How to explain a demand spike (and not just celebrate it)

April was up 18% on March. Everyone is pleased. Then someone asks the question that matters: why?

If the answer is “the promotion worked”, the next promotion gets the same budget and may not work at all, because the promotion was one of four things that happened at once. Explaining a spike means separating those things and putting a rough weight on each.

Step 1: list everything that changed

Start wide. For the weeks around the peak, write down every signal you can find, from your own calendar and from outside:

• Weather (a four-week heatwave)
• Holidays and seasons (Avurudu)
• Social attention (a lunchbox video that reached millions of views)
• Economic changes (a fuel-price cut that put money back in pockets)
• Your own actions (a promotion, a new pack, a price change)
• Competitor actions (a rival out of stock, or a rival promotion ending)

Step 2: check each one against the past

For each signal, ask whether a similar signal has moved demand before, and by how much. Heatwaves have happened before; so have holidays. Your own sales history, lined up against those past events, is the evidence. A signal with no precedent gets a wider range, not a confident number.

Step 3: attribute, roughly and honestly

The result is a split, not a verdict. Using the illustrative Crispo example from our site:

Cause	Estimated share of the +18%
Four-week heatwave	about 45%
Avurudu holidays	about 30%
Viral lunchbox videos	about 15%
Fuel-price cut	about 10%

Figures are illustrative. The point is the shape: the thing you controlled was not the biggest cause, and the biggest cause will not repeat on your schedule.

Step 4: close the loop

The attribution is only as good as the sales data behind it. Connect real sales to the signal timeline, and each new month tells you which estimates were right. Over time the ranges narrow, and “why did April happen?” becomes a question you can answer in the same week.

This is the second of the three things Forevate does: it traces every rise and dip to its causes and estimates how much each one contributed. You can see the example play out on the home page.