In one line
A seasonal index says how much a quarter or month sells against an average one, because of the time of year alone. Calculate it as the period's average sales over several years divided by the average of all periods: 1.10 means 10% above an average quarter.
What it does
- Step 1: collect two or three years of sales by quarter or month, so each period's luck averages out. One year mixes the season with everything else that happened in it.
- Step 2: average each quarter across the years, then average all the quarters together.
- Step 3: divide. Seasonal index = the quarter's average ÷ the average of all quarters. The four quarterly indices average 1.
- Worked example from the lab: over three years, Countermove's soft-drinks category sells 210.8m cans in an average second quarter against 193.9m in an average quarter, so the second quarter's index is 210.8 ÷ 193.9 = 1.09. The four come out at 0.92, 1.09, 1.11 and 0.89, close to the demand index the engine runs on: 0.92, 1.08, 1.10 and 0.90.
- Step 4: deseasonalise. Deseasonalised sales = actual sales ÷ the period's index. What is left is the trend, plus whatever you, your rivals and the economy did.
- Monthly and weekly indices work the same way. Countermove's Live screen spreads each quarter over 13 weeks along the season, so the weeks lean towards the quarter next to them: the fourth quarter's first week, just after the summer, sells about 7% above that quarter's average, and its last week about 2% below.
Business and brand
Impact on the businessP&L, cash, share price
- A great quarter is often just the season. In the lab's second year, Brightwater's sales rise 16% from the first quarter to the second with nothing changed; deseasonalised, they fall about 2%. Compare with the same quarter last year, or with deseasonalised sales, never with the quarter before.
- Time media to the peak: the same euros meet more people ready to buy. In the lab, the same extra activation adds 0.78m cans in the third quarter against 0.59m in the first or fourth, and returns €1.09 of gross profit per €1 over a year when it runs in the second quarter, against €0.91 in the fourth.
- A deal can pay better against the season. Its cost is the discount on every unit sold on deal, including units that would have sold anyway, and the peak has more of those. In the lab a 25% deal adds 18.1m cans in the third quarter against 13.4m in the fourth, but costs €9.0m against €6.9m: €0.45 of gross profit back per €1 against €0.43, and most in the first quarter, at €0.49. Outside the game, media can also cost more in the peak; Countermove keeps media prices flat all year.
- Load retailers for the peak, not beyond it. Countermove's retailers keep about four weeks of cover at the current rate of sale, so with no loading at all Brightwater's shipments run about 5% above its sell-out in the second quarter and 8% below it in the fourth. Loading more on top only moves shipments earlier and pays the loading discount: shelves in the game never run empty, and shoppers buy no more (lesson 6).
Impact on the brandMemory, loyalty, pricing power
- Brand media builds memory that lasts quarters, so it can run ahead of the peak and still sell in it (lesson 7).
- Deals in every peak teach shoppers to wait for them, which wears down the baseline (lesson 5).
- A seasonal brand that goes quiet off-season lets memory fade, and the next peak starts lower.
Try it
Move the same money between quarters
Run the same extra media, or the same deal, in each quarter of the year in turn, with everything else held. Compare the lift against the quarter before with the lift against the seasonal baseline, and see which quarter pays.
| Quarter | Index | Extra cans (m) | Back per €1 |
|---|---|---|---|
| Q1 | 0.92 | 0.59 | €1.00 |
| Q2 | 1.09 | 0.71 | €1.09 |
| Q3 | 1.11 | 0.78 | €1.08 |
| Q4 | 0.89 | 0.59 | €0.91 |
Cans Brightwater sold to shoppers over two years: Q5 to Q8 are the second year's Q1 to Q4. As sold, both lines rise and fall with the season. Back per €1 counts gross profit over the quarter and the three after; green paid back, red lost money. Every figure is the average of 8 runs of the same market with different luck.
| Quarter | Y1 | Y2 | Y3 | Average | Index |
|---|---|---|---|---|---|
| Q1 | 179.6 | 177.6 | 178.4 | 178.6 | 0.92 |
| Q2 | 212.3 | 208.8 | 211.1 | 210.8 | 1.09 |
| Q3 | 215.8 | 211.7 | 215.4 | 214.3 | 1.11 |
| Q4 | 172.1 | 171.5 | 171.8 | 171.8 | 0.89 |
| All | 193.9 | 1.00 |
The engine's own demand index for soft drinks is 0.92, 1.08, 1.10, 0.90; three years of sales recover it to within 0.02.
Notice: Extra activation in Q2: against the quarter before, sales look +25%; against the seasonal baseline the lift is +2%. The season alone lifts demand about 18% from Q1, and the quick comparison counts that as the move's. Over the quarter and the three after it returns €1.09 of gross profit per €1: it paid back. Per €1, this move works best in Q2.
Runs Countermove's market engine: the reference soft-drinks market, three brands, rivals holding their plans so only your change moves the result.
Sources
- Hyndman & Athanasopoulos (2021), Forecasting: Principles and Practice, 3rd edition, OTexts: time series decomposition and seasonal adjustment
- Makridakis, Wheelwright & Hyndman (1998), Forecasting: Methods and Applications, 3rd edition, Wiley: classical decomposition and seasonal indices
- Hanssens, Parsons & Schultz (2001), Market Response Models: Econometric and Time Series Analysis, 2nd edition, Kluwer