You set the goal in January. You find out in October that it was never going to happen. Nothing in between told you, because you only checked the one number that mattered once, at the end, when there was nothing left to do about it.
I have spent a year building a system to stop this, and describing it as four layers. Lately I think the layers were never the point. The parts matter, but the thing that makes them work is that they run as loops, at different speeds. And the reason so many goals fail quietly is that most people run only one loop, at the slowest speed there is.
One loop, and it closes too late
The default way to chase a goal is a single loop. Decide what you want, do the work, check at the deadline whether you got there. It feels complete because it has a beginning and an end. It fails because the only feedback it produces arrives when the period is already over.
A loop that slow can tell you that you missed. It cannot tell you in time to do anything. By the time it closes, the quarter or the year has gone, and all you have learned is the score. This is not a discipline problem, and treating it as one is why people respond by resolving to try harder next time, on the same single loop, at the same fatal speed.
Loops at three speeds
The fix is not one better loop. It is several, running at different speeds, because each speed catches a kind of error the others cannot see.
The fast loop is daily and simple. It checks one thing: whether the work happened. Whether the runs got done and the calls got made. It closes every day, costs almost nothing, and catches the most basic failure, which is not showing up. But it has a ceiling. A perfect daily loop tells you that you were consistent and nothing more.
The middle loop is weekly, and it is the one most people are missing. It asks whether the work is producing anything. Whether the easy pace is dropping, whether the booked-call rate is climbing, whether the thing the plan promised would move is actually moving. This loop catches the expensive failure the daily one cannot see, the plan that is being executed faithfully and not working. You can only read it at a week's distance, because a day is too noisy to trust and a month is too late to act.
The slow loop is quarterly, and it is the deepest. It asks whether the goal was the right one. You can execute the plan, watch the indicators move, and still find the objective no longer matters, or never did. That only becomes visible across months, and it is the loop people skip most often, because it can tell you that a quarter of honest, well-tracked work was aimed at the wrong thing.
Each loop is blind to what the others catch. The daily loop cannot see a failing plan. The weekly loop cannot see a wrong goal. The quarterly loop cannot catch a plan going wrong in week three. Run only one and you are protected against one kind of failure and exposed to the rest.
Speed is the whole point
The loops are not just different lengths. The length is what makes each one work, and running a loop at the wrong speed breaks it as surely as not running it at all.
Check the daily question weekly and you miss two days you could have saved. Read the response signal daily and you react to noise, adjusting a plan that was fine. Ask whether the goal is right every week and you churn through objectives you should have held for a quarter. A loop has a speed at which its signal is real, and the discipline is matching the loop to the thing it is trying to catch, not running everything at whatever cadence feels productive.
What this changed for me
I used to think tracking was storage. A place to keep the numbers so I could look at them later. Describing the system as four layers fit that idea, four kinds of thing to record.
Loops have changed what I think tracking is for. The point was never to hold the data. It was to close each loop at its own speed, to make the signal come back while there is still time to use it. A number you record and never feed back is not tracking. It is a tidier way of finding out in October.
The goal you missed was probably not a failure of effort. It was a loop that closed too late to matter. Consistency was never the whole problem. The problem was that you had one loop, running once a year, and by the time it came round the year was already spent.