Slow Season for Online Stores: How Deep It Goes, What Helps

In the slow season (your weakest months), keep your ads running at a lower, steady level rather than switching them off. Cut the budget in small steps, and only as far as demand falls. Use the quiet weeks for work the peak won’t allow: the feed, campaign structure, next year’s assortment and past customers. At stake: in the median store, the months below a typical month fell short of it by a total of 10.8% of the year’s ad revenue (GetProfit data, 96 stores, July 2025 – June 2026).

This article covers only the trough, the low point of the year. Our guide to seasonal demand in online stores covers the whole year, peak included.

The median store’s weak months leave a gap of 10.8% of yearly ad revenue

We measured how uneven the year is in 96 online stores that advertise on Google Ads (GetProfit data, July 2025 – June 2026). Each store had at least 10 complete months and 100+ conversions in the window. Revenue here means ad revenue: the conversion value Google Ads records, not the store’s total turnover. We removed each store’s own growth trend first, so that a store growing all year doesn’t look like a store with a weak spring.

Then we took each store’s median month as its “typical” month. We added up how far every weaker month fell short of it and divided the total by the year’s revenue.

The year was uneven almost everywhere. With the trend removed, the median store’s three best months brought 2.62× the revenue of its three worst. In the median store, the three best months took 41.7% of the year’s revenue; a perfectly flat year would give them 25%. In 92 of the 96 stores, the three best months brought at least 1.5× the three worst; in 72 stores, at least 2×.

Read the 10.8% as the size of the gap, not as money you can win back. Part of it is demand that isn’t there in those months. This is also one observed year: a single deep month can be a one-off event rather than a season.

When is the slow season for an online store?

In our data, the low season most often falls in July and August, then in February and April. We counted each store’s three weakest months separately; the table shows how many of the 96 stores had each month among them.

GetProfit data, 96 stores, July 2025 – June 2026. A month counts for a store only if the store’s data covers it in full.

Your off-season may fall in other months: 12 stores had November among their weakest months. Find yours in your own data:

1. Take monthly ad revenue for the last 12–24 months, from the same report every time.
2. Correct for growth. If the store grew fast during the year, compare each month with the trend line rather than with the raw average. Otherwise early months look weak only because the store was smaller then.
3. Mark the three weakest months and check whether the same months were weak the year before. If they were, it’s a season. If not, treat it as a one-off.
4. Split by category. A store can be flat overall while one category drops sharply in July. Our guide to category seasonality shows how to read this.

Is the dip demand, or your own budget cut?

Often both, though the data can’t separate the two precisely.

GetProfit data, 96 stores, July 2025 – June 2026.

Each unit of ad spend returned less in the weak months: the median store’s ROAS there was 0.70 of its peak level. So the drop in demand is real, not just a side effect of a smaller budget. Yet in two thirds of stores, spend in those months was also more than 30% below the best months, so the cut may have widened the gap.

These are observations, not an experiment. A store may cut spend because demand has already fallen, and we can’t tell how much revenue the cut cost each store.

Calculate the same two ratios for your own account:

• Spend ratio: spend in your three weakest months divided by spend in your three best.
• Return ratio: ROAS in your three weakest months divided by ROAS in your three best.

If the return ratio is close to 1 and the spend ratio is far below it, the money still worked. Then the cut likely accounts for part of the gap. If both ratios are low, demand fell and the cut followed it.

Should you pause Google Ads in the slow season?

If you sell anything at all in those months, no. Run the campaigns at a lower level and change them in steps. Here is why, and what to do instead.

The bid strategy statuses come from the Google Ads API reference (BiddingStrategySystemStatus). It lists separate learning statuses for a new or reactivated strategy, a recent budget change, a recent setting change and a change in the campaigns attached to it.

On seasonality adjustments, Google’s guide (Create seasonality adjustments) adds that they may not work as well over more than 14 days at a time. A slow season lasts months, so it is not a job for this tool.

The step sizes we use: change a campaign’s daily budget by no more than 20% at a time. Move target ROAS by no more than 15% at a time, at most once every one to two weeks. These are our working rules, not Google’s.

If your store sells nothing at all for part of the year, pausing the ads makes sense, but keep the site up. Google Search Central (Temporarily pause or disable a website) recommends keeping the site online with limited functionality, for example a disabled cart and a banner. It warns that closing a site completely, even for a few weeks, can harm how Google indexes it.

What to work on while sales are slow

The slow season is the time to change what you shouldn’t touch close to the peak.

Plan the assortment for next year’s trough

Our seasonality method treats a weak period that repeats as structural. For such a period, it advises keeping costs down and, where possible, adding products that sell in it. That is not always possible. While the trough lasts, use the time to think about next year’s assortment.

Start with what your store already has. If several categories dip at the same time and none is strong in that window, you have a seasonal gap.

This is common. Of the 96 stores, 42 had at least two significant categories (each with 5%+ of revenue and 30+ conversions). In 25 of those 42 (60%), no such category had any of its three best months among the store’s three weakest.

A category that sells in your weak months evens out the year, but in our data it did not go with faster growth. We compared each store’s last three full months with its first three (GetProfit data, 42 stores, July 2025 – June 2026). The median store with such a category grew 1.16× over the window; the median store without one grew 1.22×.

