Platform · Costco

Costco Data Scraping Services

Q: How do you handle the rotating assortment?

With a stable product_key maintained across item number changes plus first-seen, last-seen, days-in-range and reappearance counts. Costco item numbers do not persist the way retail SKUs do, so without maintained identity every rotation reads as a new product and range churn becomes unmeasurable. Reappearance in particular is a strong signal an item sold through well.

Where unit pricing is the only comparable field, because nobody else sells a 96-count pack.

Costco data scraping is the automated collection of publicly visible Costco data — pricing with unit price normalised across warehouse pack sizes, rotating assortment with appearance and disappearance detection, item numbers and own-brand classification — with member-gated pricing excluded and that boundary stated rather than worked around.

A Costco price is almost never directly comparable to a supermarket price, because the pack size is not comparable. Unit price is the field that makes the comparison possible, and it has to be computed rather than trusted.

Get a free sample See sample data

Free pilot on your own Costco list, returned in 24 hours. No card, no trial clock — and you keep the sample data either way.

costco_units.jsonl LIVE FEED

{"costco_item_number":"1204471", "product_key":"aw-cst-88120", "identity_confidence":0.96, "brand_type":"kirkland_signature", "price":24.99,"price_gated":false, "pack_description":"4 boxes of 24 ct", "pack_units_parsed":96, "unit_price_computed":0.26, "unit_basis":"per_count", "unit_price_displayed":"$6.25/box", "first_seen":"2026-03-11", "days_in_range":149, "reappearance_count":2, "warehouse_only":false} {"costco_item_number":"1209982", "price":"null","price_gated":true, "gated_reason":"member_signin_required", "unit_price_computed":"null"}

2 of 184,220 item rows · online cataloguepack parsed 93.4% · gated 11.2% · schema v2.6

Independence and trademarks. Actowiz Solutions is not affiliated with, endorsed by or connected to Costco or its owners. Costco and related marks belong to their respective owners, used here only to name the publicly accessible source this service collects from.

Our Data Powers

B2C Marketplace

amazon

D2C + Marketplace

NYKAA

D2C + Marketplace

Walmart

FMCG Marketplace

udaan

Food Delivery

Uber Eats

Quick Commerce

blinkit

Taxi Aggregator

Uber

E-Commerce

Tmall

Costco at a glance

How we handle Costco specifically

Platform-specific handling, not a generic retail template pointed at a different domain.

Platform: Costco online catalogue, with warehouse-only items noted where indicated
The critical field: Unit price computed by us, since pack sizes are not comparable to retail
Assortment: Rotating range with appearance and disappearance detection
Own brand: Kirkland Signature classified via maintained mapping
Membership: Member-gated pricing excluded, with the boundary stated in the data
Identity: Costco item number, which does not persist the way a retail SKU does
Refresh: Daily standard; more frequent during promotional periods
Region: United States, with other Costco markets on request

Platform specifics

What makes Costco data different from conventional grocery

These are the reasons a Costco dataset needs its own handling rather than a shared retail schema.

Unit price is the only comparable field

Costco sells in warehouse pack sizes that have no equivalent at a supermarket. Comparing a 96-count pack price to a 12-count pack price tells you nothing, and comparing the two headline prices is actively misleading.

Why we compute rather than trust

Displayed unit price bases are inconsistent — per item, per ounce, per pound, per 100 count — and vary within the same category.
Multipack maths is easy to get wrong: a pack of 4 boxes of 24 is 96 units, and the pack description does not always make that explicit.
Weight-based items need a different basis again, and mixing bases produces a category average that is arithmetic nonsense.

We parse the pack architecture and compute unit_price_computed on a basis we state per category, retaining unit_price_displayed separately where Costco shows one. Where the pack architecture cannot be parsed confidently, unit price is null with a reason rather than estimated — a wrong unit price on a warehouse pack is wrong by a large multiple, not a rounding error.

Assortment rotates, and that is the signal

Costco's range deliberately rotates. Items appear, sell through and disappear, and a meaningful share of the catalogue is not intended to be permanent.

This breaks assumptions that work at conventional grocers. A disappeared item is not a delisting to investigate; it is often the model working as designed. But it makes appearance and disappearance rates the most informative fields in the dataset.

