I'll be there soon, I promise.

November 1, 2005 11:23 AM

I have a tendency to set expectations I can't meet, usually in terms of time and scheduling. (When I can be there, how long it will take to finish this project, etc) How do I break this habit?

I look at a project, and I honestly, truly, sincerely believe "I can do this in 3 months!" And about 1 month in, I realize how wildly optimistic I was being, and how many unforseen variables I had not accounted for.

Worse, I'm known among friends and family for being habitually late. I'm not irresponsible or inconsiderate; I just regularly underestimate the time it takes to get from point A to point B, the time it will take to pack my suitcase, and the time it will take to stop at the post office on my way. Or, in my desire to please, I promise to be there at a time I know is unrealistic, and pray that the space-time continuum will somehow alter itself on the way. The fact that I'm usually overscheduled and overcommitted (a seperate issue, which I am also trying to curb) compounds the matter.

I'm not looking for time management help (I own Getting Things Done already); I need to learn how to strategically estimate time in my head, how to set expectations for variables I can't predict, and how to stop convincing myself that I can get through 2 toll plazas and 14 stoplights in 8 minutes.

posted by junkbox to human relations (13 comments total)

Think before opening your mouth - be absolutely, 100% sure yourself you can fulful a commitment before making it. And don't be afraid to say "No" to people.
posted by benzo8 at 11:31 AM on November 1, 2005

I do this sometimes too. I think it's part of having big expectations for yourself.

First, with regard to personal appointments, double whatever you think you need. Try this for awhile and see what happens.

Second, with regard to project planning, mid-course corrections are very useful. Don't be ashamed of them. Also, increase the level of detail for project planning and use past experience as a timeline. Then, build in a cushion.
posted by ewkpates at 11:37 AM on November 1, 2005

Here is how I as a Project Manager deal with people on my staff like you: I multiply any time estimate by 3. This is how you should operate in your life. Knowing that you are always late or chronically underestimating times, just take your gut and multiply by 2 or 3. Simple.
posted by spicynuts at 11:41 AM on November 1, 2005

Doubling any initial estimate is a good basic plan.

On the management side, however, be sure you aren't scheduling with the assumption that you'll be working 100% of the time that you're "working." We work a lot less than we think we do.
posted by callmejay at 11:42 AM on November 1, 2005

Try adopting an old business maxim - "Under-promise, over-deliver." Take a minute (like benzo8 suggests above) to think about your estimate, and then pad it by some arbitrary number. 50%? 70%? Maybe you could start logging your overages, look for a pattern, and pad your estimates accordingly. Assume that something (or everything) will go wrong on a project, budget and bid your time accordingly, and have pleased and surprised clients when you come in early. Referencing your example, honestly, truly, sincerely believe "I can do this in 3 months!" and then pad your estimate to 5 or 6 anyway. This, of course, get's a bit hairy if your in a competitive bidding situation where you want to look good against your competitors, but if you manage to establish a reputation as a conservative estimator, and consistently deliver early against your estimates, that will improve your strategic position (depending, of course, on how small your particular business community is).

Oh, and don't be afraid to say "no" to people, by which I mean don't let a desire to impress or please up front jeopardize your ability to impress or please at the finish.
posted by ZakDaddy at 11:44 AM on November 1, 2005

Worse, I'm known among friends and family for being habitually late.

Try getting to appointments 10 minutes early, and being prepared to wait. (If you did find yourself getting somewhere early, what would you do?)
posted by russilwvong at 11:51 AM on November 1, 2005

I seem to remember one person on Ask Mefi who detailed, in a thread of a different topic, a system they devised for tiem measurement that was highly effective.

It's really not that hard--if you want to apply yourself, I suggest doing this:

When you're doing little tasks around the house, time them--not with a stop watch. Just glance at the clock at the beginning and end to get a general time. Let's say--cooking instant curry. About 5 minutes. Good--now keep doing this for several tasks. Soon you will have a "stock" of "prefab" times, if you know what I mean. "__ takes about as long as a batch of curry/a cigarette/a tv show."

My big thing is time intervals, really--I'm not organised or habitually on time at all, but I do keep track of when things end and start and estimate them well too. When you have little systems you can compare--and once you drive from point A to point B not thinking how late youll be, but just trying to clock an average distance, and then give it padding, you'll be fine. It took 15 minutes? Tell the person a 20, or 25. If the situation allows for it, being early is even better.

