Pricing & Promotions
Brand & Intelligence
Digital Shelf & Search
Assortment
For Retailers
E-Commerce Dashboard
Flipkart Insights (Live) 
Grocery Intelligence.svg)
Grocery Price (U.S.).svg)
Quick Commerce (India) 
Food & Restaurant
Fashion Intelligence
Automotive
Travel & Hospitality
Real Estate
Food Delivery Intelligence 
OTT & Streaming 
By Use Case
Pricing Intelligence
Digital Shelf Analytics
MAP Monitoring 
Cross-Border Price Parity 
Share of Search
Review Sentiment
Kitchen Market Gaps 
Dynamic Pricing / AI Repricing 
Promotions & Deals Alerts 
B2B / POI & Lead Data 
By Region
Flipkart Real-Time Insights
Quick Commerce — Zepto · Blinkit
Pincode Price Tracker
GCC Q-Commerce — Talabat · Noon 
UK Grocery — Tesco, Sainsbury's, Asda
AU Grocery — Coles & Woolworths 
Marketplace Scrapers
Developer APIs
Universal APIs
Delivery & SDKs
Knowledge Center
Guides & Playbooks
Downloads & Tools
Trust & CompanyBlack Friday 2026 price tracking playbook
Written before the event, not after. What to set up now, and what nobody can tell you yet.
Most Black Friday pricing analysis fails for one reason: daily collection. Discounts in this window last hours, not days, and a daily snapshot records a market that had already moved.
What this page is, and what it is not
This is a preparation playbook, published in August 2026 for an event in November 2026. It contains no findings about Black Friday 2026, because the event has not happened.
Any page publishing 2026 Black Friday discount percentages today would be presenting a forecast as a measurement. We would rather be useful now and accurate later than the reverse.
What it does contain is the collection design: what changes in this window, which fields matter, what cadence is needed, and the measurement errors we see teams make every year. That is knowable in advance and it is what determines whether your event data is usable.
Findings will be added to this page after the window closes, based on data actually collected. Until then the sections below are a setup guide.
What actually changes in the Black Friday window
Discount duration collapses
Outside this window, a price change tends to persist for days or weeks. Inside it, discounts routinely run for hours — and several distinct offers can appear on the same SKU in one day.
The consequence for measurement is direct: a daily collection cycle will miss most of what happened. It records whatever price happened to be live at the moment of collection, which in this window is close to arbitrary.
This is the single most common reason event pricing analysis produces conclusions that contradict what teams observed on the ground.
Discounting is sequenced, not simultaneous
Retailers stage offers deliberately. Early-November teasers, a mid-month wave, the peak day itself, then Cyber Monday, then a quieter run into December.
A dataset that starts collecting on the peak day cannot see the sequence, and the sequence is where the strategy is. Whether a competitor discounted early to capture demand or held until the peak to protect margin is visible only with a run-up baseline.
Start collecting at least three weeks before the peak. Without a pre-window baseline, every discount depth figure is computed against a price you never observed.
Reference prices move before the discounts do
Discount depth is meaningless without a reliable reference price, and reference prices are least reliable in exactly this window. A "was" price shown on the peak day may not match what the SKU actually sold for in October.
This is why we recommend computing depth against your own observed pre-window price rather than against the retailer-displayed reference. Capture both, and where they diverge that divergence is itself worth recording.
In several markets, reference pricing in promotional windows attracts regulatory attention. Recording what was displayed, with timestamps, is useful beyond competitive analysis.
Availability becomes the constraint, not price
By the peak day, the binding constraint on many SKUs is stock rather than price. A deeply discounted item that is unbuyable is not a competitive threat, and a dataset tracking price without availability will overstate competitor aggression substantially.
Stock-out duration matters more than stock status here: an item out for twenty minutes and one out for two days are different events, and only transition-level collection distinguishes them.
