Camping · Germany
+57%
ARPU
A preview of available pitches directly on the search page raised booking intent.
Read the success story →A free testing feature for apps inside Google's developer platform against a tool built for websites and the teams that run them. The difference is who builds the experiment, and where it can run.
15-day trial, no credit card, no setup fees.
Over 2,700 teams rely on Varify.
What sets Varify apart
Firebase publishes its prices, and so do we. The difference is what the price covers: free app experiments that a developer prepares in code, or a flat rate per domain for every website experiment.
Every plan, price and billing model in detail
See pricing →As of November 11, 2026, read from the vendor's pricing page, documentation, privacy page and data processing terms. Prices exclude tax; the vendor's own offer always takes precedence.
Every feature side by side. Firebase A/B Testing is built for app teams: a variation is a configuration value in the app's code, measured in Google Analytics — free, and strongest in native apps. Varify is built for the teams that change the website.
Flat price per domain available
Firebase A/B Testing itself costs nothing; Remote Config, which runs the experiments, is billed per fetch request above 100,000 per day per project on the Blaze plan. Varify bills a flat price per domain.
Multiple domains in one contract
One Firebase project holds several Android, iOS and web apps. At Varify via credits, 1 credit = 1 domain per month.
Pricing published
Both publish their prices.
Company and hosting in the EU
Google may process and store the data wherever Google or its agents maintain facilities; an EU-only location is not offered for Remote Config or A/B Testing. Varify is a German company hosting in Frankfurt.
A/B experiments
Firebase runs Remote Config experiments on Android, iOS and the web, and notification and in-app message experiments on Android and iOS. Its terms still list A/B Testing as Beta.
Split URL experiments
As a URL in a Remote Config parameter that your app redirects to.
Multivariate experiments
Not offered: up to 7 variations plus the original, each built by hand, and no mode that tests combinations of changes.
Multipage and funnel experiments
At Firebase through Remote Config parameters, which apply wherever your app reads them.
Server-side experiments
Not offered: A/B Testing covers Android, iOS and web apps, and Remote Config for servers is in Preview without an SLA. Varify via the Server-side Delivery API.
Experiments in native apps via SDK
Both offer SDKs for native apps. Firebase covers Android, iOS, Flutter, Unity and C++.
Feature-flag experiments
Firebase manages feature flags in Remote Config and rolls them out gradually. Varify offers feature flags too.
Visual editor without developers
Not offered. A variation is a set of Remote Config values that your app's code already reads, so every change a test can make has to be built into the app first.
Code editor for JavaScript and CSS
Not offered; variations are parameter values, not code.
AI assistant for variations
Not among the features Gemini in Firebase lists. At Varify via Prompt-Based Experimentation.
Ready-made widgets (pop-up, USP bar, exit intent)
Firebase In-App Messaging shows modals, banners and cards and can be tested, on Android and iOS only. At Varify as Campaign Booster, for websites.
Anti-flicker protection
Not built in; Firebase recommends that your app shows a loading screen until the experiment values are active.
Changes go live without a deploy
Within what the app already reads: Remote Config values change without an app update, also in real time.
AI writes variation code from a prompt
Not offered: variations are parameter values, not code. At Varify as Prompt-Based Experimentation.
AI audit with prioritized experiment ideas
Not among the features Gemini in Firebase lists. Varify's AI CRO Audit analyzes a page URL and returns recommendations and 10 test ideas.
Control from your own LLM (MCP)
Firebase's MCP server reads and updates Remote Config templates; it lists no experiment tools. At Varify as Varify MCP.
AI-assisted result interpretation
Gemini in Firebase summarizes messaging campaigns; experiments are not among its listed features.
Evaluation in your own analytics possible
Firebase calculates results from Google Analytics data, with an export to BigQuery; other analytics tools are not connected. Varify evaluates in GA4, BigQuery, PostHog, Matomo or Piwik PRO.
Automatic import of existing analytics goals
Google Analytics events your app has triggered appear as goal metrics.
BigQuery analysis
Through the Google Analytics export to BigQuery.
Bayesian statistics
Firebase uses frequentist inference.
Frequentist statistics selectable
Frequentist is Firebase's only model: one-tailed, at a significance level of 0.05, with data refreshed once a day.
Heatmaps and session recordings
Neither offers them. Varify does not replace Hotjar or Clarity.
Segmentation by device and source
Through the BigQuery export, not in the A/B Testing console.
Device and operating system
Remote Config conditions cover platform and app version.
Geo and language
Remote Config conditions cover countries, regions and languages.
