
.Most businesses track too many metrics and understand too few.
You don't always need more data. However, most of the time, you need two questions that work on any metric you care about: Is it moving? And by how much?
Trend and magnitude. That's it.
Trend tells you direction. Revenue is climbing or falling. Customer complaints are increasing or decreasing. Time-to-close is shortening or lengthening. The trend is the vector: where the thing is headed if nothing changes.
Magnitude tells you how much. Revenue climbed 3% or it climbed 40%. Complaints doubled or they ticked up by two. Time-to-close dropped from six months to three weeks, or it dropped from six weeks to five and a half.
Together, these two dimensions answer the only question that matters for any metric: Should I do something about this, and how fast?
A metric moving in the wrong direction with high magnitude is a crisis. A metric moving in the right direction with high magnitude is momentum you can compound. A metric moving in the wrong direction with low magnitude might be noise. A metric moving in the right direction with low magnitude might be a leading indicator or it might be nothing at all.
Precision can be misunderstood
A metric that exists as a snapshot in time without a few additional data points in the future or past is incomplete. It's a number without sufficient context.
What's happening? The second-order question is: Why?
You can't answer why until you've established what. And what is always trend and magnitude.
Basecamp's famous revenue chart is a single line going up and to the right. No breakdown by product, no cohort analysis, no funnel visualization. Just the line. Jason Fried has said repeatedly that this is the only chart that matters to them because it tells them whether the business is working.
It's not an oversimplification. It's intellectually honest. The line summarizes the outcome of everything the company does. If the line is moving in the right direction at a meaningful rate, the company is healthy. If it isn't, no amount of segmentation will fix that.
Trend and magnitude force you to confront the outcome before you start explaining it.
Why Magnitude Matters
Trend without magnitude is a trap of its own.
A metric can be moving in the right direction and still not matter. Revenue growing at 2% year-over-year is technically growth, but if your cost of capital is 8%, you're losing ground. Customer satisfaction scores ticking up by half a point on a ten-point scale might be statistical noise, not a signal that your product changes are working.
Magnitude is what separates signal from noise, and what separates a problem you can ignore from a problem that will kill you.
In 2018, Facebook's daily active user growth in North America went flat. The trend was neutral, but the magnitude of the change was enormous. For a company that had grown users every quarter for a decade, flat was a crisis. The market reacted accordingly. Facebook's stock dropped 20% in a single day.
The magnitude of the change mattered more than the absolute number. Flat growth in a mature market might be fine. Flat growth in a market you've always grown in is a different story.
Most businesses don't calibrate their response to magnitude. They treat every downward trend as equally urgent, or they treat every small movement as equally ignorable. Both are mistakes.
The fog of complexity
Trend and magnitude work because they're legible under pressure.
Complex metrics require cognitive overhead. You have to remember what the metric measures, how it's calculated, what the baseline is, what the target is, and what the variance means. That's fine when you're sitting at your desk with the dashboard open and twenty minutes to think.
It's not fine in a meeting where someone asks, "How are we doing on customer retention?" and you have to answer in real time.
Trend and magnitude collapse the cognitive load. The answer is always structured the same way: "It's moving [direction] by [amount]." That structure works for any metric, in any context, with any audience.
"Retention is up 8% quarter-over-quarter." That sentence communicates everything a non-specialist needs to know. The trend is positive. The magnitude is meaningful. The business is improving on this dimension.
Compare that to: "Retention is at 87%, which is above our target of 85% but below last year's peak of 91%, and the cohort analysis shows that the improvement is concentrated in enterprise customers while SMB retention is flat."
That's more information, but it's not more clarity. The listener has to hold five data points in working memory and synthesize them into a judgment about whether things are getting better or worse. Most people can't do that in real time. They nod, the meeting moves on, and nobody actually understood what was said.
Trend and magnitude respect the listener's cognitive capacity. They give you the headline first. If someone needs the breakdown, you can provide it. But the breakdown is never the lead.
When Trend and Magnitude Aren't Enough
There are cases where trend and magnitude don't tell the whole story.
If a metric is seasonal, the trend might look negative when it's actually normal. If a metric is lagging, the trend might look fine when the leading indicators are already breaking. If a metric is gamed, the trend might look positive when the underlying behavior is destructive.
These are real limitations. Trend and magnitude are a starting point, not a replacement for understanding your business.
But I've seen teams use the limitations of trend and magnitude as an excuse to avoid using them at all. They say, "Well, it's more complicated than that," and then they build a reporting system so complicated that nobody can use it.
The correct response to "trend and magnitude don't tell the whole story" is "therefore we need trend and magnitude plus context."
Context is the third dimension. It's the thing that tells you whether the trend and magnitude you're seeing are normal, concerning, or misleading. But context is narrative, not numeric. It's the explanation you give after you've established the facts.
Stripe's internal metrics culture is built this way. Every metric has a headline number (the current value) and a trend line. If the trend is surprising, the person presenting the metric explains why. The explanation might be seasonal adjustment, or a known data quality issue, or a deliberate strategic choice. But the explanation comes after the number, not instead of it.
That's the right order. Establish what's happening. Then explain why.
The Discipline of Two Questions
The real power of trend and magnitude is that they force you to decide what you care about.
If you can't describe a metric in terms of trend and magnitude, you probably don't understand it well enough to act on it. And if you can describe it that way but the trend and magnitude don't concern you, you probably shouldn't be tracking it.
If a metric isn't moving, or if the magnitude of its movement doesn't matter, stop tracking it. You're not learning anything.
The discipline is asking two questions about every metric on your dashboard:
Is this moving in a direction I care about?
Is the magnitude of that movement large enough to act on?
If the answer to both is no, delete the metric. If the answer to one is yes, keep watching. If the answer to both is yes, do something.
That's the entire decision tree. Everything else is elaboration.
What You're Left With
When the question is just "which way is it going and by how much?" you can't hide behind complexity. When the framework is "if it's moving the wrong way fast, you have a problem," you can't defer the decision.
The businesses that move fastest are the ones that can answer those two questions for every metric that matters, in real time, without pulling up a dashboard. They've internalized the trends. They know the magnitudes. They don't need a meeting to decide whether something is urgent.
They've built their metrics around the two dimensions that actually drive decisions.
The rest is just commentary.