Use cases · Data teams
See what iPhone users see.
Many sites serve mobile Safari a different site than desktop browsers: different layouts, different flows, different prices. For price monitoring and market research, you want the version iPhone users get.
The problem
The mobile web is a different web.
Point a desktop collector at a mobile-first site and you get the desktop version: different prices, a different layout, sometimes a full-screen "please use our app" page instead of the content. Changing only the browser's name so it claims to be a phone is not enough either: the layout, fonts and rendering still come from a desktop engine, so what you record is not quite what iPhone users see.
Driftstack starts from Safari's engine instead. Every session runs a build of Apple's own WebKit — the engine family behind iPhone Safari — set up to present an iPhone model and its screen. So your collector records the mobile layout, the mobile flow and the mobile prices.
Why the data is right
The page iPhone users get.
Websites choose which version of a page to serve from what the browser reports: its engine, its screen, its fonts. Driftstack's answers are checked against real iPhones, so the page you collect is the page an iPhone visitor gets.
The mobile page, not a fallback
What a website measures to choose a page — engine, screen, fonts, graphics — comes from Safari's own engine and is checked against real iPhones
Consistent from run to run
The fixed answers — screen, hardware, audio settings — stay fixed, and the ones Safari deliberately varies are varied the way a real iPhone varies them, so repeat collections compare like with like
The country you need
Each profile goes out through your own proxy or VPN (SOCKS5, or OpenVPN and WireGuard on paid plans), so you see that location's prices and offers; the session's time zone follows that country
Want the detail? The comparison page goes signal by signal: what a website measures, and what each side returns.
Which plan
Every plan includes the API and the SDKs.
Collectors are written in code, and every plan runs them. Billing turns on one number: how many profiles run at the same time — think browser tabs. Paid plans have no hour limit (fair use). Run as many sessions as you like, up to your plan's number at the same time.
Your own code drives the sessions. We publish an SDK — a ready-made code library — for TypeScript, Python and Go. Any program that can make an HTTPS request can drive a session too. A session takes a short list of actions: navigate, interact, capture, extract.
The AI agent is on every paid plan. Webhooks are messages Driftstack sends to your own system. They tell your job when a session hits a challenge or could not save its profile; an agent session (the kind that runs the AI agent or a list of steps) sends no completed or failed webhook, so your job reads the session's status to see that it ended.
Switching from a Chrome-based tool? The FAQ below covers what changes.
Collect responsibly. This page is about market research and price monitoring. Scraping that gets around logins (authentication) or past a site's reasonable rate limits is not allowed. The Acceptable Use Policy draws that line in plain words.
Questions data teams ask
Before you switch — asked and answered.
Migrating from another vendor
Can I run a side-by-side comparison before committing?
Acceptable use
Is X allowed? (sneaker bots / scraping / ad fraud / etc.)
What happens if Driftstack as a business goes away?
Everything else is on the full FAQ.
Ready when you are
Collect the web your users actually see.
Start a free trial next to your current vendor, and compare what each one collects on your own pages. No card needed.