The Case Against SaaS Lock-In

I want to be honest with you about something uncomfortable: the business model of most SaaS companies is optimized not for your success but for your inability to leave. Lock-in isn't an accident or a side effect. It's an architectural feature. And CRM is one of the most egregious examples in all of software.

This isn't a conspiracy. It's just incentives. Once you've spent years entering contacts, building pipeline history, configuring workflows, and training your team on a CRM, switching costs are enormous. The vendor knows this. So they raise prices. Add tier restrictions. Gate features behind enterprise plans. And when you threaten to leave, they know you probably won't — because you can't easily take your data with you.

How SaaS Lock-In Works#

Lock-in operates through several mechanisms simultaneously:

Data hostage-taking: Your CRM data — contacts, notes, deal history, communication logs — lives on the vendor's servers in their proprietary format. Export options exist but are often lossy, incomplete, or restricted to certain plan tiers. HubSpot limits export to 100,000 records at a time. Salesforce exports often lose custom field configurations and relationship data. The data is "yours" in a nominal sense, but extracting and re-importing it into another system is a project measured in weeks or months.

Integration dependencies: Once your CRM integrates with your email, your calendar, your marketing automation, your billing system — and you've built workflows around those integrations — switching the CRM means re-building every integration. Integration platforms like Zapier help, but they add their own lock-in. Each integration is another anchor.

Workflow and process dependencies: Your sales team doesn't just use the software — they've built processes around its quirks. Specific views, specific pipeline stages, specific report formats. Moving to a different CRM means re-training everyone and rebuilding processes. This is often more expensive than the software migration itself.

Network effects within the product: Some CRMs create internal network effects: your data is more valuable because colleagues are also in the system, managers have built dashboards, historical data accumulates. Leaving means losing that accumulated value.

Relationship dependencies: Your customer success rep, your solution architect, your Salesforce admin — these are real relationships built around the vendor. Switching means starting over with a new vendor relationship.

The Price Escalation Pattern#

The lock-in enables a predictable pattern:

1. Vendor offers competitive pricing to win business
2. Company goes through implementation and becomes dependent
3. Vendor raises prices 5-20% annually
4. Company complains but stays (switching cost > price increase)
5. Repeat indefinitely

Salesforce has increased prices by roughly 10% annually on average over the past decade. HubSpot has moved more features from lower to higher tiers, effectively requiring upgrades to maintain the same feature set. The "platform" model means paying for the full suite even if you need only part of it.

One company I know well went from $40/user/month for a standard CRM plan to $120/user/month over five years — not because they needed more features, but because the features they were already using moved to higher tiers. The software didn't get better. The price tripled.

What Your Data Is Worth (To Them)#

Here's the thing that's rarely said plainly: your contact database, your pipeline data, your deal history — this data is genuinely valuable. Not just to you. To the CRM vendor.

They use it to:

• Train AI models that they sell to you (and others)
• Build anonymized benchmarking products
• Improve their product features
• Understand industry patterns

You're not just paying a subscription. You're paying a subscription and providing training data. The economics of cloud AI are partly subsidized by customer data. Your deal notes are teaching their AI models about sales patterns.

None of this is secret. It's in the terms of service that nobody reads. But it's worth being clear-eyed about: you're not just a customer. You're also a data source.

The MIT Alternative#

Open-source software under a permissive license (MIT, Apache 2) breaks every lock-in mechanism:

Data ownership: Your data lives in your infrastructure, in open formats you control. DenchClaw uses DuckDB — a fully open, documented database format. You can read your data with any DuckDB client. You can export to CSV at any time. No export limits. No format restrictions.

No price escalation: Software under the MIT license can be run forever with no licensing fee. The Dench team might offer premium cloud services, but the software itself is free, forever, by legal commitment.

No feature restrictions by tier: All features are available to all users. There's no enterprise plan hiding the features you actually need.

Fork freedom: If the project changes direction or stops being maintained, you can fork it. You're not dependent on a single vendor's decisions. You own the software because you own a copy of the code.

Auditable data handling: You can verify what the software does with your data. Not trust a privacy policy — verify in the code.

DenchClaw is MIT-licensed. We're not a charity and we have a business model — cloud hosting, professional support, enterprise services. But the core software is MIT, which means you're never captive to our pricing decisions. If you run DenchClaw locally, you have the full product at zero licensing cost, now and forever.

The Migration Math#

When companies realize they're locked in and try to leave, the migration math is usually sobering:

• 3-6 months of migration project time
• Engineering and data team resources
• Possible external consultants
• Training and change management
• Productivity dip during transition

That's why vendors raise prices 20% and companies stay. The switching cost is real.

But this math changes when you have clean data in an open format from the start. If your CRM data is in a DuckDB file with a documented schema, migration to any future system is a matter of transforming structured data — a days-long project, not a months-long one.

The exit strategy, paradoxically, makes you a better negotiator even if you never use it. If you can credibly leave, your vendor treats you differently.

The Philosophy#

I'll say what I actually believe: software you don't control isn't really yours. It's software you're renting from someone who has more leverage than you think they do.

This isn't about distrusting software vendors in general. Most of them are run by decent people who believe they're providing value. The problem is structural: the incentives of the SaaS business model — where recurring revenue depends on retention, not on continuous value creation — push toward lock-in regardless of the intentions of the people involved.

Local-first, open-source software doesn't have this structural problem. If DenchClaw stops being the best option for you, you should leave. Your data is in a DuckDB file on your machine. You can take it anywhere. Nothing we can do about that — and we don't want to.

That's not naivety. That's the only way to earn trust in software: by genuinely not needing to hold you captive.

Frequently Asked Questions#

Can I really export all my data from cloud CRMs?#

Most cloud CRMs offer export, but the exports are often incomplete (relationship data, custom field configurations, and activity history may not export cleanly), volume-limited (HubSpot limits bulk exports), and require knowing in advance what you'll need (once you've deleted records or fields, they may not be recoverable).

How does DenchClaw avoid lock-in if Dench goes out of business?#

DenchClaw is MIT-licensed. The source code is on GitHub. If Dench shuts down, the software continues to run. You can fork the codebase. Your data is in a DuckDB file you own. You have no dependency on Dench's servers for the core CRM functionality.

Isn't there switching cost from DenchClaw to something else too?#

Yes — any system you invest in creates some switching cost through familiarity and data accumulation. DenchClaw minimizes this by using open formats (DuckDB, CSV exports, Markdown documents) that can be read by other systems. The data format is never proprietary.

What's the catch with MIT-licensed software?#

MIT doesn't mean the software is perfect or that support is free. You're responsible for running and maintaining it. For teams without technical capacity, a managed cloud option (Dench Cloud) provides the convenience of SaaS with the data ownership of local-first.

Ready to try DenchClaw? Install in one command: npx denchclaw. Full setup guide →