Capability · Bespoke extraction

Custom Data Extraction Services

For the sources and schemas nobody has a template for.

Custom data extraction is bespoke collection scoped to your specific sources, fields and data model — the requests that no pre-built scraper covers, no standard schema fits, and no self-serve tool handles. Scope is agreed in writing before build, and delivered as a one-time project or an ongoing feed.

Most of what we build is not on a menu. A client needs eleven obscure regional sources joined to their internal product hierarchy, delivered in a schema their warehouse already expects. That is not a template problem, and pretending it is produces a dataset nobody can use.

Get a free feasibility assessment 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.

✓ Scope agreed in writing first✓ Your schema, not ours✓ Free feasibility assessment in 24 hours

custom_scope_and_output.jsonl LIVE FEED

// scope record agreed before any build {"engagement":"custom-2026-0412", "sources":11, "source_types":["regional_portal","pdf_catalogue", "legacy_html","xls_download"], "feasible":9,"partial":1,"not_feasible":1, "not_feasible_reason":"login_required_no_lawful_route", "target_schema":"client_pim_v4", "join_key":"client_internal_sku", "match_method":"gtin_then_attr_similarity", "delivery":"weekly_parquet_to_s3"} // output in the client schema, not ours {"client_internal_sku":"NW-44182", "supplier_ref":"6204-2RS", "net_price_gbp":4.85,"moq":50, "lead_days":7,"source_ref":"pdf:cat-2026.pdf#p118", "match_confidence":0.96}

scope record + 1 output recordfeasibility assessed before quoting

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

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: Bespoke extraction scoped to your sources, fields and data model rather than a standard product
Typical requests: Obscure regional sources, legacy sites, PDF and spreadsheet catalogues, non-standard formats
Your schema: Output delivered in your existing data model, joined on your internal keys
Feasibility first: Every source assessed and classified feasible, partial or not feasible before quoting
Engagement shape: One-time project or ongoing feed, whichever the requirement is
Documentation: Scope, method and limits recorded in writing before build starts
What we decline: Sources with no lawful route, stated plainly rather than attempted
Who it's for: Teams whose requirement does not fit any vendor's catalogue

Per-sourcefeasibility classificationbefore you commit

Your schemaoutput in your data modelnot ours

Written scopebefore build beginsno scope drift

Project or feedwhichever fitsnot forced monthly

Key takeaways

What it is: Bespoke extraction scoped to your sources, fields and data model rather than a standard product
Typical requests: Obscure regional sources, legacy sites, PDF and spreadsheet catalogues, non-standard formats
Your schema: Output delivered in your existing data model, joined on your internal keys
Feasibility first: Every source assessed and classified feasible, partial or not feasible before quoting
Engagement shape: One-time project or ongoing feed, whichever the requirement is
Documentation: Scope, method and limits recorded in writing before build starts

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

Definition

What is custom data extraction, and when do you need it?

Custom data extraction is what we build when the requirement does not match a product. It covers the sources nobody has templated, the formats that are not web pages, and the schemas that exist only inside your business.

The requests that come here

Obscure or regional sources. A distributor portal in one country, a municipal register, a trade body directory. Real commercial value, zero chance of a pre-built scraper existing.
Non-web formats. PDF catalogues, spreadsheet downloads, legacy systems that emit fixed-width files.
Your data model. Output that joins to your internal product hierarchy on your keys, in the schema your warehouse already expects, so nothing downstream changes.
Unusual field combinations. Fields that exist on a page but no standard schema captures.
One-time questions. A market sizing, a diligence exercise, a pitch — where an ongoing feed would be waste.
Sources that defeated an in-house attempt. Often the honest starting point, and usually the most interesting.

Why we assess feasibility before quoting

Custom work fails most often through optimistic scoping rather than through engineering. A source list of eleven typically contains nine that are straightforward, one that is partially achievable, and one that has no lawful route at all.

Quoting the eleven as though they are equivalent produces a project that misses its scope. So we classify each source before quoting: feasible, partial with the limitation stated, or not feasible with the reason. That assessment is free and it is the deliverable of the first 24 hours.

What "not feasible" usually means

Almost always one of three things: the data sits behind a login with no lawful route, the source publishes nothing that contains the field you want, or a licence covers the content. It rarely means technically difficult — difficulty is a cost question, not a feasibility one.

What we take on

Six kinds of custom engagement

The common thread is that no catalogue entry fits, so scope is written rather than selected.

Obscure & regional sources

Real value, no template.

