Everyone raised prices. We didn't.
Vendors raised prices in 2025, raised them again in 2026, and will raise them in 2027. We kept ours where they were and doubled the product instead.
Data integration vendors raised prices in 2025. They raised them again in 2026. They will raise them in 2027.
What arrived alongside those increases rarely justified them. A 30% rise buys a redesigned settings screen and a refreshed logo.
We went the other way. Prices stayed where they were, came down in some cases, and the product roughly doubled.
What doubled means
Etlworks in 2026 does about twice what Etlworks in 2025 did. Against two months ago it does about 1.5x. The list is specific:
- Native Databricks support, with three MERGE strategies and Unity Catalog naming
- AS2 send and receive, plus a schema-driven editor for building X12 messages
- An MCP server, so Claude, Cursor, and Windsurf drive the platform directly
- Resumable CDC snapshots and chunked parallel snapshots
- Security groups with fine-grained permissions enforced across UI, REST, CLI, AI, and MCP
- A Windows CLI with full backup, restore, and suspend/resume
- An AI usage and cost dashboard
- Context-aware autocomplete in the code editors
None of it arrived as a paid add-on. None of it moved to a higher tier. It landed in the product you already pay for.
Why we could afford it
We are roughly 10x more efficient than we were in 2024. Not 10 percent. 10x.
AI did that. It writes most of our code, most of our documentation, and most of our internal automation, and it answers a growing share of support requests before a person sees them. Our cost to build and run this product fell off a cliff.
When that happens you have two options. Keep the gain, or hand it to customers. We are a small company competing against firms with more salespeople than we have employees, and the second option is why we keep winning those deals.
Predictable and upfront
Open the calculator on the pricing page and get your number. Add what you need, remove what you do not, watch it change. No quote form, no discovery call to learn the price, no wall with "contact sales" behind it.
That is not a temporary state while we grow. Pricing that requires a conversation is pricing designed to find your budget.
Fair, not clever
Enterprise is unlimited. Unlimited records, real-time CDC, your choice of cloud and region, SSO, white-label, multi-tenant, and five instance sizes you can move between at any time with no re-signing.
Then compare our Starter and Business tiers against a competitor's enterprise tier. Unlimited users, unlimited connections, and unlimited flows come on every plan we sell. At a lot of vendors those three lines are the entire reason to upgrade.
The AI costs us money
Every plan includes an AI allowance that works out of the box. No key, no wallet, no configuration. Q&A is unlimited on every plan, and the agentic cap scales by tier.
When you outgrow the allowance there are two paths. Bring your own OpenAI key and pay your provider directly, with zero markup from us. Or fund an Etlworks wallet if you would rather not manage another vendor relationship. Switch between them whenever you like, since there is no contract around AI usage.
Those capabilities cost us more than they bring in. That is a deliberate decision about where this product should be in three years, not a promotional rate with an expiry date.
A test for your current vendor
Next time a price increase lands, ask what shipped since the last one. Ask what moved into a higher tier to make room for it. Then open our calculator and compare the number.