a multi-entity close I ran, intercompany mismatches alone added a full day every single month.

2. Accruals That Depend on Someone Outside Accounting

The classic bottleneck. You can't book the marketing accrual until the marketing director tells you what was spent, and they're not thinking about your close. I've watched a $180,000 accrual estimate sit for two days waiting on one Slack reply.

These tasks are dangerous because the delay is invisible in your task list. The accountant is "ready." The blocker is a person who doesn't report to you.

3. Flux Analysis and the Review-Comment Loop

Variance explanations get kicked back. The controller asks why travel expense jumped 40%, the preparer investigates, responds, and the reviewer is now in another meeting. Each round trip is half a day. Two rounds and you've lost a full day on one schedule.

4. Manual Journal Entry Preparation and Approval

Recurring entries that someone rebuilds from scratch each month. Depreciation, prepaid amortization, payroll allocations. I've seen teams spend three hours re-keying entries that were 95% identical to last month, then wait on an approver who has no context for what changed.

5. The Consolidation and Financial Statement Build

This one sits at the very end of the critical path, so every upstream delay lands here. If your consolidation lives in a workbook with 40 linked tabs, one broken reference or a hardcoded number someone "just fixed for this month" can cost hours of tracing.

How to Find Your Specific Five

Don't guess. For two closes, capture three data points per task:

• When the task became ready (all inputs available)
• When work actually started
• When it finished and got approved

The gap between "ready" and "started" is queue time. The gap between "started" and "finished" is work time. In my experience, queue time and rework are the real villains, not the work itself. If a reconciliation is ready at 9 a.m. and nobody touches it until 3 p.m., that's your problem, and working faster won't fix it.

Why Excel Hides Your Bottlenecks

Here's the uncomfortable part. A close checklist in Excel makes this diagnosis nearly impossible. A spreadsheet tells you a task is "done" or not. It doesn't tell you when it became ready, who it was waiting on, or how many times it bounced back in review. That timing data is exactly what you need, and Excel doesn't capture it unless someone manually logs timestamps, which nobody does reliably during a stressful close.

Excel also breaks down on dependencies. When Task C can't start until Tasks A and B finish, a workbook has no way to notify the Task C owner the moment they're unblocked. So people poll. They check email, they check the tab, they ask in standup. That polling is pure queue time.

A structured close platform like ComplyBar changes the diagnosis because it records the timeline automatically. Every task carries its ready time, start time, completion time, and review history without anyone logging it by hand. When you want to know why the close took seven days, you're not reconstructing it from memory, you're reading it off the actual data.

The dependency handling matters just as much. When the last bank transaction posts and the reconciliation is unblocked, the owner is notified. The variance comment lands with the preparer instead of sitting in an inbox. Recurring journal entries roll forward with prior context attached, so the approver sees what changed instead of reviewing blind. Each of those removes queue time from the critical path, which is the only time that actually shortens your close.

What to Do Monday Morning

Pick your worst month and reconstruct the timeline of your five slowest tasks. Rank them by how much they delayed your final numbers, not by how much effort they took. Then attack them in order:

• Push blockers earlier. Get accrual inputs the week before close, not during it.
• Cut review round trips by defining variance thresholds up front so preparers explain anything over, say, 10% and $25,000 before it gets reviewed.
• Standardize recurring entries so nobody rebuilds them.
• Make dependencies automatic so "ready" and "started" converge.

When I did this, our nine-day close dropped to five in two months. I didn't make anyone work faster. I stopped the waiting. Find your five, and you'll get the same outsized return from a small, focused effort.