The security rules let anyone read the database
Starter rules shipped unchanged, or rules relaxed during development and never put back. Anyone who opens the browser console can list your users or orders.
Rules that match the actual data model
Working with Firebase?
A lot of Firebase projects reach us the week after someone noticed the bill, or the week after a tester found they could read every document in the database. We lock the rules down to what the app actually needs and cut the read volume to match.
What it is
Firebase is Google's backend-as-a-service: a bundle of hosted pieces that a mobile or web app can plug into instead of writing its own server. It started in 2011 as a company called Envolve, became Firebase soon after, and was acquired by Google in 2014.
The bundle covers Firestore and the older Realtime Database for storing data, Authentication for sign-in, Hosting for static files, Cloud Functions for server-side code, Cloud Messaging for push, and Analytics and Crashlytics for measurement. It is popular with mobile teams and with web apps assembled quickly from templates and AI tools.
The honest version
No technology is all upside. This is where Firebase earns its place and where it tends to cause trouble.
Sound familiar?
The Firebase problems that bring people to us, and what you have once each one is fixed.
Starter rules shipped unchanged, or rules relaxed during development and never put back. Anyone who opens the browser console can list your users or orders.
Rules that match the actual data model
A listener in a loop, a function triggering itself, or a client-side query reading the whole collection on every page view. Usually visible in the usage graph within one morning.
Reads cut, alerts before the next spike
The old rules were covering for a client that was asking for more than it should. Fixing the rules exposes the query that needs to change too.
Rules and queries aligned
A Cloud Function that has been idle, often with a bulky dependency tree or a slow initialisation step at the top of the file.
Warm paths for the user-facing calls
A new screen subscribing to a collection without pagination, or a background job that forgot to limit its query.
Paginated, with a ceiling on read volume
Firestore does not hand you a clean dump of a running project. The move needs a staged copy, a cutover window and a way to replay writes that happened in between.
Migrated onto Postgres or Supabase
Bots or scraped SDK keys hitting Firestore directly from scripts that never opened the real app.
App Check wired up and enforced
The console does not back up your data on its own by default, and a bad client write can delete thousands of documents before anyone notices.
Scheduled exports with tested restores
What we do
Call us for one open rule, or hand us the whole project.
How it works
Tell us what's going wrong. We look at it, then tell you plainly what's wrong and what it will cost.
We fix the cause, not the symptom. If something is down, getting it back up comes first.
Backups, monitoring, updates and access, so the same problem can't quietly come back.
You get a plain-language write-up. Or we stay on for hosting, upkeep and new work.
Often found together
We work on the whole stack, so the problem can be wherever it actually is.
Open the Firebase console, go to Firestore, and look at the Rules tab. If you see 'allow read, write: if true' or rules that only check whether the user is signed in, then any visitor or any registered account can read every document. We audit the rules against the collections and the app's real queries and rewrite them to match.
Almost always it is a sudden rise in Firestore document reads, Cloud Function invocations or outbound bandwidth. The usage graph in the console shows the day it started; from there we trace it to the listener, loop or background job that caused it and cap the damage.
Parts of it, yes. Auth and file storage port over reasonably cleanly. Firestore queries and Cloud Function triggers need rework because the data model is different. We scope that honestly before you commit to a move.
Start from deny-all and open each collection only for the specific reads and writes the app actually needs, checking the authenticated user against document fields. Test the rules in the Firebase emulator against the real queries before deploying, and keep them in version control alongside the code.
A function that has been idle has to boot its runtime, load its dependencies and run any top-level setup before it serves the first request. Trimming the dependency tree, moving one-off work out of the global scope and keeping critical paths warm bring that first call back under a second.
Estimates are free. Describe what the project is doing and you get a price before any work starts.
Free estimate
Tell us what you're running and what's going wrong. A person reads it and comes back with an estimate.
Your message is with a real person, and we'll get back to you soon.
Rather talk now? Call us: