Your dashboard shows last week was up 12%.
But is that good? Was the week before down 15%?
You can see today's number. You can't see where it's heading.
Numbers tell you where you are. Trends tell you where you are going.
INTERMEDIATE - Turns raw data points into directional signals.
You have a number. Support tickets this week: 47. Is that good or bad? It depends entirely on what came before. If last week was 45 and the week before was 43, you're looking at an upward trend. If last week was 52 and the week before was 58, you're looking at improvement.
Trend analysis takes sequential data points and extracts the underlying direction. It answers the question every decision-maker actually cares about: are things getting better or getting worse?
This isn't just about drawing lines on charts. It's about detecting when a metric starts moving in a new direction before it becomes obvious. Catching a slow decline at week 3 is better than noticing it at week 12.
Get it wrong and you react to noise instead of signal. Get it right and you see problems before they become crises.
Trend analysis solves a universal problem: how do you distinguish between random fluctuation and meaningful directional change?
Compare current values to historical baselines. Look for consistent direction across multiple data points. Filter out noise to reveal the underlying trajectory.
Click on a data series to analyze its trend direction over the 7 data points.
Select a series to analyze
Smooth out the noise
Instead of looking at each data point, you look at the average of the last N points. A 7-day moving average smooths out daily variation. Compare today's average to last week's average to see direction.
Simple to implement and explain
Lags behind sudden changes
Find the best-fit line
Draw the line that best fits your data points. The slope of that line is your trend. Positive slope means increasing. Negative means decreasing. The steeper the slope, the stronger the trend.
Quantifies trend strength with a single number
Assumes linear relationships
Compare like to like
Compare this week to last week. This month to last month. This quarter to last quarter. Calculate the percentage change and track whether that change is consistently positive or negative.
Intuitive and easy to communicate
Sensitive to seasonal variations
Trend analysis sits between raw data storage and actionable intelligence. It transforms time-series data into directional signals that drive prioritization and alerting.
Hover over any component to see what it does and why it's neededTap any component to see what it does and why it's needed
Animated lines show direct connections · Hover for detailsTap for details · Click to learn more
One week of high numbers doesn't mean you're on an upward trend. It might be an anomaly. You need multiple data points moving in the same direction to call it a trend.
Instead: Require at least 3 consecutive data points in the same direction before calling it a trend.
January is always slower than December. If you compare January to December, you'll see a 'decline' that isn't real. You're comparing different seasons, not detecting a trend.
Instead: Compare same period to same period. This January vs. last January. This Monday vs. last Monday.
Yesterday was up from the day before. Is that a trend? No, that's noise. Short windows are dominated by random variation, not meaningful direction.
Instead: Match your analysis window to your decision timeframe. Weekly decisions need weekly trends, not daily.
You've learned how to extract direction from sequential data. The natural next step is using those trends to detect when something unusual is happening.