CRM Without Vendor Lock-in — What That Means in Practice

DD Pim van den Boomen • 13 June 2026 •5 min

Pillar 03 // Where to start

crm.md

$ export --everything --and-then-what

CRM Without Vendor Lock-in — what that means in practice

Every CRM has an export button. That's exactly why the button tells you nothing. You get your contacts and you lose your pipeline, your fields, your automations and your history — everything that made the CRM your CRM. Below is what lock-in really is, which test exposes it, and why we're moving to an open-source CRM. Without a cost argument underneath it.

DD DataDrift Digital • June 13, 2026 • 7 min

"Vendor lock-in" is jargon for something simple: the cost of leaving is so high that you stay, even if you'd rather not.

That cost is rarely written into a penalty clause. It's in the work you have to redo.

01 / what lock-in really isWhy isn't an export button an answer?

Because it exports the easiest part.

Ask a vendor whether you can take your data with you and the answer is always yes. That's also true: you get a file with contacts, and that really is yours.

But think about what you've built up over a year:

COMESContacts and companies. Names, addresses, phone numbers. Those come along, always.

PARTLYCustom fields. Often as a column in the file, but without the meaning: which values are allowed, what happens when one changes.

NOYour pipeline. The stages, the rules for when something moves along, the reporting on top of it.

NOYour automations. Every "if this, then that" you've set up, rebuilt from scratch.

NOThe history. Who called when, what was discussed, which email was sent. Usually unusable or incomplete.

You export your contacts. You leave your way of working behind.

That difference is the lock-in. Not the contract — the rebuild work.

02 / the testWhat do you take with you if you leave tomorrow?

Four questions, answerable today, before you need them.

Can you download an export without calling your vendor? If someone has to submit a request, the export is a favour, not a right.
Does the history come with it — conversations, notes, emails? Test it. Export one contact and look at what's actually in the file.
Is there documentation anywhere describing how your automations work, outside the system itself? If not, your way of working is only readable as long as you keep paying.
Whose name is the account under? If you work through an agency, you sometimes have access to an environment that belongs to them. That's a different story than ownership.

That last one is the sharpest, and it's the reason we're moving ourselves. Our CRM runs in an environment that works fine technically, but the setup is that we manage it and the client sits inside it. That is exactly what a client does not own.

03 / what we choseWhy Twenty?

For three reasons, one of them technical and two not.

It's open source. If the vendor disappears, the software still exists. That's a different kind of security than a service contract: you're not dependent on whether a company keeps existing.

It talks directly to language models. Twenty is the only open-source CRM we know of with a built-in connection for AI assistants. For us that's not a side note — it's the difference between a CRM an agent can pull data from and a CRM we'd first have to build a middle layer for.

It separates clients at the database level. Every workspace gets its own schema in the database. That's a stronger separation than a column saying "which client does this row belong to," which is what you get in many systems: one bug in a filter and client A sees client B.

04 / what you give up for itWhat is the price of open source?

More than the project's website tells you. Four things.

YOUNGVersion 0.2. That's an early version. Expect changes you'll need to track, and features that don't exist yet.

PARTIALIt only replaces the CRM part. No calendar, no payments, no landing pages. Those have to come from somewhere else.

RISKRecord-level permissions are currently in the free version. With open-source projects that have a commercial model, that kind of thing can move to a paid tier.

YOUYou run it. Updates, backups, outages. That's the fixed price of self-hosting.

And the most honest point: there is no cost argument. Our current CRM costs us nothing extra per additional client environment. So moving doesn't gain anything on that front. The reason is ownership, not money, and anyone who tells you open source is always cheaper hasn't counted the hours.

05 / the statusIs this running yet?

No. Decided on July 29, not yet installed.

The translation table between the old and the new system exists — which field lands where, what merges, what disappears. But that's a document, not proof: no connection has been tested live yet.

Why we're writing it down now anyway: because the trade-off is useful before we finish it ourselves. You don't have to wait for our migration to run the four questions from block 2 against your own system.

And if the outcome of that is that you're fine where you are — that's fine too. Switching for the sake of switching is more expensive than lock-in.

Frequently asked questions

What exactly is vendor lock-in?+

The cost of leaving is so high that you stay, even if you'd rather not. That cost is rarely in a penalty clause but in rebuild work: your pipeline, your automations, your custom fields and your conversation history don't come along in an export. Your contacts do — and that's exactly the easiest part.

Is an export button worthless, then?+

Not worthless, but it proves less than you think. Test it: export one contact and look at what's actually in the file. Does the history come with it? Are your custom fields there with their meaning? Usually the answer is no, and then you know where you stand before you need it.

Why did you choose Twenty?+

Three reasons: it's open source, so the software still exists if the vendor disappears; it has a built-in connection for AI assistants, which for us makes the difference between building a middle layer or not; and it separates clients at the database level instead of with a column, which is a stronger separation.

Is open source cheaper?+

In our case, no. Our current CRM costs nothing extra per additional client environment, so the move gains nothing financially. The reason is ownership: an environment within our account is exactly what the client does not own. Anyone who tells you open source is always cheaper hasn't counted the management hours.

Is this already running for you?+

No. The decision was made on July 29, 2026 and nothing has been installed yet. The translation table between the old and the new system exists, but that's a document, not proof — no connection has been tested live yet. We're writing it down now because the trade-off is useful before we finish it ourselves.

30 minutes, free, no sales pitch

Want to know how stuck you actually are?

Schedule a call

We'll look at where your time is leaking and tell you honestly whether we can help. Often the answer is: you can do this yourself. Then we say that, and you've spent thirty minutes getting a clear answer.

→ Schedule a call (30 min)

This text was produced with AI assistance and checked and approved by a human before publication.