So, yeah--try to make up little units of time in comparison to things you habitually do. Once you keep a system of comparisons, you'll do it more and more until you just know it's about 8 minutes to the store.
posted by Lockeownzj00 at 11:56 AM on November 1, 2005

A friend who is a general contractor has had your same problem. He knows exactly how many hours of work a project will consume. But of course, he can't accurately predict the weather, the reliability of other people. the availability of supplies, and so on.

It was driving him crazy because he wanted to be able to predict how long a project would actually end up taking. But that's just not possible.

He went back into his files and tried to come up with a formula, but couldn't.

So he did the "too drastic" thing: he assumed the worst-case delays for his next projects. He had to get over his rage at wasting time, and other things. But over time he got better at managing the down time. He never got better at estimating how long jobs would really take.

So get that out of your mind right now.
posted by wryly at 12:45 PM on November 1, 2005

I have two suggestions that go along with some of what others have said:
1. Regarding personal lateness - in your estimate, allow some time for "extras". For example, when I first moved to Seattle, I found that I was chronically late because I had so much trouble finding streets/exits/learning the geography. So I started building in a half-hour within which I could "get lost." This worked like a charm, sometimes I was early (I'd explore a bit), but I was never late. As I got better at this, I allowed less time - but I still ALWAYS build in about an extra 15 minutes for any appointments (this way I am ALWAYS on time).
2. Regarding project lateness - you say that you nearly always come across complications that cause your project to delay. I suggest that you document these as they come up, then when the project is complete, estimate how much of a delay they caused and note what kinds of things they were. Then, when you need to make your next estimate, look at what happened last time and build in that amount of time (it's the same sort of cushion as in #1). After a bit, you should find your estimates are more accurate.
I do believe that we can get better at estimating - once we acknowledge history (and those who don't are doomed to repeat it).
posted by dbmcd at 1:19 PM on November 1, 2005

Stop looking forward, and start looking back.

The best predictor of how long things will take next time, is how long they took last time.

You're going to drive to visit your family. Ask yourself how long the trip took last time. That's how long it will take next time, IF NOT LONGER.

You're commissioned to deliver a piece of work. Look at how long it took you to deliver a similar sized piece of work last time. That's how long it will take next time, IF NOT LONGER.

It's also obvious that you know your problem is that you are setting a desired deadline or time for yourself, and then rationalising how you can achieve it. Instead, you need to figure out what steps are required, their likely duration BASED ON PAST EXPERIENCE, add those durations up, and only then look at what the total time or estimated delivery date is.

If the answer is unacceptable, you can either forgo that undertaking, or look for steps you can omit. What you cannot do is start believing you can complete steps in less time than it took last time.

Also, stop thinking in terms of idealized performance. Eg, with your car trip, maybe you had a bad run of lights, maybe you needed to stop for fuel, maybe a parade with elephants blocked your route for 10 minutes. You may be tempted to think "that won't happen next time!" Well, tough shit. I guarantee some other, roughly equivalent thing will happen. That is why we call those perfect deviation-free experience "dream runs" - they only happen in your dreams. So when you look back at past times and durations, I want you to use the total time things really took, without factoring out things you think don't count.

Finally, if you fudge numbers out of a desire to please others, try to hold in your mind a picture of how DISPLEASED they will be when you are late, again.

Good luck.
posted by i_am_joe's_spleen at 1:28 PM on November 1, 2005

I haven't tried this much myself, but here's a formula I got in a time management seminar...

Te = [To + 4(Tm) + Tp] / 6

Te = Time Estimate
To = Time Optimistic
Tp = Time Pessimistic
Tm = Time Most Likely
posted by clh at 4:20 PM on November 1, 2005

Here's what I did back when I had my own consulting firm.

1. In a spreadsheet, make two columns: estimated and actual.

2. Go back through your projects and try to recall the estimate and actual times you had.

3. Using the spreadsheet, find your own Average Personal Misestimation Coefficient. (Mine was about 3.1 for web projects.)

4. *Force yourself* to use this coefficient in your planning, but keep on tracking the original estimated and final actual times. Over a few months, try to reduce your APMC. I got mine down to about 1.2.

Worked for me!
posted by josh at 5:10 PM on November 1, 2005

Rather than saying "no" to people, try saying "Let me get back to you on that." I often find that when I'm pressured to give an answer RIGHT NOW, I default to underestimating the time needed. If I can intentionally break the conversation, look at my files, make an estimate, tack on padding time, and then get back, there's less pressure to please. Sometimes, I'm also able to buy enough time to e-mail them, which also takes some of the pressure off (versus talking to them).

Ultimately, though, I'm in the same boat as you. I'm late and blowing deadlines all. the. stinkin'. time.
posted by Alt F4 at 7:56 PM on November 1, 2005