One data layer, 29 apps

Most software that calls itself all-in-one is really a bundle: separate applications behind one login and one invoice, each keeping its own copy of your data. LinearPilot ships 29 built-in apps, but the count is the least interesting thing about it. What actually matters is what sits underneath them.

The phrase we use is “one data layer,” and it has a concrete meaning. A customer isn't typed into the CRM, then typed again into the mail tool, then typed a third time into the thing that sends invoices. There is one customer record. The apps are different views and workflows over the same shared records, not islands that occasionally sync.

A few objects act as spines that everything else leans on. Clients is the master registry of the people and organizations you serve, and it holds both: a client can be a person, an organization, or a person attached to one. Drive holds the files, with real version history you can open and restore rather than a folder full of names ending in final-v3. Knowledge Base is the clearest illustration of the whole idea, because it isn't a second filing cabinet at all — it's a curated reading layer over Drive folders. Different interface, same underlying records. When the assistant answers a question, it reads across these instead of guessing from one app's narrow slice.

This is a genuinely different thing from “integrations,” and the difference is the entire point. Integrations move copies of data between separate systems on a schedule, which is exactly why they drift, collide, and constantly need reconciling. There is nothing to sync here, because there is nothing separate to reconcile. The apps were never holding their own private copies; they were reading and writing the same records the whole time. A lot of the daily pain of running on many tools is really just sync failing quietly in the background, and one data layer doesn't have that failure mode because it doesn't have the copies that cause it.

This is unglamorous, and it pays off in the unglamorous places. The sales-to-delivery handoff that turns into copy-paste. The “which version of this contract is final.” The “who talked to this client last, and about what.” Those problems exist because the answer lives in five different tools. Put the data in one layer and the seams between the tools stop being where work goes to die.

It matters even more once you add AI. An assistant is only ever as good as what it's allowed to see. When your records are shared instead of scattered, the assistant can reason across the business rather than inside a single box, because there's a coherent picture for it to read in the first place. Fragmented data produces a fragmented assistant, no matter how good the model is.

A small, concrete example. A signed agreement carries a link to the client record, so once it's signed it turns up on that client's timeline — the same client the account manager already has open — rather than sitting in the signing app waiting to be remembered. It's worth being precise about the seam, though, because there is one. A signer is entered as a name and an email when the envelope goes out, and LinearPilot matches that back against your clients afterwards and proposes the link for someone to accept. The record ends up in one place. It gets there by matching, not by magic.

It's worth being honest about the cost of consolidating. One data layer means concentration: more of your operation living in one system, one vendor to trust, one place you need to get security right. That is exactly why the isolation and encryption work exists, and it's a fair thing to weigh. Moving off the tools you already use is real effort, too. A shared foundation is a strong reason to make that move, not a wand that makes the move free.

So the number, in perspective. 29 apps is enough that most teams find the handful they actually need and comfortably ignore the rest, which is the right outcome. The pitch was never the length of the list. It's the shared ground the entire list stands on, and the fact that adding the next app makes the ones you already use a little more useful instead of adding one more island to keep in sync.