“Good, 4% it is.”
That is how it often starts. The analytics team builds a forecast that says sales will probably grow by around 4% next year, give or take a few points. The number goes into a slide, the slide goes to the management meeting and someone approves it. A week later, 4% is in the budget. A month later, it is the sales team’s quota.
Nobody decided it, but the forecast has become the target. And from that moment, it stops working as a forecast.
Two numbers, two different jobs
A forecast is an objective view of what is most likely to happen, based on what we know today. A target is what we would like to happen. One describes the road ahead. The other describes where we want to go.
Think of a weather forecast. It says there is a good chance of rain on Saturday. Your target is a dry wedding. You do not change the forecast to “sunny” because you want sunshine. You rent a tent.
Both numbers are useful. But they answer different questions, and they need to stay separate to do their jobs.
How a forecast quietly turns into a target
It rarely happens on purpose. Usually, it is a mix of small, everyday things:
- It is the only number on the slide. When people need a number to plan around, the forecast is the easiest one to grab.
- The uncertainty gets lost on the way. The analyst says “around 4%, most likely between 1% and 7%”. Two meetings later, it is just “4%”. The further a forecast travels from the people who made it, the more certain it sounds.
- Someone “just needs a number”. Finance needs a budget, operations needs a capacity plan. A careful, uncertain forecast quickly becomes a fixed input for someone else.
An uncertain forecast can very quickly become a certain target for someone else.
What goes wrong when they merge
Once people know the forecast will become their target, they start to treat it differently. This is a version of Goodhart’s law, named after the economist Charles Goodhart and often summed up as: when a measure becomes a target, it stops being a good measure.
In practice, it looks like this:
- Sandbagging. If my forecast becomes my quota, I have every reason to forecast low. A target I can beat looks better than an honest forecast I might miss.
- Wishful thinking. If management wants 10% growth, a 4% forecast can feel like bad news nobody wants to deliver. So the forecast is “adjusted” until it looks better.
- The early warning disappears. If forecast and target are the same number, there is no gap to look at. Problems only show up when the actual results come in, and by then it is too late to act.
- The wrong people get blamed. A forecast is judged by whether the target was hit, instead of by how good a forecast it was. Forecasters learn that honesty does not pay.
All of these have the same effect: the forecast stops carrying real information about the future.
It also works the other way around
Sometimes the target leaks into the forecast instead. This is very common when the same people make the forecast and are responsible for hitting it. A sales manager asked “what will we sell next year?” naturally thinks about everything the team will do to sell more. Hope slips into the number.
This is not dishonesty. It is a normal human bias, and it is very hard to see from the inside. It is easier to spot in short-term forecasts, where the results come back quickly. In long-term forecasts, it can hide for years.
Another bias to watch is anchoring. If people see a number first, their own estimate tends to stay close to it. That makes timing a choice. Show the sales team the statistical forecast first, and it can pull their numbers away from optimism, but you lose their independent view. Show it after, and you keep that view, but optimism has more room. And if they can see the target while they forecast, the target becomes the anchor instead.
The gap is the most useful number you have
Here is the good news. When you keep forecasts and targets apart, the difference between them becomes one of the most valuable signals in your business.
If the forecast is below the target, that is not a failure. It is a warning, early enough to do something about it. You can change prices, shift marketing spend, open a new channel or adjust the plan. Looking at the gap at different horizons (next quarter, next year and three years out) tells you where you need to act first.
This also means a good forecast can end up looking “wrong”. If the forecast warns that sales will drop, and management acts so that they do not drop, the forecast did exactly its job. It told you where you were heading, so you could change course.
A forecast is not a promise. It is information you can act on.
Five ways to keep forecasts and targets apart
- Give them different names and owners. Write “forecast” and “target” on every slide, and keep them on separate lines. The forecast belongs to the forecasting process. The target belongs to management.
- Always show the range. Share the forecast with its uncertainty, not as one number. A range lets you ask a much better question: how likely are we to hit the target?
- Ask what has to be true. Use scenarios to work backwards. What would the market, prices or demand need to do for the forecast to reach the target? Is that realistic, or is it a wish?
- Start with an objective baseline. Let the statistical forecast come first, then add human knowledge on top. Models and people are good at different things: models do not talk to customers or read the news. Keep a record of every adjustment and why it was made, so over time you can see which adjustments actually improve accuracy.
- Judge forecasts as forecasts, not on whether the target was hit. For short-term forecasts, that means accuracy. For long-term forecasts, accuracy alone can mislead, because a good forecast often triggers action that changes the outcome. There, ask whether the assumptions held and whether the forecast got the direction right. Either way, reward honest forecasts, especially the ones that bring bad news early.
How Indicio helps
Indicio is built to give you an objective forecast that is separate from anyone’s wishes. It tests many econometric, statistical and machine-learning models on your data and leading indicators, backtests them and combines the best ones. The result is a baseline that comes from evidence, not from targets.
Explainable forecasting with SHAP values shows which indicators push the forecast up or down. They are not proof of cause and effect, but they show the story the models are telling. That turns the conversation with management from “why is your number so low?” into “what is the forecast seeing that we are not, and does it change our plan?”
With scenario analysis, you can change a driver and see how the forecast moves, side by side with your baseline. It is a practical way to test what it would take to close the gap between forecast and target. Keep the scenarios realistic, though. A model learns from the past, and a scenario far outside anything it has seen can push it into territory it was never built for.
Takeaway
Forecasts and targets are both important, but they are not the same thing. When the forecast becomes the target, you lose your early warning system, and people start forecasting what is safe instead of what is likely.
Keep them apart, show the uncertainty and use the gap between them to make better decisions.
A forecast tells you where you are heading. A target tells you where you want to go. You need both, and you need them to disagree sometimes.
Want an objective forecast you can compare your targets against?
Click to start a virtual demo today.
