Planned against actual: what to do with your statistics

Every time you close a block you generate a piece of data no other planner has: what you meant to do, next to what you did. This article explains how to read it so it makes you plan better rather than feel worse, which is the easy outcome.

Where the data comes from

From the closures, and from nothing else. When a block ends you answer complete, reschedule or cancel, and that is all you have to do: there is no report to fill in and no sheet to maintain.

That sounds like a detail and it is the reason Dilo's statistics mean anything. Any system that asks you to log your time at the end of the day is asking you to remember, and you do not remember: you reconstruct. And when reconstructing, everybody rounds in their own favour. The question at the end of the block arrives at the one moment when the answer is exact, and it costs one tap.

You can look at the last 7, 30 or 90 days. The first seven are noise: do not draw conclusions before the second month.

The completion rate

What percentage of what you scheduled ended up completed. It is the most visible number and the easiest to misread, so let us start with what it is not: it is not a grade. There is no good number to reach.

The 90 % deserves its own paragraph, because it is the most common trap. When a number is watching you, the natural move is to start scheduling only safe things: lunch, the gym you already did, the meeting that was going to happen anyway. The percentage goes up, it feels good, and your planner stops telling you anything at all about yourself. A measurement system that can never give you bad news is not a measurement system.

Hours planned against hours kept

The other side of the same coin, and the one that corrects you fastest: how many hours you put into the day and how many actually came out.

Almost everybody discovers the same thing the first time: they plan between eight and ten hours of blocks and keep four or five. That is not a lack of discipline, it is an estimation error so common it has a name in the psychology literature: when estimating how long we will take, we imagine the version with no setbacks, and the version with no setbacks never happens.

The side effect is the one you are after: when the day you plan resembles the day that happens, you stop going to bed feeling behind. That feeling almost never came from having done little. It came from having promised yourself too much in the morning.

Your worst band of the day

The statistics split your blocks by band and show you the completion rate of each. It is the metric most people ignore and the one that improves days fastest.

There is practically always one band dragging the result down, and practically never the one you would name. Everybody thinks they know their best hour, but what you remember are the exceptional days, not the average of sixty.

What you do with the data is logistics, not morality:

• The important thing moves to your best band. Full stop. There is no virtue in doing the hard thing at the hour where you have been failing for two months.
• Your worst band gets mechanical work: email, reviewing, sorting, errands. Things that come out even when done badly.
• If your worst band is after lunch — by far the most common — consider putting the scheduled break or the walk there, rather than a battle you have already lost many times.

The uncomfortable list: what you cancel most

The tasks you schedule again and again and never do. It is the least pleasant metric to look at and, by a distance, the one that changes the most.

A task cancelled four times does not need more willpower. It needs a diagnosis, and there are only four:

1. It is too big «Do the tax return» is not a task, it is a project. Split it until the first piece is ridiculously small: «find the certificate and put it on the desk». If it does not feel ridiculous, it is still too big.
2. It is at the wrong hour Cross this task against your worst band. Very often there was nothing wrong with the task: you were putting it at four in the afternoon.
3. Something has to happen first You cannot «call the insurer» if you cannot find the policy number. The real block that was missing was finding the number. Schedule that one.
4. You are never going to do it It happens. Some tasks have been travelling from list to list for six months because you think you ought to want to do them. Delete it. The relief tells you whether you were right, and the cancelled list is the only place where these tasks are exposed — on an ordinary list they hide forever among the rest.

Once a month, take the most-cancelled task and apply one of the four. Just that one. It is probably the best use of twenty minutes you can make with this app.

The rule you cannot break

Never adjust what you plan so it matches what you did.

It is tempting and it is fatal. If you allowed an hour for something and it took two, there are two possible readings: «this takes me two hours» or «I'll put down two hours so it comes out green». The first is information. The second is makeup, and the moment you start applying makeup the numbers stop meaning anything and all you have left is a planner with pretty colours.

The practical distinction: adjust the duration because you have learnt how long you take, not because you want the percentage to go up. They sound identical and they are not, and only you know which one you are doing.

The same goes for cancellations. Cancelling is not a failure and does not break any streak: it is an honest answer that closes the matter. What is a problem is leaving blocks unresolved, because a block in limbo is neither data nor anything else, and limbo is what makes planners get abandoned.

The twenty-minute review

Once a week, always at the same hour. It gets scheduled like everything else:

Four questions, in this order, and none of them is «did I do well?»:

1. How many hours did I plan and how many did I keep? If the gap is large, next week I plan the figure I kept.
2. What was my worst band? I move one single thing. One.
3. What task tops the cancelled list? I apply one of the four diagnoses above.
4. What went well? It sounds like self-help and it is pure method: if you do not identify what worked, next week you will change it without noticing.

Twenty minutes, four questions, no judgement. The goal is not to become a more disciplined person: it is that in three months, the day you plan at night resembles the day you are about to live.