CMS or CRM, and why your firm keeps confusing the two

Two systems, two jobs

Take the definitions plainly. A case management system, the CMS, is the system of record for matters. It holds the open files, the parties, the key dates, the documents, the workflows and the ledgers. Its job is the accurate conduct and administration of live legal work, and in England and Wales the familiar names are LEAP, Clio, Proclaim and their peers. When a fee earner asks where the file is, the answer is the CMS.

A client relationship management system, the CRM, is the system of record for relationships. It holds the contacts, the enquiries that have not yet become matters, the referral sources, the marketing lists and the history of who spoke to whom. Its job is the front of the firm, winning and keeping clients, rather than conducting their matters. In a larger firm this is a separate product with its own owner in the business development team.

Why small firms blur them

Small firms blur the two for a sound reason. Their CMS does some of the CRM's job. It stores contacts, it records enquiries, it keeps a history against the client, and for a firm of a dozen people that is often enough that no separate CRM is ever bought. So the CMS quietly becomes the place where both the matters and the relationships live, and the word CRM gets used loosely to mean whatever holds the client's details. That is not a mistake to be ashamed of. It is a reasonable response to a firm's size. It becomes a problem only when the blur hides where your data sits, because you cannot add intelligence to a system you cannot locate.

Here is the practical way to think about it. Ask two questions of any process you want to improve. First, is this about a live matter or about a relationship? Chasing a missing document, checking a limitation date and preparing a bill are matter questions, and they live in the CMS. Following up a stale enquiry, spotting a client with no recent instruction and keeping a referrer warm are relationship questions, and they live in the CRM, or in the part of your CMS doing the CRM's job. Second, where does that data physically sit in your firm? For most small firms the honest answer to both is the same system, and that is fine. Knowing it is the same system is what matters.

The question that changes the buying decision

Now the part that changes the buying decision. Adding intelligence to either system does not require you to migrate off it. Both classes of product expose an API, the interface that lets approved external software read and write their data under permissions you control. That is how your accounts package and your e-signature tool already connect. An AI layer connects the same way. You do not move your matters to put an assistant on top of them, and you do not replace your contact history to add intelligence to it. The system of record stays. The intelligence plugs in.

This is why the CMS or CRM question matters before you buy anything with AI in the name. If a tool only reads the open file, it is helping with one matter at a time and touching neither the caseload view nor the relationship view. What a small firm usually wants is something that reads across the whole record, wherever that record sits, and works on both the matter side and the relationship side, because in your firm they are often the same database. The right question to a vendor is not which system they replace. It is which of your systems they connect to, and whether one tool can read across both.

Bracton comes in two forms, and which one suits your firm turns on whether you are free to change system. Bracton is the whole case and client management platform, holding the matters, the documents, the diary, the ledgers and the compliance record, with the assistant built in and reading across every file. A firm free to change system takes that. A firm tied into a contract with another provider takes the Bracton AI Assistant on its own, running inside the case management or client relationship system you already have through the API it exposes, and what you pay for the assistant is credited against the platform in full if you move to it within twelve months of starting. Either way the assistant reads across the caseload rather than one open file, prepares routine work on the matter side and surfaces the drifting relationships on the other. Every research answer cites a source, every output waits for a qualified fee earner to sign off, the audit trail records what was done, hosting is in the UK and your client data is not used to train a model.

The way to see the difference is to watch the whole platform work an invented firm, which the demonstration runs in your browser. The way to start on something of your own is the client account, read against the rules and answered in writing: book a client account review.