Reliable datasets, scoped to your brief.

We build the pipeline, run it on your schedule and repair it when a source changes. Your team receives verified rows where it already works — Sheets, Snowflake, S3, API or MCP.

Three things define your dataset.

We scope all three with you before the first run.

  1. 01

    Coverage

    The sources, markets and mobile apps it draws from.

  2. 02

    Depth

    The fields, matching rules and checks that make every row usable.

  3. 03

    Frequency

    How often it refreshes and where each run delivers.

One pipeline, end to end.

00 / BRIEF

Monitor competitor prices, promotions and availability across retailer websites and mobile apps. Match products. Deliver a daily dataset with changes since the last run.

01 / SOURCES Products & sources Your market, defined 02 / COLLECT IN PARALLEL Read public websites Prices, stock, promotions Read mobile apps Offers and availability 03 / PROCESS & VERIFY Match & compare Same products, new snapshot 04 / PUBLISH Email report Market prices A dataset that stays current Snowflake XLSX
  1. 01 / Sources

    Products & sourcesYour market, defined

  2. 02 / Collect in parallel

    Read public websitesPrices, stock, promotions

    Read mobile appsOffers and availability

  3. 03 / Process & verify

    Match & compareSame products, new snapshot

  4. 04 / Publish

    Market pricesA dataset that stays current

    Email reportSnowflakeXLSX

Websites and apps read in parallel. Products matched; every published row passes type, null and duplicate checks. Delivered to Snowflake, XLSX or an email report.

Automatic repair

Repaired automatically.

A failed run is detected, the pipeline is repaired and the schedule resumes. Each repair ships as a new version; you see what changed and when.

Automatic pipeline repair Oat drink1 litre £2.20 ! Layout changes Jason repairs and verifies the pipeline Price field moved RepairedSample verified ✓ product_prices pricecurrency 2.20GBP 2.35GBP ✓ Checked and ready Data flows again v13 → v13.1 · detected 04:10 · verified 04:22

Escalation

A human when repair stops.

Some failures need investigation. Your agreement names the escalation contact, coverage hours and response times. An unverified fix is not a completed recovery.

Escalation to a named contact Repair attempts 1 · Sample failed 2 · Sample failed 3 · Sample failed Run held · not marked done ! Repair stops Named contact from your agreement within coverage hours Investigating… Fix verified ✓Type, null and duplicate checks Schedule Next runpausedWaiting for a verified fix Next run06:00 tomorrowRecovery complete✓ Verified before resuming Schedule resumes

Continuity

Code you can leave with.

Enterprise agreements can include a perpetual licence to your pipelines’ Python source: a snapshot as deployed when you leave, with dependencies and handover notes, to run and modify for your own use. The licence outlasts the contract. Repairs and updates stop at exit, because keeping pipelines working is the service.

Pipeline source handed over under your agreement Your pipeline on Jsonify v24 · live Runs every day Export package pipeline.py requirements.txt HANDOVER.md Licence set in your agreement Packaged under your agreement Your repository Nothing here yet pipeline.pyrequirements.txtHANDOVER.md✓ Runs on your infrastructure Licensed to keep

Security review

Security and data handling.

Everything your security, legal and procurement teams will ask about, with the document each answer comes from.

  • SOC 2 Type IIIndependently audited
  • GDPRDPA & SCCs available
  • EU AI ActTransparency built in
Independent assuranceSOC 2 Type II audited; report scope, period and findings available for your security review.
Review the report’s scope, period and findings with your security team. Request the SOC 2 report →
Collection scopeSources, fields, intended use and frequency agreed in writing; you confirm authority to access and use the data.
Sources, fields, intended use and frequency are agreed in writing. You confirm your authority to access and use the requested data. Access terms →
Permitted accessRobots.txt respected on websites; no gated content without your credentials, used within account permissions.
Pipelines respect robots.txt on websites and do not access content behind logins or paywalls except where your agreement provides for it. Authenticated collection uses your credentials, within your account’s permissions. Access terms →
Personal dataPseudonymisation and filtering of personal-data fields configured per agreement; your data is not reused, sold or exploited for Jsonify’s purposes.
Pseudonymisation and filtering of personal-data fields are configured per agreement; where configured, they run on every row. Data processed on your behalf is not reused, sold or exploited for Jsonify’s purposes. Privacy policy →
Hosting and transfersHosting, subprocessors, retention, deletion and international transfers are covered in your review.
Hosting, subprocessors, retention, deletion and international transfers are covered in the review. Subprocessors → Privacy policy →

Procurement

Four steps to a first production run.

Bring your security questionnaire. We work through it with your security, legal and procurement teams.

Start a security review
  1. 01

    Security evidence

    SOC 2 report and the controls relevant to your service.

  2. 02

    Data and access scope

    Sources, access method, fields, intended use, delivery destinations.

  3. 03

    Processing terms

    Roles, contractual terms, data handling, transfers.

  4. 04

    Operating agreement

    Validation criteria, support, and how changes to the pipeline are handled.

Questions before you get started.

Why not use a managed collection service?
Jsonify offers managed collection alongside self-serve. We agree the sources, fields, checks, schedule and delivery with you, then build, run and repair the pipeline. Your agreement sets out support and how changes are handled. See how we scope your dataset →
Can we keep our existing proxy or browser provider?
An infrastructure contract and a maintained dataset solve different parts of the job. Keep the infrastructure that works for your team, and discuss how it fits into the collection and delivery scope before rollout. Compare infrastructure and dataset responsibilities →
What happens to our data if we stop using Jsonify?
You can export your datasets. Enterprise agreements can also include a perpetual licence to a snapshot of your pipeline code as deployed at exit, for your own internal use, with dependencies and handover notes. Repairs and updates stop when the agreement ends. Review continuity and vendor risk →
How does Jsonify handle security and compliance?
Jsonify is SOC 2 Type II audited, offers a GDPR Data Processing Addendum with Standard Contractual Clauses, and discloses Jason as an AI system in line with the EU AI Act. Pseudonymisation and filtering of personal-data fields can be configured per agreement. For a managed deployment, we agree collection scope, processing terms and delivery requirements, and provide the SOC 2 report for your security review. Explore enterprise security and compliance →

Two ways to collect your market.

RadarNow available

Monitor what’s published.

Collect prices, menus, listings and availability from websites and apps. Follow what changes across sources and markets.

BenchmarkEnterprise

Compare what’s offered.

Run defined scenarios through forms to collect comparable insurance quotes, rental rates and personalised offers.

Enterprise

Let’s scope your dataset.

Share your sources, delivery needs and security requirements.

Connect your data assistant

Build datasets and work with your data in Claude, ChatGPT, Copilot or another assistant.

Connect in Claude

  1. Open the Jsonify connector: Open on Claude.ai ↗
  2. Select Connect and sign in to Jsonify.
  3. In a chat, enable Jsonify under + → Connectors and describe the data you need.
Advanced

Team or Enterprise account? Your workspace owner enables Jsonify once for everyone from Admin settings → Connectors. Jsonify is a community connector in the directory — some admins allow those by default, some review them first.

No directory? Add it by hand: Settings → Connectors → + Add custom connector, name it Jsonify, paste the server URL below, then Connect.

Server URLhttps://factory.jsonify.com/mcp

Official Claude custom-connector guide ↗

Then say: “build me a dataset of competitor product prices and availability, refreshed daily”. Full instructions per client on /connect.