Data type · Image enrichment

Product image enrichment — sourcing, with the rights question answerable

A catalogue with missing images converts worse and ranks worse, and most teams know exactly which SKUs are affected. The reason it stays unfixed is rarely sourcing. It is rights.

Product image enrichment is filling catalogue image gaps at scale: locating imagery per SKU across manufacturer and distributor sources, deduplicating, normalising to your target specification, and recording the source of every file. Finding an image is not the same as being licensed to publish it, and we do not imply otherwise — but we make the rights question answerable instead of unanswerable.

A catalogue with missing images converts worse and ranks worse, and most teams know exactly which SKUs are affected. The reason it stays unfixed is rarely sourcing. It is rights.

Get a free enrichment sample See sample data

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

✓ Source recorded on every file✓ Perceptual dedupe across sources✓ We never grant or imply usage rights

image_enrichment.jsonl LIVE FEED

// one record per sourced file per SKU {"client_sku": "EX-4471", "file_id": "img-9f3c1a22", "source_type": "manufacturer_site", "source_domain": "examplebrand.example", "source_url": "https://examplebrand.example/…/kettle-1.jpg", "retrieved_at": "2026-08-25T09:12Z", "perceptual_hash": "p:9f3c1a…", "width": 2400, "height": 2400, "background": "white", "is_primary_candidate": true, "meets_spec": true, "rights_cleared_by_actowiz": false} {"client_sku": "EX-4471", "file_id": "img-9f3c1a22", "duplicate_of": "img-9f3c1a22", "source_domain": "distributor-b.example", "note": "same asset, second source — kept, not merged, so you can pick the source you have rights from"} {"client_sku": "EX-9902", "file_id": null, "not_found_reason": "no_manufacturer_imagery_published"}

3 of 412,880 sourced filesfound for 87.2% of requested SKUs · source on every file · schema v1.0

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

Summary What it is What's included Schema Markets Who it's for Examples How we engage SLA FAQ

Written and reviewed by the Actowiz Solutions Data Engineering team Published 5 August 2026 Last updated 5 August 2026 Reading time 10 min

Key facts at a glance

What it is: Sourcing imagery for catalogue gaps, at SKU scale
What we supply: Files, plus the source of every one
What we cannot supply: The right to publish them. That is not ours to give
Why it still works: Most supplier agreements already grant retailer usage rights
Dedupe: Perceptual hash across sources, duplicates retained not merged
Normalisation: To your target dimensions, background and format
Not found: Reported with a reason, per SKU
Delivery: To your own bucket or PIM, never held longer than agreed

87%typical find rate on requested SKUsreported before you commit

100%of files carry their sourceso rights are answerable

0usage rights granted by uswe source, you clear

24hfree sample on your own SKU listwith the find rate stated

Key takeaways

What it is: Sourcing imagery for catalogue gaps, at SKU scale
What we supply: Files, plus the source of every one
What we cannot supply: The right to publish them. That is not ours to give
Why it still works: Most supplier agreements already grant retailer usage rights
Dedupe: Perceptual hash across sources, duplicates retained not merged
Normalisation: To your target dimensions, background and format

Last verified 5 August 2026 by the Actowiz Solutions Data Engineering team.

Definition

Why the rights question comes first on this page

Every other service on this site delivers data. This one delivers files that somebody created, and copyright in a photograph belongs to whoever took it.

What we will not do

We will not tell you an image is safe to publish. We are not your counsel, we did not write your supplier agreements, and a vendor implying otherwise is handing you a liability with a delivery note attached.

What we do instead, and why it is enough

We record the source of every single file — domain, URL, source type and retrieval timestamp. That turns an unanswerable question into an answerable one:

Manufacturer site, and you have a supplier agreement with that manufacturer? Your agreement almost certainly covers retailer usage. Your legal team can confirm in minutes because they know which manufacturer.
Authorised distributor? Same question, one step removed, and now askable.
An unknown third party? Now visible, and excludable, rather than silently in your catalogue.

Most brands' supplier agreements already grant retailer image usage. The reason catalogues stay unfixed is not that rights are absent — it is that nobody can tell which image came from where. That is the problem this solves.

rights_cleared_by_actowiz is a constant false on every record, so nothing downstream can mistake sourcing for clearance.

What we deliver

Six components of an enrichment run

Sourcing is the easy part. The rest is what makes it usable.

Sourcing per SKU

Across manufacturer and distributor sources.

Manufacturer sites first, distributors second
Matched on identifier where available, attributes where not
Multiple candidates per SKU, ranked
Not-found reported with a reason

Source provenance

The field that makes the rights question answerable.

Source domain, URL and type on every file
Retrieval timestamp
Source type — manufacturer, distributor, retailer, unknown
rights_cleared_by_actowiz constant false

Deduplication

Perceptual, not filename-based.

