Question 1

What exactly does Justmetrically do?

Accepted Answer

We design, build, and maintain custom web data extraction pipelines for enterprise teams. That means scoping your data source and delivery requirements, building collection infrastructure that survives real-world scale, cleaning and structuring the output using AI-powered normalization, and delivering it into the systems your team already uses — on a schedule that fits your workflow.

Question 2

How long does it take to launch a new project?

Accepted Answer

Most projects go live within 5–10 business days from brief to first delivery. A Validation Sprint is typically delivered in 3–5 business days. More complex recurring pipelines with custom schemas and integrations take 2–3 weeks.

Question 3

Do I need technical knowledge to work with you?

Accepted Answer

No. You describe your requirements in plain language and we handle all the engineering. You'll receive clean, structured data without writing a single line of code.

Question 4

What information do I need to provide to get started?

Accepted Answer

The target website(s), the specific data fields you need, your preferred delivery format, your desired refresh cadence, and any business context that helps us prioritize edge cases.

Question 5

Is there a minimum engagement size?

Accepted Answer

Our smallest engagement, the Validation Sprint, starts at $100 and is designed to test viability on a new source before committing to a larger pipeline contract.

Question 6

How is the service priced?

Accepted Answer

We offer three tiers: a Validation Sprint ($100, one-time) for testing a source; Managed Pipelines ($500/month) for recurring delivery; and Custom Data Products (custom quote) for dashboards, warehouses, and enterprise programs.

Question 7

What is a Validation Sprint?

Accepted Answer

A scoped, one-time extraction that proves whether a given source is viable before you commit to a recurring contract. Deliverables include a sample dataset, schema overview, and a complexity assessment.

Question 8

Can I do a one-off project without committing to a subscription?

Accepted Answer

Yes. The Validation Sprint is designed explicitly for one-time engagements. If you need recurring delivery later, it converts naturally into a Managed Pipeline contract.

Question 9

What happens if my requirements change mid-project?

Accepted Answer

Scope changes are handled as separate line items with a revised estimate before any additional work proceeds. There are no surprise charges.

Question 10

Do you offer refunds?

Accepted Answer

If we determine a source is technically unviable after a Validation Sprint begins, we'll discuss a partial or full refund case by case. Managed Pipeline contracts allow cancellation with 30 days' written notice.

Question 11

What types of websites can you extract data from?

Accepted Answer

We can extract from most publicly accessible websites, including e-commerce platforms, real estate portals, job boards, directories, news sites, financial data sources, travel platforms, and more.

Question 12

Can you handle websites with anti-scraping measures like CAPTCHAs or bot detection?

Accepted Answer

Yes. Our infrastructure includes rotating proxy pools, browser automation, CAPTCHA-solving pipelines, and session management. We engineer around bot detection as standard.

Question 13

Can you scrape JavaScript-rendered pages like React or Angular SPAs?

Accepted Answer

Yes. We use headless browser automation (Playwright-based) for dynamic content that standard HTTP fetching cannot reach.

Question 14

How frequently can you refresh the data?

Accepted Answer

From real-time webhook triggers to hourly, daily, weekly, or custom scheduled cadences. Refresh frequency is factored into pricing.

Question 15

What output formats do you support?

Accepted Answer

JSON, CSV, XLSX, JSONL, and Parquet natively. We also support delivery into BigQuery, Snowflake, Redshift, Google Sheets, REST API endpoints, webhook receivers, and Slack.

Question 16

What happens if a website changes its structure and breaks the pipeline?

Accepted Answer

We monitor for schema drift and broken selectors as part of ongoing pipeline maintenance. When a structural change breaks extraction, we fix it proactively and notify you. Included in all Managed Pipeline contracts.

Question 17

How accurate is the extracted data?

Accepted Answer

We typically deliver above 98% field completeness on well-structured sources after our normalization layer. Every pipeline includes QA logic to flag anomalies before delivery.

Question 18

How do you handle missing or inconsistent data?

Accepted Answer

We apply cleaning rules, fallback selectors, and AI-powered normalization. Unresolvable gaps are flagged with a null value rather than silently dropped.

Question 19

Can you enrich the extracted data with AI?

Accepted Answer

Yes. We offer LLM-powered enrichment — classifying products, summarizing descriptions, extracting structured fields from unstructured text, and more. Available on Managed Pipeline and Custom Data Product tiers.

Question 20

Do you deduplicate the data before delivery?

Accepted Answer

Yes. Deduplication logic is included in all pipeline QA layers, configurable by key fields.

Question 21

Is web scraping legal?

Accepted Answer

Scraping publicly available data is broadly legal in most jurisdictions, supported by precedents including hiQ v. LinkedIn. Legality depends on the specific website, use, and data type. Clients are responsible for ensuring their use case is lawful in their jurisdiction.

Question 22

Do you sign NDAs?

Accepted Answer

Yes. We routinely sign NDAs and are comfortable with standard enterprise confidentiality agreements. Mutual confidentiality is our default operating posture.

Question 23

Are you GDPR compliant?

Accepted Answer

Yes. Our core services target publicly available, non-personal business data. We do not extract or store personal data as part of standard pipelines. We will enter into a Data Processing Agreement (DPA) for enterprise clients who require one.

Question 24

Do you scrape personal data?

Accepted Answer

Generally no. Our standard services target business data. If a specific use case requires working with personal data, we assess legality and put appropriate safeguards in place before accepting the project.

Question 25

What is your data retention policy for delivered datasets?

Accepted Answer

Delivered datasets are stored on our systems for 30 days post-delivery, then purged. Longer retention is available by arrangement. We do not retain copies of client data beyond what is needed to deliver and support the engagement.

Frequently asked questions.

Talk to a human

See pricing

Read our blog