The other 54 of the 96 stores could not be tested: most had fewer than two such categories at Google’s top category level. For them, a counter-seasonal category means a new niche. Our guide to counter-seasonal products looks at whether that is worth it.

For ideas, check what shoppers buy in your categories during your weak weeks. According to Google’s Merchant API reference (BestSellersProductClusterView), the best sellers report ranks products, grouped into product clusters, by popularity in a category and country, weekly or monthly. The ranking is based on estimated units sold. Each cluster also shows whether it is in stock in your data, out of stock or not in your inventory at all.

Fix the catalogue and the structure before the run-up

Check your assortment coverage: which products in the feed never got an impression, and why. Fix feed errors and add missing products to campaigns now, while each change puts less revenue at risk.

If the campaign structure needs rebuilding, do it in the trough and finish before the run-up starts. Our rule is to keep the structure as it is right before and during the peak, because structure changes can send bid strategies back into learning.

Sell to people who already bought

You can reach past buyers through your own channels, such as email, without paying for each click. Our guide to customer retention covers what an online store can realistically do there.

Put the dates in the calendar

Start preparing for the trough two months before your weakest month: that is the deadline in lesson 13 of our course on product advertising. For the peak that follows, we start scaling campaigns up gradually one to two months ahead. By the time the season starts, they have enough data to grow. For a November peak, that means late August or September.

A worked example: one store’s slow season

Example store, not client data.

A tableware store with 3,000 products spends an average of 40,000 a month on ads and earns 2,160,000 a year from them, or 180,000 a month. Its ROAS for the year is 4.5. Here is its year:

The typical month, the median of the twelve, is 150,000. The months below it fall short by 210,000 in total, or 9.7% of the year. That is close to the median store in our data. The three best months (October to December) bring 970,000 and the three worst (February, July, August) bring 280,000, so the peak-to-trough ratio is 3.46×.

Now the spend. In the three worst months the store spent 68,000, against 168,000 in the three best. Its spend ratio is 0.40. ROAS was 4.1 in the worst months (280,000 ÷ 68,000) and 5.8 in the best (970,000 ÷ 168,000), so its return ratio is 0.71.

Demand did fall: the return ratio is close to the 0.70 median, and with both ratios low, the two-ratio test says the cut followed demand. But spend fell much further than return, and the weak months still returned 4.1, close to the store’s overall ROAS of 4.5. So the cut probably went deeper than demand required.

Next year the store cuts its budget in 20% steps, watches ROAS after each one and stops cutting once ROAS shows the money still works. In June, two months before August, its weakest month, it also starts looking for products that sell in July and August.

Your slow-season checklist

1. Name your three weakest months by ad revenue, corrected for growth, and check them against the year before.
2. Size the gap. Add up how far each month below your typical month (your median month) falls short of it. Divide the total by the year’s ad revenue. Compare it with the 96 stores in our data: a quarter had up to 7.0%, the median 10.8%, a quarter more than 14.7%.
3. Calculate the spend ratio and the return ratio for your three weakest months against your three best. They show how much of the dip is demand and how much is your own cut.
4. Set the steps for your ads. Keep campaigns running. Cut the budget by no more than 20% at a time. Hold target ROAS, or move it by no more than 15% at a time and at most once every one to two weeks. Use seasonality adjustments only for short events.
5. Mark the weeks for structure changes, feed fixes and coverage work, and finish them all before the run-up to your peak.
6. Pick one direction for next year’s trough: a category you already have, a product type that sells in those weeks, or work with past buyers.
7. Put the start date in the calendar: two months before your weakest month.

The portal’s assortment section helps with the first step, finding your weakest months, in your own account. It shows revenue by month over two full years and says whether the season repeats. Each store, and each notable category in it, gets one of five states, from “Confirmed by two years” to “Too little data — analysis impossible”. A store may have no season while a category in it does.

How much of the catalogue reached sales — and what the rest is doing. The portal walks the whole catalogue through the stages — from the feed to sales, shows where the budget goes by product group, and answers whether you really have a season. The portal changes nothing without your consent.

Break down your catalogue →

Sources

• BiddingStrategySystemStatus — Google Ads API reference — a bid strategy shows a learning status when it was recently created or reactivated, after a recent budget change, a recent setting change, or a change in the campaigns attached to it. Checked 2 October 2026.
• Create seasonality adjustments — Google Ads API — seasonality adjustments are meant for short events of 1–7 days and may not work as well over more than 14 days at a time. Checked 2 October 2026.
• Temporarily pause or disable a website — Google Search Central — keep the site online with limited functionality; closing it completely, even for a few weeks, can harm indexing. Checked 2 October 2026.
• BestSellersProductClusterView — Merchant API reference — the best sellers report ranks product clusters by popularity in a category and country, weekly or monthly, based on estimated units sold, with your inventory status for each. Checked 2 October 2026.
• GetProfit data: 96 stores, July 2025 – June 2026 — size of the trough, peak-to-trough ratio, weakest and best months, spend and ROAS in the weakest months, categories that peak in the trough and growth with and without them.
• GetProfit methodology — how we treat a repeating weak period and the off-season, step sizes for budget and target ROAS, structure changes before the peak, preparation one to two months ahead.
• GetProfit portal — seasonality over two years and its five states.