Regional distributor and dealer portals
Municipal and local registers
Trade body and association directories
Single-country platforms
Legacy sites with no modern structure

Non-web formats

Where the data is not a web page.

PDF catalogues and price lists
Spreadsheet and CSV downloads
Scanned documents requiring OCR
Fixed-width and legacy exports
Email-delivered reports where permitted

Your data model

Output that fits what you already have.

Delivery in your existing schema
Joins on your internal keys
Your category and attribute taxonomy
Your null and unit conventions
No downstream rework required

Matching & enrichment

Connecting external data to internal records.

Match to your product or company master
Confidence-scored matching with your threshold
Attribute-level enrichment
Deduplication against your existing records
Unmatched records reported not dropped

One-time projects

Where an ongoing feed would be waste.

Market sizing and landscape studies
Diligence support with documented method
Pitch and proposal data
One-off audits
Fixed scope, single delivery, QA report

Rescue engagements

Sources that defeated a previous attempt.

Assessment of why an in-house build failed
Rebuild with documented method
Migration onto your existing schema
Parallel running against your current output
Handover documentation if you internalise later

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

Free per-source feasibility assessment before any quote is given
Written scope document with fields, absence behaviour and out-of-scope list
Output in your schema and units, joined on your internal keys
Unmatched records reported rather than silently dropped
Project pricing where the requirement is genuinely one-time
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

Sources behind logins, paywalls or credentialed portals
Fields that no source publishes, such as competitor cost or unit sales
Licensed content where a licence rather than extraction is the route
Quoting sources before they have been assessed for feasibility
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

What you receive

Two artefacts: the written scope agreed before build, and output in your own schema. The scope document is what prevents drift.

Custom engagement deliverables — structure varies by engagement
Field	Type	What it captures	Refresh
`Written scope document`	artefact	Sources, fields, feasibility classification per source, method and stated limits	Before build
`Per-source feasibility`	enum	feasible, partial or not_feasible, with the reason recorded for each	Before quoting
`Target schema definition`	artefact	Your data model, field names, types, units and null conventions	Before build
`Join key and match method`	string	Which of your keys we join on and how matching is performed	Before build
`Output records`	your schema	Delivered in your data model rather than ours, so nothing downstream changes	Per delivery
`match_confidence`	decimal	Where matching to your records is involved, a confidence score you threshold	Per record
`source_ref`	string	Document, page or URL reference for traceability, particularly on document extraction	Per record
`unmatched_report`	artefact	Records that could not be matched to your master, listed rather than silently dropped	Per delivery
`QA report`	artefact	Coverage, fill rates and validation results per delivery	Per delivery
`Method documentation`	artefact	How each source is accessed and what is excluded, for your legal file	Per engagement
`Handover pack`	artefact	Collection design documentation, provided on request if you internalise later	On request

Unmatched records are reported rather than dropped. A custom feed that silently discards what it could not match to your master looks cleaner and hides the exact population you need to investigate.

Coverage

What we have built custom extraction for

Illustrative rather than exhaustive. If your requirement is not here, that is normal for this service.

Industrial and MRO supplier cataloguesRegional distributor portalsMunicipal and planning registersTrade association directoriesLegacy B2B ordering systemsPDF price lists and tariff cataloguesScanned document sets requiring OCRSpreadsheet download portalsNon-English single-country platformsAuction and tender archivesMembership and accreditation registersPublic sector open data with awkward formatsInternal product master matchingCompany master enrichment

Every engagement begins with a per-source feasibility assessment. Where a source has no lawful route we say so and remove it from scope rather than quoting it and delivering nothing. 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
Germany & United Kingdom	Large industrial and B2B supplier bases made up of regional portals and PDF catalogues that no vendor has templated.
United States	Broad demand for one-time project work supporting diligence, market sizing and corporate development.
European Union	Multi-country regional sources and municipal registers where single-country platforms require bespoke handling.
India & Southeast Asia	Fast-growing businesses with awkward legacy sources and multilingual catalogues that defeat standard templates.

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

Research needs sources that are public but awkward, with reproducible methodology.

What we deliver

Documented extraction with method recorded, source references retained and limits stated for publication.

Metric that moves

Reproducibility

Use cases

How custom engagements typically run

Four shapes, with the outcome each is judged on.

Long-tail source coverage in your own schema

A defined list of awkward sources is assessed for feasibility, built, and delivered in your existing data model joined on your internal keys, with unmatched records reported.

Outcome: Sources covered without downstream rework, and without a vendor schema to translate.

Document catalogue extraction

Supplier PDF catalogues and price lists are extracted with page-level references retained and matched to your SKU master with confidence scoring.