Perceptual hash across all sources
Duplicates retained with a link, not merged
So you can choose the source you hold rights from
Near-duplicates flagged separately from exact

Specification compliance

Against your standard, not ours.

Your target dimensions, format and background
meets_spec flag per file
Resolution and aspect recorded before normalisation
Upscaling never applied silently

Normalisation

Only what you ask for, and reversibly.

Resize and format conversion to your spec
Background handling where requested
Original retained alongside the normalised file
No retouching, no compositing, no generation

Delivery and retention

Into your systems, on your terms.

Delivered to your bucket or PIM directly
Manifest with source, hash and spec status per file
Retention window agreed in writing
Deleted from our side on completion unless you say otherwise

Service scope

What the ecommerce data scraping service includes

A managed engagement, not a tool licence. We own the pipeline and everything that breaks in it.

✓ Included in every engagement

Sourcing imagery per SKU across manufacturer and distributor sources
Source type, domain, URL and retrieval timestamp on every file
Perceptual deduplication with duplicates retained and linked
Normalisation to your stated dimensions, format and background
Specification compliance flagged per file, with originals retained
Not-found reported per SKU with a reason
Delivery into your own bucket or PIM, with agreed retention
Source discovery, scoping and a written collection plan
Free pilot on your own sources before any commitment
Full pipeline build, hosting and proxy infrastructure
Schema design, validation and sampled human QA on every run
Ongoing maintenance when source layouts change — our cost, not yours
Delivery to your warehouse, bucket, SFTP or API endpoint
Documented methodology and compliance notes for your legal review

× Not included — stated upfront

Any grant, clearance or implication of usage rights
Retouching, compositing or generation of product imagery
Silent upscaling of files below your specification
Imagery behind a login or a paywall
Retention beyond the window agreed in writing
Anything behind a login, paywall or credentialed session
Personal data beyond a documented lawful basis
Licensed third-party datasets we do not hold rights to
Guarantees about fields a source simply does not publish

Schema

Enrichment manifest fields

One record per sourced file per SKU. The manifest is the deliverable as much as the files are.

Image enrichment manifest v1.0
Field	Type	What it captures	Refresh
`client_sku / product_key`	string	Your identifier and our matched identity	Every record
`file_id`	string	Our identifier for the file, null where nothing was found	Every record
`source_type`	string	manufacturer_site, distributor, retailer or unknown	Found files
`source_domain / source_url`	string	Where it came from. The field that makes rights answerable	Found files
`retrieved_at`	timestamp	When we retrieved it	Found files
`perceptual_hash`	string	For dedupe across sources and for your own matching	Found files
`duplicate_of`	string	Links to the first instance, so both sources stay visible	Duplicates
`width / height / format / background`	number / string	As retrieved, before any normalisation	Found files
`meets_spec`	boolean	Against your stated standard	Found files
`is_primary_candidate`	boolean	Our suggestion for the lead image, not a decision	Found files
`not_found_reason`	string	Why nothing was sourced for this SKU	Not-found records
`rights_cleared_by_actowiz`	constant	Always false. We source; you clear	Every record

The manifest is designed to be joined to your supplier agreement list. Source domain is the join key that makes that possible, and it is the reason this service works at all.

Coverage

Where imagery is findable, and where it is not

Find rates differ enormously by category, and we report yours before quoting.

Consumer electronics — high, manufacturers publish extensivelyHome appliances — highIT hardware — highIndustrial and MRO — high on branded lines, low on genericsAutomotive parts — moderate, fitment images varyBeauty — high on brands, low on private labelFashion — moderate, seasonal ranges disappear quicklyFurniture — moderate, room sets complicate matchingGrocery and FMCG — moderate, pack shots often via distributorsToys — moderateBooks and media — high via identifiersPrivate label and own brand — low, by definitionHandmade — not applicable, each item is uniqueDiscontinued lines — low, imagery is withdrawn with the product

Private label is the honest weak spot. If a product exists only under your own brand, there is no manufacturer source to enrich from, and the answer is photography rather than sourcing. Request a source we don't list →

Markets served

Countries and markets where this service is in highest demand

We deliver into 40+ countries. These are the markets where this particular service is requested most, and the reason demand concentrates there.

Highest-demand markets for this service, and why demand concentrates there
Market	Why demand concentrates here

North America

United StatesCanadaMexico

United Kingdom & Ireland

United KingdomIreland

Western Europe

GermanyFranceNetherlandsBelgiumSpainItalySwitzerlandAustria

Nordics

SwedenNorwayDenmarkFinland

Middle East

United Arab EmiratesSaudi ArabiaQatarKuwaitIsrael

Asia Pacific

SingaporeAustraliaNew ZealandJapanSouth KoreaMalaysiaIndonesiaThailandVietnamPhilippines

South Asia

IndiaBangladeshSri LankaPakistan

LATAM

BrazilArgentinaChileColombia

Africa

Wants retailers using approved assets rather than substitutes.

