How to track competitor promotions and discounts

Most "competitor price monitoring" advice tells you to scrape product pages. That catches a price change after it happens — it misses the sale your rival is actively paying to put in front of your customers. This guide covers the faster signal: tracking promotions through competitors' ads, using the free ad libraries, a simple recording system, and real numbers from 29,536 UK ads we've qualified so far, so you know what a normal promo cadence in your sector actually looks like.

Why promotions surface in ads before anywhere else

Price-monitoring tools watch a competitor's website. That's useful, but it has two blind spots:

• You only see what's on the page you're watching. A homepage banner is the public face of a promotion. The 20% new-customer code, the abandoned-basket voucher, the "first box free" offer — those often run exclusively in paid social, targeted at people who aren't existing customers. They never touch the homepage, so a page monitor never fires.
• You see the change, not the push. A price drop with no media behind it is housekeeping. A price drop with thirty fresh creatives behind it is an offensive — your rival is paying to put that offer in front of your prospects. The ad spend is the part that hurts you, and it's the part a page scraper can't see.

The ad libraries fix both. A promotion has to be advertised to work, and the moment it is, it appears in Meta's Ad Library, Google's Ads Transparency Center and TikTok's Commercial Content Library — creative, copy, start date and all, no account required. If you're new to those tools, our guide to seeing competitors' ads for free walks through each one step by step.

What counts as a promotion? The offer types we actually see

"Promotion" covers more than percentage-off sales. Across the 29,536 live UK ads we've AI-qualified to date (112 brands, Google, Meta and TikTok), 8,103 — 27% — carry some form of offer. Here's how they break down:

Two things worth internalising. First, nearly half of all offers aren't a simple discount — free trials, finance and gifts are ways of discounting without cutting the sticker price, and they're easy to miss if you're only searching ad copy for "% off". Second, BOGO is essentially extinct: 24 ads out of 29,536. If your category still leans on it, that's a genuine anomaly worth knowing about.

How to track competitor offers manually (free)

A workable manual system needs three things: a fixed list of competitors, a fixed weekly slot, and a fixed set of fields you record every time. Here's the routine we'd run:

1. Pick 3–5 competitors. More than that and the weekly sweep dies within a month. Choose the ones whose promotions actually move your numbers.
2. Sweep Meta first. The Meta Ad Library shows every currently active ad for a Page, with start dates. Filter to the UK, sort by newest, and scan for offer language. Meta is where promo creative usually lands first, because it's the cheapest place to test it.
3. Then Google. The Ads Transparency Center catches Search and YouTube — where "sale" messaging goes once a brand commits budget to it. Be aware it caps how much it will show you for big advertisers; we've documented the ~900-creative pagination ceiling in our Transparency Center guide.
4. TikTok if your audience is there. The Commercial Content Library covers UK-shown ads with first- and last-shown dates.
5. Record what you find, every time, in the same format.

The fields worth recording

These are the fields we extract from every ad, and they're the right spreadsheet columns for a manual version too:

• Offer type — one of the eight categories above, or "none".
• Depth — the percentage or amount, exactly as stated. "Up to 60%" and "60%" are different promotions; note the "up to".
• Promo code — if there's a code in the creative, record it. Codes tell you which channel and audience an offer is aimed at.
• Deadline — "ends Sunday", "this weekend only", or nothing. A dated deadline is a real event; permanent urgency is a copywriting habit.
• Conditions — new customers only, minimum spend, first order.
• First seen / last seen — your dates, not the platform's. This is what turns a spreadsheet into a cadence record.

How often do competitors actually discount? What our data shows

The most common mistake we see is reading a rival's promo activity without a baseline. "They're running a 25%-off ad" means one thing in a sector where nobody discounts and the opposite in one where everyone does. From our qualified-ad data, the share of live ads carrying an offer varies enormously by sector:

And within a sector, strategies diverge sharply. In fashion, 36% of ASOS's live ads carry an offer against just 5% of Next's — same market, opposite promotional postures. In package holidays, Jet2 runs offers on 49% of its live creative while easyJet sits at 22%. At the extreme end, meal-kit and pet-food subscription brands are almost pure promo machines: 93% of Gousto's live ads and 89% of Butternut Box's carry an offer, because the entire acquisition model is built on a discounted first box. If you compete with a brand like that, tracking whether they're discounting is pointless — track the depth and the conditions instead.

Reading discount depth and urgency properly

Where an ad states a percentage, the average discount across our data is 33.7% (3,574 ads with a stated figure). But averages hide two traps:

1. Intro-offer distortion. B2B software shows the deepest stated discounts of any sector we track — averaging 76.5% — not because accounting software is permanently on sale, but because the standard acquisition offer is "90% off for six months" style pricing. A 90% intro discount and a 20% sitewide sale are different animals; record the conditions, not just the number.

2. "Up to" inflation. Retail sale ads overwhelmingly lead with the ceiling ("up to 70% off"), which typically applies to a thin slice of clearance stock. When Dreams runs 73% of its live creative with an offer attached, the headline percentages cluster around event pricing — that's the mattress sector's always-on sale culture, and it's why mattress "sales" shouldn't trigger any reaction from a competitor at all.

Urgency deserves the same scepticism. Our qualification scores every ad for urgency signalling, and offer-carrying ads score dramatically higher than non-promo ads — countdown language, deadline framing, stock warnings. But a deadline that appears in every ad, every week, isn't a deadline; it's a template. The only way to tell manufactured urgency from a genuine sale event is a record over time, which is exactly what the weekly sweep gives you.

Caveat on all of the above: these figures describe the ads brands run, not the sales they make. An offer-heavy ad account can reflect a deliberate acquisition strategy or a brand in margin trouble — the ad data tells you what they're doing, not whether it's working.

When the spreadsheet stops scaling

The manual system genuinely works for a handful of competitors on one or two channels. It breaks in predictable ways: the weekly sweep gets skipped, offer language in image creative (rather than copy) gets missed, and nobody spots the promo that launched on Tuesday and ended Friday between sweeps.

This is the problem AdRival was built for. We track every live ad for 105 UK brands across Google, Meta and TikTok daily, and AI-qualify each creative — extracting the offer type, discount percentage, promo code, deadline, conditions and urgency signals from the actual creative, including text baked into images that keyword searches can't see. That's what turns "check four libraries and eyeball thumbnails" into a query: which of my competitors launched an offer this week, how deep, and with what deadline? You can browse the live data free on our brand ad libraries — no account needed — and see the promo picture for brands like Pets at Home or LookFantastic before deciding whether the manual route covers your needs.

Either way, start the record this week. The value of promotion tracking compounds: every week of data makes the next promo easier to read, and the rival who surprises you with a sale in November is the one you weren't watching in August.