Outcome: Varied documents turned into one comparable dataset with every figure traceable.

Fixed-scope one-time project

A market sizing or diligence question is scoped in writing, delivered once with a QA report and documented method, priced as a project.

Outcome: The question answered without committing to an ongoing feed nobody needed.

Rescue of a failed in-house build

We assess why the previous attempt failed, rebuild with documented method, and run in parallel against your existing output so the comparison is direct.

Outcome: A working source and an explanation, rather than a second attempt at the same wall.

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.

Industrial distributor · DE

Eleven sources were quoted as equivalent and two could not be delivered

Situation

A previous vendor quoted a full source list without assessment; two sources sat behind logins with no lawful route and the project missed scope.

What we ran

Per-source feasibility assessment before quoting, classifying nine feasible, one partial with stated coverage limits and one not feasible with the reason.

Result

Scope was agreed against what was actually deliverable, and the infeasible source was removed rather than billed.

Manufacturer · UK

External data always needed a translation layer

Situation

Every new vendor delivered its own schema, so each source added downstream mapping work the data team maintained indefinitely.

What we ran

Output delivered in the client PIM schema with their field names, units and null conventions, joined on their internal SKU key with confidence scoring.

Result

New sources stopped requiring downstream rework, and unmatched records were reported rather than silently dropped.

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 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.

Build vs buy

Custom build, in-house attempt, or a standard product?

If a standard product covers your requirement, buy that. Custom is for when it does not.

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

Why the written scope document matters more than the code

Custom data work fails through scope ambiguity far more often than through engineering difficulty. Both sides agree enthusiastically at the start, build proceeds, and the delivered output is not what the client pictured.

Where the ambiguity hides

Which sources, exactly. "All the major distributors" is not a source list. Eleven named domains is.
Which fields, and what counts as present. A price that appears only after selecting options is a different field from a listed price.
What happens when a field is absent. Null, omitted record, or estimated? These produce very different datasets.
Which schema. Ours or yours, with whose field names, units and conventions.
What matching is expected. Joined to your master on which key, at what confidence, and what happens to unmatched records.
What completeness means. Every product on the site, or every product in a category, or every product matching your master.

What the document contains

Every source named with its feasibility classification. Every field with its definition and absence behaviour. The target schema with field names, types and units. The join key and match method with a confidence threshold. Delivery format and cadence. And an explicit list of what is out of scope, which is the section that prevents most disputes.

It is agreed before build starts and it is short — usually two to three pages. Clients occasionally find the process pedantic at the time. Nobody has regretted it at delivery.

Feasibility: the three reasons we decline a source

We assess every source before quoting, and roughly one in ten comes back not feasible. That assessment is free, and it is more useful than a quote for work that cannot be delivered.

The three reasons

No lawful route. The data sits behind a login, a paywall or a credentialed portal. We do not create accounts or use client credentials, so there is no route we will take. This is the most common reason.
The field does not exist publicly. The client wants supplier cost, competitor margin or unit sales. The source publishes none of these, and no amount of extraction produces data that was never published.
Licensed content. Exchange market data, MLS listings, licensed sports feeds, commercial patent databases. These are licensed products, and a licence is the route rather than extraction.

What "partial" means

Between feasible and not feasible sits partial, and it is worth understanding because it is common. A source might publish prices for two thirds of its catalogue, or expose availability only in a basket flow we will not use, or publish a field inconsistently.

Partial sources are quoted with the limitation stated in the scope document — expected coverage, which fields will be sparse, and why. That way the gap is a known parameter rather than a delivery surprise.

Why we would rather decline

A quote that includes an infeasible source produces a project that misses scope, an unhappy client and a refund conversation. Declining one source in eleven costs a fraction of the engagement and keeps the other ten credible. For sources that are feasible but genuinely hard, see AI-powered scraping — variety and awkward layouts are often solvable even where templates are not.

How it works

How a custom engagement runs, from assessment to delivery

Feasibility assessment comes first and is free. Nothing is quoted until every source has been classified.

Scope the sources and fields

You send us target sites, regions, SKUs or keywords. We return a field-level schema proposal, coverage estimate and refresh recommendation — usually within two working days.

Pilot sample, free

We extract a real sample from your actual targets so you can inspect field fill rates, edge cases and match quality before any commitment.

Production build and QA harness

Our engineers build extractors, then wire validation rules: type checks, range checks, duplicate detection and golden-record comparison against a manually verified subset.

Scheduled delivery into your stack

Feeds run at your chosen cadence and land in the warehouse or bucket you already use. Schema changes are versioned and announced before they ship.

