Grocery chains under measurement, each counted once however many ways it sells. Eight marketplace and delivery platforms are measured alongside them, for 31 retailer identities in all.
Investors
Independent price measurement for a market that runs on price.
Basket Bee is past the hardest technical risk. Measurement across 23 grocery chains and 8 marketplace and delivery platforms, a working layer that resolves rival retailers' listings into the same product, and a licensed access model for the brands it is built for are all in place. What remains is a customer-facing portal, moving collection onto a schedule, and the commercial work of selling it.
Listings carrying a price in the most recent complete capture, counted exactly rather than estimated. 99.8% of the capture's listings carry one.
The figure that decides whether a price can be compared between rivals: distinct products with a price at two or more different retailers. The catalog holds far more products than this; most are sold by only one of them.
Catalog as of 2026-08-23 · exact counts over that capture, not estimates
The investor case.
The opportunity is not a price-comparison screen. It is that the brands competing in this market need product-level measurement they do not control, and that the system which produces it here is already running end to end: 23 grocery chains and 8 marketplace and delivery platforms, resolved into one comparable catalog, with every past capture retained.
The data asset exists
23 grocery chains and 8 marketplace and delivery platforms under measurement, and 359,260 priced listings in the most recent capture — built and held, not projected. Every capture is retained, so the catalog carries history rather than a single snapshot.
Resolution, not scraping
The value is not the collection. It is that mixed Arabic-English listings from rival retailers resolve into the same product, which is the step that makes the data sellable.
Accuracy with a stated scope
Match quality is checked against barcode evidence the matcher never sees. Only a minority of matches admit that check at all, and the share that cannot be checked is disclosed rather than hidden inside an average.
Both channels, measured not assumed
Ten of the retailers are measured on both the in-store and the delivery channel, which is what makes it possible to quantify a channel price difference rather than assume one.
Built for brands, not for retailers
The product is designed for FMCG suppliers and their agencies. Retailers are the subject of measurement and are not the intended buyer, so the commercial model would not depend on the cooperation of the companies being measured.
Boundaries written down
Collection operates within defined limits, with the stop conditions recorded in advance and reviewable.
The frontier we operate on.
Turning every major Egyptian grocer into one comparable market view is a genuinely hard problem, and how well it is solved is what separates the products in this market. The difficulty is not access to prices. It is that rival retailers describe the same product in different languages, at different pack sizes, with nothing in common to join on.
Breadth and depth in one view
23 grocery chains and 8 marketplace and delivery platforms, each measured on the channels it actually sells through and 10 of them on both at once, resolved into one comparable catalog of 359,260 priced listings — with every past capture kept, so the catalog carries history rather than a single snapshot.
Cross-language product equivalence
Off-the-shelf approaches fail on mixed Arabic and English grocery data. Ours links the same item across retailers that describe it completely differently, and leaves an auditable trail for every decision it makes.
Channel economics, measured
For the 10 retailers observed on both the in-store and the delivery channel, the price difference between the two can be quantified rather than estimated: which channels carry a premium, and which sell at parity. That comparison needs both sides of the same retailer, which is why coverage breadth alone does not produce it.
Accuracy with its limits disclosed
Match quality is scored against barcode evidence deliberately hidden from the matcher, so the check is independent of the judgement being checked. Only a minority of matches carry clean barcodes on both sides and can be checked at all — the rest have no ground truth, and that share is disclosed rather than averaged away. No accuracy rate is published here until it has been measured against the current catalog rather than an earlier one.
The market, and why now.
Egypt’s grocery spend is enormous, recurring, and newly price-sensitive. The conditions that make independent price measurement valuable are in place now, and the depth of history that makes it decision-grade accrues only by measuring week after week.
The largest recurring spend
Groceries are the biggest repeating household outlay in Egypt, which makes shelf pricing the most consequential and most frequently changing data in consumer goods.
Inflation made price the decision
Sustained food inflation turned pricing from a background variable into the decision itself — for shoppers choosing where to buy, and for brands defending position.
Brands are measuring blind
Suppliers largely see what retailers choose to tell them. Independent measurement of what is actually on the shelf changes what they can negotiate with.
A position built on breadth and history
Others measure Egyptian grocery prices. What compounds is the combination: 31 retailer identities read together, rival listings resolved into one comparable product, and every past capture retained. Breadth can be bought; a back-catalog of prices cannot be bought late.
No market-sizing figure is stated on this page. None has been independently derived, and an unsourced number is worth less than an acknowledged gap.
A staged business model.
Brands and their agencies are the first customers the product is being taken to. Retailers are the subject of measurement and never the buyer, so the model would not depend on the cooperation of the companies being measured.
Licensed intelligence
Scoped access to the measured catalog, sold to brands, suppliers and agencies. This is the first commercial product and the current focus; no accounts have been signed yet.
Deeper analytics
Alerting, reporting and comparison tooling layered on the same measurement, sold into the accounts the first product opens.
Category intelligence
Aggregated market movement and promotion signals as a distinct product line for analysts and market builders.
Where the build stands.
The measurement layer and the client access model are built. The current stage is making operations release-grade — moving collection from operator-run to scheduled — and putting the licensed product in front of its first customers.
The measurement layer
Broad multi-chain collection, cross-retailer product resolution, offer modelling, uncertainty review, controlled publication, and the operator console that runs all of it.
The licensed product
Taking the client data model, scoping and packaging from a controlled environment to a customer-facing intelligence portal, and moving collection from operator-run to scheduled.
First revenue, then breadth
Land the first licensed accounts, then widen category depth and the analytics sold on top of the same measurement.
The ask.
The hardest technical risk is behind us: the measurement fleet, the cross-retailer resolution layer, and the licensed access model are built and running. The raise funds the move from a founder-run pipeline to a scheduled, customer-facing intelligence product.
Move collection from operator-run to scheduled and monitored, so coverage and freshness are guaranteed rather than supervised.
Take the licensed intelligence product to market and land the first brand and agency accounts.
First engineering and data hires around the founder-built core, to remove key-person risk.
Widen coverage depth and the analytics layer sold on top of the same measurement.
Pre-seed — Round size, terms and allocation are being finalized with counsel. No figures are stated here because none have been set.
Founder-built, engineering-deep.
Basket Bee is led by two Co-CEOs, Marwan Abdelmaged (BSc Computer Science) and Adam Nasser (BSc Business Accounting). The measurement fleet, the cross-retailer resolution layer, the operator control plane and the licensed access model were designed and built in-house. That depth, rather than a slide deck, is the credibility.
What we would show in a meeting.
A live walk-through rather than a deck: the measured catalog and where its coverage is strong and weak, the auditable trail behind the matching decisions, how accuracy is measured and where it cannot be, and the operator console running a real collection cycle end to end.