Appearance detection shows what is entering the range and when.
Time in range before disappearance indicates sell-through on treasure-hunt items.
Reappearance is a strong signal an item worked.
Item numbers do not persist the way retail SKUs do, so identity has to be maintained across number changes.

We keep a stable product_key across item number changes with match confidence, alongside first_seen, last_seen and reappearance_count. Without that, every rotation reads as a new product and range churn is unmeasurable.

Member-gated pricing: what we will not do

Some Costco pricing and content requires a membership sign-in. We do not create memberships to reach it, and we do not use client credentials.

What that means in practice

We collect what an anonymous visitor sees, which on the online catalogue is a substantial and useful set.
Where a price is member-gated, the field is null with a reason code rather than substituted from another source.
Warehouse-only items are noted where the site indicates them, since they may have no online price at all.
In-warehouse pricing is not published online and we do not claim to have it.

This means Costco coverage is genuinely partial, and we say so during scoping rather than after. For clients who need a full picture, the honest answer is that public collection gets you the online catalogue with unit-normalised pricing and range rotation, and does not get you warehouse-floor pricing. Any vendor claiming otherwise is either logging in or making it up.

Scope

What we collect on Costco, and what we do not

The right column matters more than the left. Anyone can list fields; the limits are what tell you whether the dataset will hold up.

✅ What we collect

Unit price computed by us on a stated basis per category
Displayed unit price retained separately where Costco shows one
Pack architecture parsed, with null and a reason where it cannot be parsed confidently
Stable product identity across item number changes, with match confidence
Appearance, disappearance, time in range and reappearance counts
Kirkland Signature classified via maintained mapping
Warehouse-only indication where the site provides it
Category structure and online availability
Ratings and review text without reviewer profiles

❌ What we do not, and why

Member-gated prices, which are reported null with a reason code
Membership creation or use of client credentials to reach gated pricing
In-warehouse floor pricing, which is not published online
Estimated unit prices where pack architecture could not be parsed
Inventory quantities or warehouse-level stock

Core Costco fields

The full dictionary is agreed during scoping. These are the fields specific to this platform.

Field	What it is on this platform
costco_item_number	Costco item number as published, which does not persist reliably
product_key / identity_confidence	Our stable identity across item number changes, with confidence
price / price_gated	Price where public, and a flag where it requires membership
gated_reason	Why a price is unavailable, so a null is interpretable
pack_description / pack_units_parsed	Published pack text and the parsed unit count
unit_price_computed / unit_basis	Our computed unit price and the basis used, stated per category
unit_price_displayed	Costco's own unit price where shown, retained separately
brand_type	kirkland_signature or third_party via maintained mapping
first_seen / last_seen / days_in_range	Range entry, exit and duration for rotation analysis
reappearance_count	How many times an item has cycled back, a sell-through signal
warehouse_only	Where the site indicates an item is not sold online

Use cases

What teams do with Costco data

Like-for-like comparison against supermarkets

Unit price is computed on a stated basis with pack architecture parsed, making warehouse pack pricing comparable to conventional retail rather than misleading.

Range rotation and treasure-hunt analysis

Appearance, time in range and reappearance counts with stable identity across item number changes make range churn measurable instead of reading every rotation as a new product.

Kirkland Signature versus branded positioning

Own-brand classification with unit-normalised pricing supports the own-brand price index that warehouse pack sizes otherwise make impossible to compute.

Honest coverage assessment before commitment

Member-gated fields are reported null with reason codes, so the share of the catalogue that is publicly visible is measurable rather than assumed.

The 24-hour sample — run on your sources, not ours

Send us a Costco item or category list. We run real collection against it and return the output within 24 hours, with the platform-specific fields populated so you can check them yourself rather than take our word for it.

Real extraction from your actual sources
Returned inside two business days
Coverage and QA note included
You keep the data either way
No card, no trial clock
Named engineer on the call

Get my free sample Book a 20-min scoping call Reply within one business day. Reference calls available under NDA.

How we engage

Three ways to engage us

Same collection pipeline and QA underneath. The difference is who holds the schedule and how the data reaches you.

Managed service (most common)

We own the collection, the QA and the delivery. You receive clean data on a schedule and never touch a scraper.

Dedicated engineer assigned to your account
Site changes fixed by us, not reported to you
Scheduled delivery to your warehouse or S3
Named contact on Slack or email

Best fit: Teams who need the data, not the infrastructure.

