Combatting Our Unrealistic Estimate of Time
by Jim Freund
Granted, there are a lot of worse things to be concerned about when you’re close to hitting 90, but one personal shortcoming really perturbs me – my consistent underestimating of how long it will take me to complete a task I’m about to set out on. Given all those years of experience behind me, you would think I’d be better at this chore than I am, but you’d be wrong.
So that’s why I perked up recently when coming across two articles on the website zapier.com: “Why We’re Bad at Estimating Time (And What To Do About It)” by Jessica Greene, and “How To Fix Your Broken Perception of Time” by Ivana Isadora Devcic. It was refreshing to learn that I’m not alone in this regard, and that it’s possible – even at this late date – to improve my performance.
“Let’s face it,” Ms. Devcic began knowingly, “humans are notoriously bad at setting deadlines and estimating time. The comforting fact is that it’s largely not our fault.” I gratefully welcomed that judgment and was prepared to move on to the “How to Fix” segment promised in her title.
But although the article is interesting, it’s mostly concerned with our “Broken Perception of Time” and all the manifold reasons why. If this is something that interests you, by all means read the article; but for me, having quickly accepted her premise, I wanted to advance to the fixing part – which is just what Ms. Greene’s article has attempted to do.
She sets the mood by telling a tale of having painted five rooms in her house over the past few years. In each case, she had been convinced that the room would be finished in a week or two, but it never took less than a month. Yet she admits that, even for rooms painted subsequent to that first eye-opening experience, “I couldn’t let go of my confidence that the next room would be easier and faster to paint.” But it never was.
It turns out that the false confidence leading to her unrealistic estimates has a name – the Planning Fallacy – “something we all occasionally suffer from when estimating how long tasks and projects will take to complete.” I felt better already, having been linked with the vast number of others who suffer from the same malady.
It seems that these estimation mistakes can primarily be attributed to two key factors:
Failing to consider how long it has taken us to complete similar tasks in the past; and
Assuming that this time we won’t run into any complications that will cause delay.
This second factor is a result of what’s termed “optimism bias” – our tendency to believe that the future will be better than the past. Upsetting as it may sound, we’re inclined to estimate for a best-case scenario, while disregarding historical data that proves the best-case scenario is unlikely.
So Ms. Greene warns us to stop estimating by intuition only. Instead she offers six techniques to overcome the Planning Fallacy and avoid underestimating the time required.
Before turning to these six, however, it occurred to me (having never been known to paint even one room) that it might be well to have a more typical model in mind for this catalog of timing issues – something that actually happens in my life on a recurring basis, so that I can measure the degree of optimism I’ve mistakenly applied to the latest instance. (Here I encourage readers to summon up their own bugaboo.)
In my case, a certain car trip I make on a regular basis – the drive in from our CT house to our NYC apartment after each weekend – seems like an ideal candidate for estimating how long it will take this next time. It contains a number of measurable elements – how much traffic I’m likely to encounter, roadwork that might cause a delay, whether I’ll need to make a pit-stop along the way for gas or to pick up coffee and donuts (or, most often, to relieve myself), whether it’s likely to rain or snow and how much time that would set me back . . . . Hey, wait a minute . . . the car’s electronic gizmo that provides route directions also includes an estimated time of arrival at your goal – maybe I should just peek at that in advance . . . .
Okay, here’s a better one. I’m dining at a restaurant (one of my favorites, which provides factual input from past such forays). Later that evening there’s a basketball playoff game on the tube that I want to watch from the first quarter on. So the question is, when will I be finished with dinner in order to race home in time. Oh, boy, here’s some good stuff – will the restaurant be busy or fairly empty, what kind of service might I receive, if I’m out with other geezers, will they waste a lot of time participating in an organ recital and then imbibing before ordering dinner, are they likely to insist on fancy desserts and coffee, is rapid transportation to my house available from the restaurant . . . – Hey Jim, you jerk, tape the goddamn game and stop worrying about whether you’ll make the tip-off . . . .
Hmmm . . . . this isn’t as easy as I thought it would be . . . Well, probably the best one for me is how much time I’ll need in the morning to be ready to face the day – how long will my various ablutions last, what about breakfast, will I receive an urgent phone call, etc. etc. Everything seems to take so goddamn long for an old fogy like me nowadays. And even though I invariably underestimate, I find it hard to face up to this in advance. Is it my optimism bias kicking in, or just incredulity at how slow I move nowadays in the morning. . . ? (Actually this has the makings of a good standalone topic for a later blog post . . .)
Okay, enough of that stuff. Here are Ms. Greene’s six keys to success in estimating time.
1. USE HISTORICAL DATA
This makes eminent sense, but sad to say, it requires you to actually collect the data you’ll be using to create the future estimates. I’m not sure I’m up to this (although I’ll be the first to admit it would likely provide some protection against completely unrealistic predictions).