Ongoing monitoring and SLA support

We watch coverage drift, fill rates and source changes daily. A named engineer owns your account, and layout breaks are fixed by us — not queued for you.

Formats & destinations

JSON, JSONL, CSV, Parquet or XLSX, delivered to Amazon S3, Google Cloud Storage, Azure Blob, SFTP, Snowflake, BigQuery, Databricks or a REST/GraphQL endpoint. Webhooks fire on completion, and every batch ships with a manifest containing row counts, schema version and QA results so your pipeline can fail loudly instead of silently ingesting a bad file. Output is delivered in your schema and format rather than ours, including your field names, units and null conventions.

Compliance & data ethics

Custom engagements operate under the same boundaries as our standard services: publicly accessible sources only, no credentialed access, no personal data resale, no licensed third-party content. Each engagement includes a written scope document naming every source with its feasibility classification, the method used and what is excluded, suitable for your legal review.

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.

Feasibility classification: Labelling each source feasible, partial or not feasible before quoting. Quoting sources as equivalent when one has no lawful route produces a project that misses scope.
Partial source: A source that publishes the required field inconsistently or for part of its catalogue. Quoted with expected coverage stated, so the gap is a known parameter rather than a delivery surprise.
Unmatched report: The records that could not be matched to a client master, listed rather than dropped. A feed that silently discards them looks cleaner and hides the population worth investigating.

FAQ

Custom data extraction: frequently asked questions

What data, platform and strategy teams ask during evaluation.

Anything where no catalogue entry fits: obscure or regional sources, non-web formats like PDF and spreadsheet catalogues, unusual field combinations, or output that must arrive in your existing data model joined on your internal keys.

If a standard service covers your requirement, we will point you at it rather than scoping custom work. Custom costs more and takes longer, and using it where a product fits is waste.

After the feasibility assessment, never before. The assessment is free and takes about 24 hours: every source classified feasible, partial or not feasible, with reasons.

Ongoing feeds are quoted as a fixed monthly retainer. One-time projects are quoted as a project fee with a single delivery and QA report. We do not force a monthly retainer onto a question that only needs answering once.

We tell you before quoting and remove them from scope. Roughly one source in ten comes back not feasible, almost always for one of three reasons: no lawful route because it sits behind a login, the field does not exist publicly, or licensed content where a licence is the correct route.

Quoting an infeasible source produces a project that misses scope and a refund conversation. Declining one source in eleven keeps the other ten credible.

Yes, and this is one of the main reasons clients choose custom work. Output arrives in your data model with your field names, types, units and null conventions, joined on your internal keys.

The practical benefit is that nothing downstream changes when a source is added. The usual hidden cost of external data is not the vendor fee, it is the translation layer your team maintains.

Yes, with a confidence score you threshold. Matching uses identifiers where available then attribute and name similarity, and the method is recorded in the scope document.

Critically, unmatched records are reported rather than dropped. A feed that silently discards what it could not match looks cleaner and hides exactly the population you need to look at.

Yes, and rescue engagements are among the more common requests. We assess why the previous attempt failed, rebuild with documented method, and can run in parallel against your existing output so the comparison is direct rather than a claim.

Sometimes the assessment concludes that the source genuinely has no lawful route, in which case the honest outcome is that your team was right to stop. We will say that rather than sell a rebuild.

Both, priced differently. A market sizing, diligence exercise or one-off audit is a project: fixed scope, single delivery, QA report and documented method.

If the requirement turns out to be recurring, a project converts to a managed service on the same schema without rebuilding. We would rather run the project first than sell a retainer for something you needed once.

You get a handover pack on request: collection design documentation, source-by-source method, schema definition and known limits. Plus full historical data export with no exit fee.

Making that transition painful would be a poor way to operate. Custom work in particular tends to become strategic over time, and a client who internalises it well is a better reference than one held in place by lock-in.

Feasibility assessment within 24 hours. For engagements where sources are classified feasible, production delivery in 5 to 10 business days matches our standard timeline.

Genuinely awkward scopes — heavy OCR, many document formats, complex matching against a large internal master — run longer, and we give a realistic estimate in the scope document rather than quoting the standard window and missing it.

Get a free feasibility assessment on your own source list

Send us the sources that no vendor covers. Within 24 hours you get each one classified feasible, partial or not feasible, with reasons — before any quote.

Get a free feasibility assessment Book a 20-min scoping call

Free assessment, no card, no obligation. If a source has no lawful route, we tell you rather than quoting it.

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