We're Building the European Vercel of Data

Vercel made front-end deployment painless: you push code, it is live. Feels like magic, feels like you do not have to take care of a single deployment anymore. What we want to do is the same thing for data. We want you to connect a source, sync, generate query with your agents or your little hands, and share the results with yourself or your colleagues. That's it.

All that in a single product, hosted in EU servers, priced for middle sized organisations that do not need a consulting team on top and know that their main KPI is not the number of dbt models they have in prod or the nested complexity of their dags.

Now some organisations do indeed need that. We are not the tool for them yet; we want to fill in the little gap between "I need it" and "I don't want to get into a sales call".

Vendors. Nobody likes them.

If we are honest, we all have way too many damn vendors. Some SDR somewhere convinced someone at your org to buy it, then you had 5 team meetings about it, and now you have to use it for reasons obscura. Fivetran or similar to ingest. Snowflake or similar to store. dbt or alternatives to transform. Looker or whatever to visualize. Much vendors, quite a few bills, endless login pages, countless Slack channels where you beg for support.

Something like a 50-person scaleup paying $50k/year for a data stack is likely not an oddity anymore. For that company, a competent senior you're not hiring. And the little cherry on top of that disappointment cake? Half of that effort and money is spent on gluing the other halves together.

The modern data stack was a great idea that turned into an objectively amazing business model. For the vendors. For everyone else, it's twenty meetings and a deprecating will to live.

We Have One Product. Does Most of What You Want.

rawquery does most of what you want from something that scales; ingestion, storage, compute, and serving. The way Vercel is one product from git push to production URL. You need more? We are gladly leaving you with your demons and architecture reviews. You need data somewhere to be queryable and get stuff out of it? Just use us. We are cheap, we are easy. We will get better at everything else eventually.

Right now you can connect; Stripe, Postgres (or simply use live connectors!!), HubSpot, Shopify, MySQL, Salesforce and even Google Sheets, or any API you can describe in a JSON file. Your data will land in Apache Iceberg. You will query with DuckDB. You can make Transforms with SQL. Share charts with a link to the data as REST endpoints. Connect your BI tool via Postgres wire protocol if you really feel fancy.

rq connections create my-stripe --type stripe -p api_key=sk_live_xxx
rq connections sync my-stripe
rq query "select status, count(*) from my_stripe.subscriptions group by 1"

rq connections create my-stripe --type stripe -p api_key=sk_live_xxx
rq connections sync my-stripe
rq query "select status, count(*) from my_stripe.subscriptions group by 1"

Here are 3 commands. Where we went from "I need my Stripe data somewhere queryable" to "here are my subscription metrics" without provisioning something somewhere to do it, without a 40-page setup guide, without a "let's schedule a call to discuss your needs."

If your data needs more care than that. Well that is indeed very nice. We are not there yet. Come back in six months once we integrated DataFusion engine and a few fancier transformation tools. Or not.

We are Europeans.

We like regulation. Our job is to provide a data service not political opinions. But if you really need to know; I like having a non detachable cap on top of my water bottles. I also enjoy the idea of not building a company that will be artificially x40 its EBITDA and that we will have to have the burnrate of Montenegro to justify that valuation.

rawquery runs in European servers (Frankfurt, Germany). Compute, storage, API: all of it sits in an EU datacenter. No "data residency option" buried in legal. We are European, so the product is European. Voila.

The European angle is not a marketing angle with branding on top. It is a fact. We don't mind American technologies; we love them. But we do not live in San Diego, our goals and ambitions are not the same; we want to build a sustainable long term business, not a moon shot from which we can add "ex rawquery" on linkedin in a year.

Pricing that Respects Your Intelligence.

The data industry invented forward deployed engineers, consultants, SLAs, and consumption based pricing, which sounds fair until someone runs a bad query on Friday and you get a $12k surprise on Monday. Then they invented "contact us for pricing," which is code for "we'll charge you as much as we think you can pay while not feeling too bad about it."

We have a free tier that actually works, you could actually run on it. Not a "free trial" that expires before you finish. Paid plans are flat-rate and visible on the website. We are planning some overage costs. Because some people want more, ok, but you do not need it, and you will need to activate it.

Your data sits in Iceberg tables. An open format. If you leave, you take everything. We don't do lock-in. If the product isn't good enough to keep you, we're sorry. We hope you enjoy the enablement workshop with whoever else you pick.

The DX is not a moat. But it's damn good?

Vercel won because deploying felt like nothing. Data Platforms should feel the same way. You should not need to understand Iceberg table maintenance or S3 bucket policies to know how many customers churned last month.

rawquery ships a CLI, a REST API, a Postgres wire protocol, and a web UI. The CLI came first. The UI was built on top of it. If you can type SQL, you can use rawquery. If you can't type SQL, your agent surely can, and worst case you can still push a CSV and get a chart in two commands.

# Push a CSV, query it, done
rq push sales.csv --table sales
rq query "select region, sum(revenue) from sales group by 1 order by 2 desc"

# Or just connect with psql. It speaks Postgres wire protocol.
PGPASSWORD=rq_your_api_key psql -h rawquery.dev -p 5433 -U any -d your-workspace

# Push a CSV, query it, done
rq push sales.csv --table sales
rq query "select region, sum(revenue) from sales group by 1 order by 2 desc"

# Or just connect with psql. It speaks Postgres wire protocol.
PGPASSWORD=rq_your_api_key psql -h rawquery.dev -p 5433 -U any -d your-workspace

The bets we make

We're betting that most teams don't need a warehouse that scales to petabytes. They need one that works at gigabytes without a finance meeting. DuckDB on Iceberg handles 90% of analytical workloads, and the other 10% can afford anything else anyways and built it.

We're betting that "connect and query" beats "integrate, configure, model, orchestrate, materialize, serve, maintain, pray, cry." We're also making the odd bet that building for SMEs and from Europe is a market.

Try it.