Those of you who aren’t daunted by the need to keep historical time might want to read another article on zapier.com by Erin Greenawald titled “Time Tracking Experiment: What I learned after analyzing every minute of my life for 30 days.” It contains discussions of such matters as determining your ideal allocation of time, being as specific as possible, labelling your time while categorizing it, deciding where lost time goes, and analyzing the results. Good stuff, I’m sure, but those of you who know me are unlikely to be surprised at how quickly I skipped over the details here . . . Well, maybe the next key will prove more helpful . . . .
2. HAVE SOMEONE ELSE ESTIMATE FOR YOU
I like this one, but WHAT A COP-OUT! The theory behind asking a friend to help is that evidently people are much more pragmatic when figuring out how long it will take someone else to complete the task. Were I to attempt this, however, with my luck the friend would likely respond: “Sure, Jim – and there are also a few things I’ve been meaning to ask you to do for me . . .”
3. ESTIMATE IN RANGES, OR BUILD IN TIME FOR DELAYS
I admit having found it a little off-putting when the author began this point by referencing a well-known Donald Rumsfeld quote, the gist of which is:
There are known knowns; there are known unknowns; and there are also unknown unknowns – the ones we don’t know we don’t know.
The point is, I guess, that you need to take all of these into account when estimating how long a task or project will take – and in particular, things you never previously considered could happen. Ms. Greene illustrates this by diagramming a “Cone of Uncertainty” – a helpmate that I imagine may prove useful, but one I’m just not going to get into.
4. USE THREE-POINT ESTIMATIONS
Here’s a technique that forces you to confront your optimism by identifying three estimate scenarios: best case, worst case, and most likely. The best case is our old optimistic self; the most likely is often based on historical data gathered from similar past tasks; and the worst case is how much extra time it might take if everything goes wrong. Then you take the average of them (actually this isn’t so difficult – just add the three numbers together and divide by three), and presto – there’s your estimate.
The ticklish part here is to identify all the things that could go wrong for inclusion in the worst case. Those with a negative bent of mind might have no trouble with this – in fact they might enjoy the exercise – but then the three-point average may be inexorably inclined toward the troublesome end. Meanwhile those of us with an optimistic bias might omit some of the possible problems (so that the worst case isn’t really that bad), thereby tilting the average in the trouble-free direction . . . . Hey, I know – get ahold of that neutral guy you used to estimate for you in the #2 Key to Success and have him select a friend to share the estimating here – one of them on the best case and one on the worst – they’re likely to balance each other out . . . .
5. CALCULATE YOUR FUDGE RATIO
This one is interesting. Here Ms. Greene refers us to Steve Pavlina, who has written a book on the subject; I compromised by reading an article he authored on his website (stevepavlina.com), titled “How to Make Accurate Time Estimates”.
This is how it goes. You estimate (by intuition alone) how long several upcoming tasks will take. Then you track the actual time spent on them. Add the estimated values together; then add the actual times together; divide the sum of the actuals by the sum of the estimates. What you end up with is what Mr. Pavlina calls a Fudge Ratio.
So let’s say your estimates average 40 percent lower than the actual time needed. Then multiply your initial estimate on a new job by the Fudge Ratio of 1.40 in order to come up with a realistic estimate.
Mr. Pavlina admits that, for individual tasks, your Fudge Ratio can vary widely, but maintains that for groups of tasks that collectively require a few days to complete, the Fudge Ratio will settle into a fairly narrow range. He also concedes that knowing his own Fudge Ratio has not made his initial estimates more accurate, but then points out that using the Fudge Ratio does help solve the problem. He advises you to recalculate your Fudge Ratio periodically – “once a quarter should be fine” – and update it if the nature of the tasks you undertake changes notably.
6. ESTIMATE DURING THE LOW POINT OF YOUR DAY
This one is worth noting. Evidently people tend to be less positive when they hit their low point in the day – the so-called “trough” – and this negative creep will help in creating more realistic estimates. For most people, the trough is mid-day after lunch.
Ms. Greene has some good stuff about this in another zapier.com essay titled “Find Your Chronotype and Schedule Your Productivity”. First, though, I had to figure out what that big word means. Now I know – chronotype is the scientific term for what’s commonly referred to as an internal clock.
Apparently most people have a “peak” productivity period each day during which they’re most focused, least easily distracted, and thinking quite clearly – which makes it a “perfect time to work on analytical tasks.” For most of us, those are the two or so hours before lunch. Then you eat, after which comes the trough where you’re less energetic and focused.
Still, it’s not a complete loss – it’s actually better for creative tasks. Ms. Greene cites Daniel H. Pink’s book, “When: The Scientific Secrets of Perfect Timing,” for the proposition that creativity thrives when we’re more easily distracted, because distraction actually help us make connections we might not have made if we were highly focused.
There’s a lot more on this technique, focusing on three different types of chronotypes (Larks, Third-Birds and Owls) and “Building Your Personal Productivity Schedule,” but it’s lunchtime now and I think I’ll just leave things at that. After all, we’ve come a long way in a short span to fix our “broken perception of time”.
Still, maybe I should put in a little more work on my Fudge Ratio before lapsing into today’s trough . . . .