Late with today’s post, sorry (not that I expect anyone’s waiting on these lol)
#81: “Futility Is Proof that You Aren’t Doing Enough”
Published in 2017/04/06 issue
In the last column, I wrote about how making accurate estimates for your work is a form of kindness towards others. This time, I’d like to press further into how you can make your estimates more accurate.
When I write this column for Famitsu, I estimate that it’ll take a minimum of 1.5 hours and a maximum of 4 hours. Since I’m on the 81st column now, having this much past experience means that my estimation is more or less fact. However, I wouldn’t be able to make such an accurate estimate for something I was doing for the first time. I’m sure everyone has experienced a task taking longer than expected.
When I first started writing this column, I thought I’d be able to do it quickly, so I set the minimum time to 1 hour. I couldn’t imagine spending more than 2 hours on it, so I set that as the maximum time. In the end, it wasn’t that easy.
In that case, is there any other way to increase accuracy besides actually performing the task and using the time taken as feedback…?
What I recommend is doing a detailed task breakdown. As the name implies, this means taking a single task and breaking it down into several smaller tasks. Let’s use writing this column as an example. First, we break it down into the most detailed tasks that we can:
Task Title: Writing a Column (2,500 Characters Average)
(Broken down into 6 tasks)
- Turn on PC and open column manuscript file (starts up quickly because it’s on an SSD. 1-3 minutes)
- Think of an idea for the column (min. 2 minutes if I have an idea, max. 15 minutes if I have to think about it)
- Write around 2,500 characters (1 hour if I write 41.66 characters per minute, 3 hours if I write 13.8 characters per minute)
- Think of a title for the column (3-10 minutes)
- Proofread the entire text (5-30 minutes)
- Take a break in the middle (0-30 minutes)
When we look at it like this, we can see that the singular task of “writing a column” is actually made up of 6 tasks (or perhaps it would be more accurate to say that we define it as 6 tasks). When you’re doing this task breakdown, you should purposely make it overkill. Now, let’s add up the minimum and maximum times for the 6 tasks above.
Adding up all of the minimum estimates gives us 1 hour and 11 minutes, while adding up all of the maximum estimates gives us 4 hours and 28 minutes. The nice thing about task breakdowns is that they become more objective as you do them. For example, for writing a 2,500-character column, we can identify fixed tasks that will always be required no matter the column’s content. So, regardless of whether it’s done spontaneously or with a lot of hard thinking, we know that we need to hammer out around 2,500 characters on the keyboard. Then, we can objectively ask, “How long does it take to type 2,500 characters?”, divide the character count by numbers of hours, and get a numerical visualization. Instead of thinking “something like this,” we now have a precise amount of time required.
When we divide 2,500 characters by 60 minutes, we realize that we’d need to write 41.66 characters per minute in order to reach that goal. That’s around 0.7 characters per second. We need to write 3 characters every 4 seconds. Quantifying it like this shows us that depending on the person, if they aren’t good at composing from their thoughts or if they type slowly, it may be impossible to write 2,500 characters in one hour.
If you compare estimating a broad task like “writing a column” as one task to breaking it down into smaller tasks and estimating each individual one, the latter is obviously going to be more accurate. However, surprisingly many people don’t come to this conclusion. They don’t consider that their tasks are composed of many smaller tasks. When I first started writing these columns, I missed tasks 2, 5, and 6, so my estimate fell behind.
Task #2: even if I already have an idea for the column, I have to spend some time imagining the rough outline before I start writing. If I don’t have an idea for the column, then I have to think of one. Topics that don’t naturally come to mind probably won’t be very interesting to write about, so I give myself 15 minutes to get that flash of inspiration.
Task #5: I’m the kind of person that proofreads as I write rather than writing everything in one go and proofreading at the end. When writing is going smoothly, it doesn’t incur much of a proofreading cost, so sometimes I can include it with the writing cost in task #3.
Lastly, task #6. This is where the infamous “laziness” and “slacking off” come into play. Sometimes I don’t slack off at all, like when I concentrate and write the entire thing in one go, fueled by cigarettes. But usually, I get tired after 500 characters and start chatting in FFXIV which is open in the background.
Now, calling #6 a “task” is kind of iffy, but in reality, it makes sense because it’s something you should include in your estimates. Sometimes you get people who say, “No, I can’t include slacking off in my estimates! I finish my work diligently, so I don’t need a task like that!”, perhaps because they want to be praised by their superiors. You’re not going to be praised for saying that, and you’re definitely not going to maintain schedules that don’t have any leeway, so just relax. In my opinion, what deserves praise is when you finish your work as scheduled.
For tasks like #6, you can think of them like “if I don’t use it, then the task will proceed ahead of schedule.” Alternatively, that time can be used as insurance in case someone assigns you another urgent task. You rarely see people estimating their slacking time and incorporating it into their expectations, and some may think that putting this much effort into estimating times is a form of slacking off in itself. However, an estimate that gives no leg room is just begging to be broken by the tiniest of causes.
After failing to stay on schedule multiple times, they’ll start to think that making estimates is futile, because things aren’t going well regardless. But that’s just an excuse. Estimates are definitely not futile, and the reason you fell behind schedule in the first place was because your estimate was too optimistic. Even if you fail once, your estimation accuracy will rise if you fix the cause of the problem. Anyone who says that estimation is futile is either doing it wrong, thinking of it as a chore, or both… There are probably a lot of such people, but would you try giving thorough estimation just one more chance? *grins*