What we deliver

The reverse view — where retailers are using imagery that is not yours.

Metric that moves

Asset compliance

Use cases

How enrichment gets used

Four patterns, and the last one runs in reverse.

Filling known catalogue gaps at scale

Sourcing per SKU across manufacturer and distributor sources with a find rate reported per category, so a catalogue completeness project is scoped on real numbers rather than hope.

Outcome: Gaps closed on the SKUs where imagery exists, and honestly identified where it does not.

Making the rights question answerable

Source domain on every file, joined to your supplier agreement list, turning 'can we use this' from an investigation into a lookup.

Outcome: Legal sign-off in hours rather than a project that stalls.

Marketplace listing compliance

Files normalised to a marketplace's dimension, background and format requirements, with meets_spec flagged so rejections are predictable.

Outcome: Listings that go live first time.

Reverse check: are retailers using your assets?

For a brand, the same perceptual hashing run against retailer listings shows who is using approved imagery and who has substituted their own.

Outcome: An asset compliance picture most brands cannot otherwise get.

Engagement examples

Two engagements, anonymised

Clients rarely permit naming. These are real engagement shapes with identifying detail removed, so you can judge whether the work resembles your situation.

Examples are anonymised at client request. Named references are available on request under NDA. See published case studies →

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

Before you commit to anything, we run this service against your own sources and send you the output. If the coverage isn't there, the sample will show you that too — which is the point. We would rather lose the deal at the pilot than at month three.

Real extraction from your actual sources
Returned within 24 hours
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.

Build vs buy

Source imagery in-house, commission photography, or enrich?

All three are right in different places, and enrichment is not always the answer.

In-house build vs self-serve tool vs Actowiz managed service
Consideration	In-house scraping team	Generic proxy / DIY tool	Actowiz managed feed
Time to first usable data	6–12 weeks of engineering before anything is trustworthy	Days, but output needs manual cleanup before use	Free pilot in 24 hours, production in 5–10 business days
Who fixes it when a source changes	Your engineers, at the cost of their roadmap	You do — tools report failures, they don't resolve them	We do, same business day, inside the retainer
Data quality assurance	Whatever your team has time to build	None beyond HTTP success	Schema validation plus sampled human QA on every run
Compliance documentation	Rarely produced, then requested urgently by legal	Not provided; terms risk sits with you	Sources, method and lawful basis documented for review
Accountability	Distributed across a team with other priorities	A support ticket queue	A named engineer and an account owner
True annual cost	Engineer salaries, proxies, hosting, ongoing maintenance	Low licence fee plus significant hidden analyst time	One fixed monthly retainer, quoted after scoping

Duplicates are kept, not merged, and that is deliberate

The same manufacturer photograph frequently appears on the manufacturer site, on two distributor sites and on three retailer listings. A naive dedupe keeps one and discards the rest.

We keep them all, linked by duplicate_of, because which source you got it from is the whole question.

You may hold a supplier agreement with the manufacturer and not with the distributor.
You may hold one with a distributor and not with a retailer.
The file is identical; your right to use it is not.

Discarding duplicates would throw away exactly the information that makes the deliverable usable. So the manifest shows every source we found an asset at, and you pick the one you can stand behind.

Near-duplicates are flagged separately from exact matches, because a lightly recropped version is a different file with potentially a different creator.

What we will not do to an image

Normalisation means resizing, format conversion and background handling to your stated specification. It does not mean improving the picture.

Not done, in any circumstance

No retouching. Colour, shape and finish must remain what the source showed, because a shopper will compare the delivered product to the image you published.
No compositing. We do not assemble a product into a scene it was not photographed in.
No generation. We do not create imagery for a product that has none. A generated product image is a misrepresentation, whatever the disclaimer says.
No silent upscaling. Where a source image is below your specification we flag meets_spec: false and leave it. Upscaling a 400px image to 2000px produces a file that passes a dimension check and looks worse than having no image.

The original is retained alongside any normalised version, so a normalisation decision is reversible and auditable.

How it works

How an enrichment run works

Send the SKU list. The find rate comes back before the quote.

Send the SKU list and your image specification

Identifiers where you have them, attributes where you do not, plus target dimensions, format and background.

We report the find rate per category

Before quoting. Branded electronics runs high; private label runs near zero, and that is worth knowing before commissioning a project.

Free sample within 24 hours

Real sourced files for a subset of your SKUs, with the full manifest, so your legal team can see the provenance before you commit.

You clear rights against the source list

Join source_domain to your supplier agreements. That is a lookup rather than an investigation, and it is the step that makes the whole thing work.

