Soft demand before peak: How to move seasonal stock without discounting everyone

UK shoppers are buying less, yet seasonal stock remains on shelves. What an awkward place to be heading into peak.

The latest ONS figures show UK retail volumes fell 0.4% in September 2026. Warm weather hit sales of autumn and winter clothing, and mail order sales fell too.

The quick fix is to mark everything down early, but it’s also the expensive one.

Here is what the data says, why blanket cuts hurt, and how to move stock without handing margin away.

What the September numbers say

The ONS reports that online sales values fell 0.4% over the month to September 2026. In August they had risen 3.0%. (Not a crash, but not a party either.)

Over the year, online is still up. Online sales were 6.8% higher than in September 2025.

The online share of all sales was 28.8%, down from 28.9% in August.

So the picture looks like this:

• Demand is soft, not broken. Online is still growing year on year.
• Clothing took the hit. Warm weather held back autumn and winter lines.
• Mail order sales fell too.

Our reading is that seasonal stock may need a push. Retailers will feel pressure to cut prices early.

A warm September may delay purchases rather than cancel them. We cannot know yet. But it is a reason not to slash prices on day one.

Why blanket markdowns cost more than they look

Cutting prices for everyone is simple. It is also blunt. Plenty of those shoppers would have bought anyway.

The research on broad discounting is not kind:

• Analysis by Les Binet and Nielsen, reported by Marketing Week, found that about 84% of price promotions are unprofitable.
• BCG estimates that 30-40% of retail promotions are inefficient. It also found that personalised offers generate 3x ROI.

Clothing adds a second wrinkle. Capital One Shopping reports that 62% of clothing shoppers delay a purchase. Cut early and wide, and you may teach them that waiting pays.

Then there is peak itself. These are US figures, but they show where the market is heading. Adobe forecasts US online holiday spend of $275.1bn, up 6.7% year on year. It expects discounts of up to 30% during Cyber Week.

The NRF forecasts that retail sales in November and December will be up 4.2% to $1.013 trillion. Last year's growth was 3.1%. Faster growth means more brands fighting for the same attention.

In a crowded peak, a blunt discount is easy to copy and hard to profit from. (Everyone else has a banner too.)

How to identify which stock needs to move

Before you can act, you need to know exactly which products are at risk of tying up cash and shelf space. Start by reviewing your seasonal inventory levels against sales velocity.

Look for items with a high weeks-of-supply remaining relative to their typical sales rate. These are your slow movers.

Pay attention to products where sales have stalled despite strong marketing, and where incoming stock or seasonality means time is running out to sell at full price.

Key indicators that stock needs to move include:

• High inventory relative to historical sales for this period.
• Approaching end-of-season, with little likelihood of demand returning soon.
• Declining product views or add-to-cart rates.
• Repeatedly bypassed in customer searches or site navigation.
• SKUs with upcoming new arrivals that will push them further down the pecking order.

Use your data tools to segment at the SKU or category level. Don’t rely only on instinct or averages; let the numbers guide you to the lines where action will have the biggest impact.

Once you know what needs to move, you can tailor your incentives and communications to maximise impact without eroding margins across your entire range.

Who actually needs the offer?

Start with a simple question: Does this shopper need an incentive to buy?

Some will pay full price, some need a nudge, some will leave whatever you do.

Targeted offers go only to the middle group. That protects margin and lifts incremental revenue. It also means fewer offers overall, so you spend less on promotions. It is not the same as spreading the budget around more cleverly.

Here is how that can look on seasonal stock:

• A shopper with a winter coat in the basket and signs of leaving gets a nudge.
• A shopper browsing knitwear with no clear intent gets a softer prompt, such as free delivery.
• A shopper ready to buy at full price gets nothing. (Yes, nothing. It is a lovely feeling.)

Timing matters here. About 70% of carts are abandoned. Reacting after the shopper has gone is a weak position.

Tempo is our predictive intent engine. It spots the signs of leaving before it happens.

That is Precovery: acting before a shopper abandons, not chasing them afterwards.

It is a predictive machine learning method that forecasts intent. It is not generative AI. Prediction beats reaction.

A plan for the weeks before peak

You do not need a new strategy. You need a sharper order of play.

• Sort your stock into three groups: fast sellers, steady sellers and slow seasonal lines.
• Leave fast sellers at full price. They do not need help.
• Try softer incentives on slow lines. Free gifts or free shipping can move stock without cutting the price tag.
• Keep deeper discounts for shoppers who show they need one.
• Measure with a control group so you know which sales the offer actually drove.

Set your rules before the pressure arrives. Decide the deepest discount each product group can take. Decide your margin floor. Write both down.

Peak is a season, not a single day. A plan that covers the whole run stops one bad week from setting the tone for the rest.

Track the right numbers too. Redemptions tell you how many people used a code. They do not tell you whether the code changed anything. Redemptions are the wrong metric for judging a discount. Look at incremental revenue and margin per order instead.

Peak does not have to be a panic. Our post on how to stop panic wrecking your discount strategy covers the common traps. And turning slow-moving inventory into profit shows how to clear stock without devaluing your brand.

Prediction beats reaction

The ONS data is a prompt, not a verdict. Demand is soft. Seasonal stock is stuck. Peak is crowded.

But this is also an opportunity to rethink old habits. Peak season rewards retailers who act with precision, not panic - who make every offer count and reserve margin for the shoppers who actually need a nudge.

There is no one-size-fits-all approach: segment your stock, segment your customers, and respond to real signals in real time. Smart use of data, predictive intent, and incremental testing will not just protect profit: they will build resilience for the seasons ahead.

The next few weeks will be noisy, but you do not have to join the race to the bottom. The winners will be those who stay disciplined, learn fast, and stick to a strategy that serves both their bottom line and their brand. Plan now, act thoughtfully, and let the data guide you through peak and beyond.

It is tempting to react with a bigger sale. A smarter move is to decide who needs an offer, and when, before they leave.