Case study · Personal data removal · USA and worldwide
REMOVEThree people built a privacy product with 2M+ removals from 500+ data brokers, 95%+ with no human involved
REMOVE finds people's names, addresses and phone numbers on data-broker sites, gets them taken down and keeps watching in case they come back. I was hired as fractional CTO to build the product from zero. I designed its architecture, built a team of two engineers and led the work on hundreds of custom parsers and removal flows, one of the hardest problems in privacy. Three people now cover more than 500 brokers at a 99% success rate, work that is usually done by hand by far larger teams. More than 95% of removals run with no human involved, a new customer's first removal is done within 24 hours, and each removal costs the company cents.
- Client
- REMOVE
- Product
- Automatic removal of personal data from data brokers
- Market
- USA and worldwide
- My role
- Fractional CTO, hired to build the product
- Team
- 2 engineers and me
- Scale
- 500+ brokers, 2M+ removals, 50,000+ people protected
- Cost of one removal
- Cents, with no human involved in 95%+ of removals
- Website
- remove.dev
In numbers
- of removals run with no human involved, so a team of three is enough
- 95%+
- removals completed, each one costing the company cents
- 2M+
- to a new customer's first removal, with nobody filing it by hand
- 24 h
- removal success rate, even against brokers that resist
- 99%
- data brokers, each with its own parser, all built by three people
- 500+
- people protected by a product three of us built
- 50K+
The product
Scan, remove, monitor: the whole process runs automatically with no manual step, and the customer sees every broker and every removal in one dashboard. Images from remove.dev.
The starting point
Data brokers collect people's names, home addresses, phone numbers and relatives, and they sell all of it to anyone, legally. Getting that data taken down is one of the hardest problems in privacy. Every broker has its own site, its own search, its own removal process and its own ways of resisting, and the listings come back after they are removed.
REMOVE set out to do it automatically, for hundreds of brokers at once, where most of this work is still done by hand. That meant building a product from zero and a team to build it, and the company hired me to lead both.
What we did
Hired to build the product
I came in as fractional CTO to create the product from scratch. I designed its architecture and every system in it, and the company got a founding CTO's work without a full-time executive hire.
A team of three, built for output
I built the technical team: two engineers and me. I decided how we work, what we build first and the standard every broker integration has to meet, so each new broker is repeatable work instead of a new project.
Hundreds of custom parsers
Every broker publishes people's data in its own way. We wrote hundreds of custom parsers that find a person's listings on each of them: names, addresses, phone numbers and emails. Three people built coverage that usually takes a far larger team.
Removal that works 99% of the time, for cents
Getting the data removed is the hard part. For each broker our systems send the request it will accept: through its API, through browser automation, or as a legal demand under CCPA and GDPR when the broker resists. That is how the product reaches a 99% success rate with no human involved in more than 95% of removals, and why each removal costs the company cents.
The first removal within 24 hours
A new customer does not wait in a queue of manual requests. The system scans the brokers, files the removals on its own and gets the first one done within 24 hours, so the customer sees a result on their first day.
Monitoring 24/7
Scheduled re-scans catch listings that reappear and new exposures, and the system removes them again automatically, 24/7, with nobody on the team keeping watch.
Privacy first
The product holds exactly the data it exists to protect. It keeps that data encrypted with AES-256, never shares it and uses it only to get it removed, and that is why customers trust it with their personal details.
A dashboard for every customer
Each customer sees in real time every broker that had their data and where each removal stands, so the product proves its value every time they open it.
The result
Three people built a product from zero that has completed more than 2 million removals from more than 500 data brokers, with no human involved in more than 95% of them. A new customer's first removal is done within 24 hours, each removal costs the company cents, and at a 99% success rate the product protects more than 50,000 people around the clock.
Full case · PDF
REMOVE
Three people built a privacy product with 2M+ removals from 500+ data brokers, 95%+ with no human involved
- 01The architecture of the product
- 02How a broker integration is built: parser, removal flow, verification
- 03APIs, browser automation and legal demands: choosing the channel
- 04Monitoring and automatic re-removal, with nobody on watch
- 05Protecting the data we exist to remove
- 06How a team of three covers 500+ brokers
The full REMOVE case
The public story stops here. The PDF has the rest: the architecture before and after, the migration plan, how the team works with AI agents, what it all costs to run and where the savings came from.
More case studies
All case studies
Want the next case study to be about your company?
In a 30-minute call we pick the first task to hand to AI and estimate what it will save you.