UTM and query parameters
As a custom signal your code sets.
Referrer
As a custom signal your code sets.
Cookie, session and local storage
As a custom signal your code sets.
Login state and custom JavaScript
Through Google Analytics user properties or custom signals that your code sets.
URL match and patterns
As a custom signal your code sets; Remote Config has no URL condition of its own.
CSS selector as a condition
As a custom signal your code sets.
JavaScript condition
Through custom signals that your code computes and sets.
Exclude internal IPs
Not as an IP condition; your code can pass an internal-traffic flag as a custom signal.
Named contact person
Not offered: Firebase states it has no enterprise contracts or support.
Guaranteed response time
Not for A/B Testing: it is Beta, Google's paid support excludes Beta services, and Firebase's free email support has no published response time. Varify answers within 24 hours on working days.
Guided onboarding
Not offered: Firebase states it has no enterprise support; getting started is self-serve.
Works without cookies
Firebase measures through Google Analytics, whose web tags use first-party cookies by default. Varify stores the assignment in local storage.
No personal data stored
Remote Config selects values by Firebase installation ID, and results come from Google Analytics data. Varify stores no personal data beyond the IP address.
SSO via Azure or SAML
Through Google accounts, which can federate with Microsoft Entra ID via SAML.
Two-factor authentication
Through Google's 2-Step Verification.
DPA in German
Published in English only; the German page serves the English text.
Visual editor, targeting, personalization, BigQuery — the full feature set
Explore all features →As of November 11, 2026
Source: Claude Research, November 2026. All competitor information comes from publicly available sources: the vendor's website, pricing page, product documentation, privacy page and data processing terms. Last checked on November 11, 2026. Feature sets and plan limits change constantly, and the vendor's own statement always takes precedence. If a row is out of date, email us at info@varify.io and we will correct it.
Add the snippet, launch your first experiment, analyze what you learn. Describe the change in one sentence and Prompt-Based Experimentation builds the variation — no app release, no developer queue.
Add the snippet
Via your tag manager or straight into the head. We check the implementation and the anti-flicker setup.
Launch your first experiment
Varify picks up the goals from your existing analytics setup. Describe the variation as a prompt, then fine-tune it in the visual editor.
Analyze and learn
Analyze in Varify or in your own analytics — with the same goal definitions you use today.
App experiments in Firebase can keep running alongside. We help you launch.
AI CRO Audit, Prompt-Based Experimentation, Varify MCP
More on the AI features →Cookie-less, with no personal identifiers attached to experiment participants. German company hosting in Germany.
More on data protection →Three examples from customer programs. Each story shows how the number was reached.
Camping · Germany
+57%
ARPU
A preview of available pitches directly on the search page raised booking intent.
Read the success story →+33.7%
booking rate
A before-and-after of the windshield repair replaced the list of USPs — and the booking rate rose sharply.
Read the success story →
Software · Germany
+16%
sales
Real user numbers and star ratings in the hero area built trust where a product claim had not.
Read the success story →Yes, within the Remote Config quota of 100,000 fetch requests per day per project; above it, the Blaze plan bills per request. What it needs instead of money is development work: every value a test can change has to be built into the app as a Remote Config parameter and shipped in a release before the experiment can start.
You pay a flat price per domain, regardless of visitor numbers and with unlimited team members. The price is published openly on the pricing page.
For mobile app teams that already build on Firebase and Google Analytics and want to test configuration values, notifications and in-app messages in their Android and iOS apps at no cost. Varify is built for teams that change the website without waiting for a developer or a release, and evaluate in the analytics they already run.
Yes. The two tools do not get in each other's way: app experiments can stay in Firebase while website experiments run in Varify, and both can report into the same Google Analytics property.
Three features work together. The AI CRO Audit analyzes a page URL and returns concrete recommendations and 10 test ideas. Prompt-Based Experimentation then builds the variation and writes its code from a description in plain language, and you refine it as often as you like. Varify MCP lets you drive Varify from your own AI assistant.
They keep running until you stop them, and Firebase ends them automatically after 90 days. Varify rebuilds the website experiments including their targeting, and the numbers are cross-checked in your analytics. Snippet and goals are usually done in a morning.
No. Varify stores the variation assignment in local storage, without personal data. Whether that needs consent on your site is a decision for you and your data protection officer.
Firebase and Google Analytics are trademarks of Google LLC. Varify.io is not affiliated with Google LLC. All competitor information comes from publicly available sources as of the date stated and may change. The vendor's own offer always takes precedence.