Why You Keep Checking Your Work Over and Over

Abstract

This article answers a question many finance professionals ask late in the evening: why do I keep checking my work over and over? Four pieces of research bear on it. In two sets of experiments, healthy volunteers who repeated the same check became less confident in their memory of it, although in the first set their accuracy did not change. A meta-analysis of 43 studies linked the worry about mistakes to burnout more strongly than the drive for high standards. A review of perfectionism at work found no clear overall link with job performance. The three steps at the end are practical readings of this research, not tested methods, and they are not advice on accounting, audit or clinical matters.

The reconciliation that already ties

It is 6:40 p.m. on a Thursday. The bank reconciliation for the quarter has tied to zero for the third time. You agreed the ledger to the bank statement, then the open items to the ledger, then the ledger to the bank statement again. The file goes to your controller at 9 a.m. and the floor has emptied.

You open it a fourth time. Nothing has changed since the third. You could not say what you are looking for, only that you cannot send it yet.

It does not always look like this. You may reopen a file after you have sent it. You may re-foot a column you footed an hour ago. You may look in your sent folder to see that the email really went, or ask a colleague whether they got it.

How to Tell Your Boss You Made a Mistake in the Numbers is about what to do when an error does get through. This article is about the other pattern: the file ties, and the checking carries on.

Why do I keep checking my work over and over?

Research points to two things that can feed each other. The first is the checking itself, which can leave you less sure of what you checked. The second is the worry behind it, which is closer to fear of mistakes than to a high standard. A third question, whether the extra checking makes the work better, has no clear answer in the research cited here.

None of these studies watched accountants check files. Each section below says what was studied, and what that does and does not tell you.

What repeated checking does to your confidence

Van den Hout and Kindt (2003) ran three experiments with healthy volunteers, 39, 40 and 40 people. In the first two, volunteers repeatedly checked a gas stove in a computer animation. A comparison group checked virtual light bulbs instead. The stove group's memories of the stove became much less vivid and less detailed, and their confidence in their memory fell. Their memory accuracy was unaffected. A third experiment, without a pre-test, gave the same pattern. The authors suggest that checking can be a counterproductive way of staying safe, because it increases doubt.

Radomsky, Gilchrist and Dussault (2006) moved the idea into a real kitchen. Fifty undergraduates went through a routine of switching a device on, switching it off and checking it, for 19 trials. One group used a stove and the other a tap. Everyone then did a final check of the stove. Repeated checking of the stove lowered the volunteers' reported confidence in their memory, and made it less vivid and less detailed. Repeated checking of the tap did not. The authors argue that this memory distrust can in turn drive more checking.

Read the Thursday evening through this. Nothing in the file changed between the third check and the fourth. What can change is how sure you feel that you did the check at all. The thinner that memory, the stronger the pull to do it again. This is an inference. A bank reconciliation is not a gas stove, and the volunteers were healthy people and students doing short tasks. The studies measured memory of an action, not the accuracy of a figure, so they cannot tell you whether your file is right.

What the worry behind it does

Hill and Curran (2016) pooled 43 studies, with 9,838 participants, on how perfectionism relates to burnout. They looked at two sides of it. Perfectionistic strivings are the drive for perfection and very high personal standards. Perfectionistic concerns are worry about making mistakes, fear of negative evaluation and a sense of falling short of your own expectations.

Strivings had small negative or non-significant links with burnout. Concerns had medium-to-large positive links with burnout overall. Most of the pooled studies measured people once, so the authors say the findings show possible links, not established causes.

Harari, Swider, Steed and Breidenthal (2018) reviewed research on perfectionism at work. Their abstract reports an equivocal overall relationship with job performance, and sizable and consistent relationships with several other work-related factors.

Read the Thursday evening through this. A fourth pass over a file that ties looks less like a high standard and more like worry about being found wrong, and the worry is the side the meta-analysis links to burnout. The review does not show that perfectionism makes work better. This is an inference. Neither study counted rechecks, and none tested whether a fourth pass catches what three passes missed. The studies covered work, sport and education, so they are not specific to finance.

How to stop checking your work over and over: three steps

None of these is a cure, and none was tested on rechecking. They aim at three things the research points to: a check with no end, a memory that thins with each repeat, and the worry behind it.