Delivery into your bucket or PIM

With the manifest, originals retained, and our copies deleted on completion unless you have asked otherwise in writing.

Formats & destinations

Files delivered to your own Amazon S3, Google Cloud Storage or Azure Blob bucket, or pushed into your PIM. Manifest as JSON, JSONL, CSV or XLSX.

Compliance & data ethics

We collect only publicly accessible imagery and record the source of every file. Copyright remains with the rights holder; we do not grant, clear or imply any usage right, and rights_cleared_by_actowiz is set false on every record. Retention on our side is agreed in writing and files are deleted on completion by default.

Service commitments

What we commit to, in writing

These are contractual, not marketing copy. They appear in the engagement document.

Service level commitments written into every managed engagement
Commitment	What we hold ourselves to
Pilot turnaround	A real sample from your own sources within 24 hours of scoping, at no cost.
Go-live	Production collection running within 5–10 business days of sign-off.
Delivery punctuality	99.5% on-schedule delivery, measured monthly and reported to you.
Breakage response	Source layout changes triaged same business day; critical sources inside 4 hours.
Data quality	Schema validation on every run plus sampled human QA before any delivery leaves us.
Escalation	A named engineer and an account owner, not a shared ticket queue.
Change requests	Field additions and source changes handled inside the retainer, not re-quoted.
Exit	Your historical data exported in full on request. No lock-in, no export fee.

Why teams pick Actowiz for this work

Engineers, not a dashboard. You get people who fix breakages, not a self-serve tool you maintain yourself.
We tell you what we can't do. Scope limits and coverage gaps are stated before you sign, not discovered in month three.
QA is part of the service. Schema validation and sampled human review run before delivery, every run.
Compliance is documented. Sources, method and lawful basis written down so your legal team can review them.
Fixed monthly cost. No per-request metering, no surprise overage on a month when a competitor adds SKUs.
Six years, 40+ countries. Long-running production pipelines across retail, travel, mobility and finance.

Definitions

Terms used on this page

Plain definitions of the terms used on this page, so procurement and legal reviewers are working from the same vocabulary as your data team.

FAQ

Image enrichment: frequently asked questions

The rights questions come first because they are the ones that matter.

That depends on your agreements, and we cannot answer it for you. What we can do is record the source of every file — domain, URL, type and timestamp — so your legal team can answer it quickly.

Most retailer supplier agreements already grant image usage. The reason catalogues stay unfixed is usually that nobody can tell which image came from where, and that is exactly what the manifest solves.

Because which source you obtained a file from is the whole rights question. The same photograph on a manufacturer site and a distributor site is one file and two different permission situations.

We link duplicates with duplicate_of so the manifest shows every source we found it at, and you pick the one you can stand behind.

No. A generated product image is a misrepresentation of a physical object a customer will receive, whatever disclaimer accompanies it.

Where no imagery exists — which is the normal case for private label — the answer is photography, not sourcing, and we will say so rather than quoting for something that will find nothing.

No to both. Colour, shape and finish must remain what the source showed, because a shopper compares the delivered product to the published image.

Where a file is below your specification we flag meets_spec: false and leave it. Upscaling produces a file that passes a dimension check and looks worse than no image at all.

It depends heavily on category. Branded electronics and appliances run high; private label runs near zero because there is no manufacturer source to enrich from.

We report the rate on your own SKU list before quoting, so a completeness project is scoped on real numbers.

That one is monitoring — watching what imagery retailers publish, detecting changes, checking whether your approved assets are in use. This one is acquisition, filling gaps in your own catalogue.

Different purpose, different deliverable, and this one carries a rights question that monitoring does not.

Only for the retention window agreed in writing, and by default we delete our copies on completion. Files are delivered into your own bucket or PIM rather than held on a platform you access through us.

Yes, and it is a strong use. The same perceptual hashing run against retailer listings shows which retailers are using your approved assets and which have substituted their own.

That is a fixable digital shelf finding most brands cannot otherwise see.

We quote individually. Drivers are SKU count, category mix and whether normalisation and delivery into your PIM are in scope — category mix matters most because find rates vary so widely.

One scoping call, a free sample within 24 hours with the find rate per category, then a fixed quote. Enrichment is often a project rather than a subscription. Request a quote.

Send a SKU list and see the real find rate

We return sourced files for a subset with the full provenance manifest, so your legal team can assess before you commit.

Get a free enrichment sample Book a 20-min scoping call

No sales sequence. If your catalogue is mostly private label, we will tell you enrichment will find little and photography is the answer.

Enterprise Data Extraction

AI Training Data

24-hour sample

AI Dynamic Pricing

Social Commerce

GCC Quick Commerce

KitchenIntel

UK Grocery Price Tracker

What we can collect

Sample data

Ready-Made Scrapers

API Playground

Need it managed instead?

Case Studies

Sample Datasets

ROI Calculator