API access

The same collection pipeline exposed as an authenticated REST endpoint your systems query directly.

On-demand and scheduled endpoints
Rate limits agreed to your load profile
Sandbox keys for integration testing
Versioned schema with deprecation notice

Best fit: Product and engineering teams building on live data.

One-time or project extraction

A defined pull for a specific question — market sizing, diligence, a pitch, a one-off audit.

Fixed scope agreed in writing upfront
Single delivery with full QA report
Methodology documented for your records
Converts to managed if you want continuity

Best fit: Research, strategy and diligence work with a deadline.

Pricing

Every engagement is quoted individually, because the honest answer depends on your scope: how many sources, how many records, how often, and how the data reaches you. We scope it with you, run a free pilot on your own sources, and then quote a fixed monthly figure — no per-request metering and no overage billing when volumes move. Request a quote and you will have a number after one call.

Costco is usually collected alongside its competitors

Almost nobody buys a single platform in isolation. Costco data becomes useful when it sits next to the competitor set on one schema, refreshed on one schedule, so a price index or availability comparison is genuinely like-for-like.

That is what grocery data scraping covers, and a Costco-only engagement can be expanded into it without rebuilding. If you already know you need several platforms, start there instead — it is the same pipeline and usually the better scoping conversation.

FAQ

Costco data scraping: frequently asked questions

Platform-specific questions, including what cannot be collected here.

Because displayed unit price bases are inconsistent — per item, per ounce, per pound, per 100 count — and vary within the same category. Mixing bases produces category averages that are arithmetic nonsense.

We parse pack architecture and compute on a basis stated per category, retaining Costco's displayed figure separately. Where the pack cannot be parsed confidently, unit price is null with a reason: a wrong unit price on a warehouse pack is wrong by a large multiple, not a rounding error.

No. We do not create memberships or use client credentials to reach gated pricing. Where a price is member-gated, the field is null with a reason code rather than substituted from elsewhere.

This makes Costco coverage genuinely partial and we say so during scoping. Public collection gets you the online catalogue with unit-normalised pricing and range rotation. It does not get you warehouse-floor pricing, and any vendor claiming otherwise is either logging in or inventing it.

With a stable product_key maintained across item number changes plus first-seen, last-seen, days-in-range and reappearance counts.

Costco item numbers do not persist the way retail SKUs do, so without maintained identity every rotation reads as a new product and range churn becomes unmeasurable. Reappearance in particular is a strong signal an item sold through well.

Usually not. Costco's range deliberately rotates and a share of the catalogue is not intended to be permanent, so disappearance is often the model working as designed.

That is why we report appearance and disappearance as events with duration rather than flagging them as delistings. Applying conventional grocery assumptions here produces a stream of false alarms.

No. It is not published online, so it is not collectable from public sources. We note warehouse_only where the site indicates an item is not sold online.

If in-warehouse pricing is essential to your analysis, public web collection is the wrong method and we would say so rather than sell you a partial dataset described as complete.

We quote individually. Drivers are category scope, refresh frequency, and whether unit-price normalisation is needed across the full catalogue — pack parsing is where the work concentrates.

A defined category at daily refresh sits at the lighter end. One scoping call, a free pilot on your own categories within 24 hours including a coverage figure showing what share is publicly visible, then a fixed monthly quote. Request a quote.

See real Costco data before you commit to anything

Send us an item or category list. We return the output within 24 hours with the platform-specific fields populated.

Get a free sample Talk to an engineer

Free pilot, no card, no obligation. If we cannot collect a field you need on this platform, the sample shows you that too.

Hear It Directly from Our Clients

Watch how businesses like yours are using Actowiz data to drive growth.

▶

1 min

★★★★★

"Actowiz Solutions offered exceptional support with transparency and guidance throughout. Anna and Saga made the process easy for a non-technical user like me. Great service, fair pricing!"

Thomas Galido

Co-Founder / Head of Product at Upright Data Inc.

▶

2 min

★★★★★

"Actowiz delivered impeccable results for our company. Their team ensured data accuracy and on-time delivery. The competitive intelligence completely transformed our pricing strategy."

Iulen Ibanez

CEO / Datacy.es

▶

1:30

★★★★★

"What impressed me most was the speed — we went from requirement to production data in under 48 hours. The API integration was seamless and the support team is always responsive."

Febbin Chacko

-Fin, Small Business Owner