1. Decide what "checked" means before you start. Write the checks and the exit at the top of the file, while you are fresh. For instance: "1. Ledger agrees to bank statement. 2. Open items listed with owners. 3. Difference is zero. When all three are ticked, the file goes." Without a list, the feeling of being sure decides when you stop, and in the studies repeating a check wore down confidence in it. This is an inference, and no study cited here tested it.

2. Log the check instead of repeating it. When a check is done, write one line: what you checked, the result, the time. For instance: "Ledger to bank statement, difference 0.00, 18:12." When the urge to open the file again arrives, read the line. A written line does not get thinner with each repeat, as the memory of the last check did in the studies. It is an inference that this settles the urge, and no study cited here tested it.

3. Name what the check is protecting you from. Before you open the file again, write one sentence: what do I expect to happen if this is wrong? For instance: "My controller finds a difference and decides I am careless." Then ask whether the sentence is about the file or about you. A fact about the file has a check that answers it, and you have done it. A fear about you has no check that answers it, which is why another pass does not settle it. This moves the question from the checking to the worry behind it. It is an inference that naming the worry loosens its hold.

Questions People Ask

How do I stop checking my work over and over?

Start small. Pick one low-risk file this week, write the list and the exit from step 1, and close the file when the list is ticked. Expect the urge to be loud when you leave it closed. That discomfort is not evidence of an error. No study cited here tested this, so it is a judgement. If the figure is going to an auditor, a lender or a regulator, agree with your reviewer what "checked" means for it and follow your company's review rules.

Is it bad to double-check my work in finance?

No. A planned check against a list is part of the job, and so is a second reviewer. The trouble starts with a check that has no list and no end. A repeat of the same check on an unchanged file adds less than a different check would, such as another source or another method. No study cited here counted how often a fourth pass finds something.

Is checking my work over and over a sign of OCD?

Not necessarily. Many careful people check. The Health Service Executive lists checking as a common compulsion in obsessive compulsive disorder, and describes compulsions as actions taken in response to anxious thoughts that bring short-term relief but tend to increase anxiety over time. What matters is how much of your life the checking takes and whether you can stop. This article cannot tell you which it is. A doctor can.

What if I find a real error on the fourth check?

Then the fourth check found something, and it is worth asking why the first three did not. Add that step to the list from step 1. The aim is a list with no gaps, not a higher number of passes. If extra passes often turn up errors, ask what sits behind them. Rushed deadlines are one possible cause, and How to Push Back on an Unrealistic Deadline in Finance covers the conversation before the work starts.

When it is more than a habit

Checking is a habit when it ends. It is more when it takes hours, or when you cannot stop even though you know the file is right. It is more when it spreads to things outside work, such as locks, appliances or sent emails, or when sleep, mood or health are suffering. In those cases talk to your doctor. Coaching is educational support for work patterns. It is not treatment and not a substitute for it.

If the checking sits inside a longer run of exhaustion, Burnout in Finance: Why High Performers Hit the Wall describes what that can look like. If you have tried to reason your way out of the loop, Why High Performers Cannot Think Their Way Out of Burnout is about the limits of that.

Sources

• Harari, D., Swider, B. W., Steed, L. B., & Breidenthal, A. P. (2018). Is perfect good? A meta-analysis of perfectionism in the workplace. Journal of Applied Psychology, 103(10), 1121–1144. DOI: 10.1037/apl0000324

• Health Service Executive (Ireland). Obsessive compulsive disorder (OCD) symptoms. hse.ie/mental-health/issues/ocd

• Hill, A. P., & Curran, T. (2016). Multidimensional perfectionism and burnout: A meta-analysis. Personality and Social Psychology Review, 20(3), 269–288. DOI: 10.1177/1088868315596286

• Radomsky, A. S., Gilchrist, P. T., & Dussault, D. (2006). Repeated checking really does cause memory distrust. Behaviour Research and Therapy, 44(2), 305–316. DOI: 10.1016/j.brat.2005.02.005

• van den Hout, M., & Kindt, M. (2003). Repeated checking causes memory distrust. Behaviour Research and Therapy, 41(3), 301–316. DOI: 10.1016/S0005-7967(02)00012-8