What to track, and at what frequency
Cadence recommendations assume a tiered design: the highest frequency on the SKUs and zones where a competitor move changes your decision, lower on the tail.
| What to track | Why it matters in this window | Frequency |
|---|---|---|
| Price, with timestamps | Discounts last hours. Without a precise observation time you cannot reconstruct the sequence or compute duration. | Hourly on priority SKUs |
| Pre-window baseline price | Depth computed against a retailer-displayed reference is unreliable in this window. Your own observed October price is the defensible baseline. | Daily, from early November |
| Displayed reference price | Capture what the retailer showed as the was-price, separately from your own baseline. Divergence between them is worth recording. | Hourly |
| Availability with duration | A discounted item that is out of stock is not a threat. Stock-out duration separates a twenty-minute gap from a two-day one. | Hourly |
| Promotional mechanic | Percentage off, multibuy, bundle, voucher and loyalty-gated offers are not comparable as a single discount figure. | Hourly |
| Offer start and end times | The window each offer ran is the field that makes sequence analysis possible. | Derived from hourly capture |
| Sponsored placement share | Paid placement density rises sharply in this window and changes what shoppers actually see. | Daily |
| Assortment changes | Event-only SKUs and bundles appear and disappear. Counting them as range changes distorts assortment analysis. | Daily |
Five measurement mistakes we see every year
Daily collection
The most common and most damaging. In a window where offers run for hours, a daily snapshot samples a market at one arbitrary moment. Teams then wonder why the data disagrees with what they saw.
No pre-window baseline
Starting collection in late November means every depth figure is computed against a reference you never observed. Three weeks of run-up is the minimum.
Trusting the displayed was-price
Reference prices are least reliable in exactly this window. Capture the displayed reference, but compute depth against your own observed baseline.
Ignoring availability
Price-only collection overstates competitor aggression, because the deepest discounts frequently sell out fastest. A discounted unbuyable SKU is not competition.
Blending event SKUs into assortment analysis
Event-only bundles and exclusives inflate apparent range expansion and then appear as delistings in December. Flag them as event SKUs or exclude them.
What we will publish after the window
After the window closes we will add findings to this page based on data actually collected: discount depth distributions by category and market, offer duration distributions, the sequencing pattern across the run-up, and how much of the observed discounting was availability-constrained.
Those figures will be dated, will state the collection cadence and SKU population they came from, and will note where coverage was thin. A report without those qualifiers is not a report.
If you want the same analysis on your own competitive set rather than our aggregate view, the collection design above is what we would run. A pilot takes 48 hours and there is time to have it live well before the run-up begins.
Questions about this playbook
Including why there are no 2026 figures on it yet.
Because Black Friday 2026 has not happened. It is late November; this page was published in August.
A page publishing 2026 discount percentages now would be presenting a forecast as a measurement. Findings will be added after the window, dated, with the collection cadence and SKU population stated.
At least three weeks before the peak day, so early November. Without a pre-window baseline, every discount depth figure is computed against a reference price you never observed.
Starting on the peak day also makes the sequencing invisible, and sequencing — who discounted early versus who held — is where competitor strategy actually shows.
On priority SKUs, yes. Offers in this window routinely run for hours rather than days, and several distinct offers can appear on one SKU in a single day.
Not on everything, though. We tier it: hourly on the SKUs where a competitor move changes your decision, several times daily on the working set, daily on the tail. Hourly across a full catalogue is expensive and mostly confirms nothing happened.
Yes, comfortably. Production collection goes live in 5 to 10 business days after scoping, and a pilot returns real data within 48 hours.
Setting up in September or early October gives you a clean pre-window baseline, which is the part teams most often lose by starting late.
Yes — UK, EU, GCC, India and Australia all run this window with different intensity and different mechanics. Loyalty-gated pricing matters more in UK grocery; voucher stacking dominates in Southeast Asia.
A single global view of Black Friday is misleading. Market is a dimension on every record in our collection.
Get this running before the window opens
A pilot on your own SKUs returns real data within 48 hours, and production goes live in 5 to 10 business days. The part teams lose by starting late is the pre-window baseline.
Free pilot, no card, no obligation.
4.8/5 Average Rating
50+ Video Testimonials
92% Client Retention
50+ Countries Served
 377-7584 (USA).svg){kind=icon}
