Most MGAs we talk to have a CRM. Most of them bought it in the last three years, and most of the records in it stop about a fortnight after onboarding.
The standard reading of that is an adoption problem, and the standard answer is more training and a standing item in the Monday meeting. We think that's the wrong answer. But the case for the CRM is better than it usually gets credit for, so it's worth putting properly before taking it apart.
The case for the CRM is a good one
A CRM does several things nothing else in an MGA does.
It holds a producer record that outlives the person who built the relationship. Underwriters leave. What's in their head goes with them and what's in the system doesn't, and anyone who has taken over a book at short notice knows exactly how much difference that makes.
It gives whoever runs distribution a list rather than a set of impressions. Sixty-odd producers, sorted, with something written against each one. Even a half-maintained list beats going round the table.
It makes the compliance side tractable. TOBAs, appointment records and who's authorised to bind what. Most MGAs are running that on a spreadsheet and a shared drive, which is a rubbish way to keep an appointment register, though at least an honest one.
And for some teams it plainly works. If you've got business development people whose entire job is opening producer relationships from a prospect list, a pipeline tool is the right purchase and we'd tell you to buy one. That's real work, it's shaped like a pipeline, and the people doing it are measured on it.
So the case is sound. Where it breaks is narrower than "CRMs don't work at MGAs", and it's worth being exact about where.
It assumes a person who doesn't work here
Every mainstream CRM is built around the enterprise salesperson. Someone who lives in a pipeline, whose day is dictated by the list, who logs the call because logging the call is the job. For that person, filling in the CRM is the work.
An underwriting team is a different animal. Underwriters are hired to price risk, they're good at pricing risk, and a fair number of them came into the industry because that's the part they enjoy. A quote request arrives from a broker they rate and they answer it. A bind instruction arrives and they bind it. The work happens in the thread.
So the CRM turns up asking for a tax. Log the contact, set the stage and put a next action against it. And it hands nothing back that helps price the next risk. People pay a tax like that for a few weeks while somebody is watching, then volume rises and they stop, which means the record goes thin at exactly the point in the year you most wanted it.
The record the CRM was asking for already exists in email
Here's the part the adoption conversation misses. The data the CRM wanted typed in is generated anyway, every day, as a by-product of the work itself.
Every submission, every quote request, every bind instruction and every renewal thread that reopens six weeks after everyone wrote it off arrives in somebody's inbox. Every reply leaves a timestamp. Which threads stayed alive, which went thin, and which producer has been sending more than they used to since capacity started coming back. Nobody had to log anything for any of that to exist.
What it isn't is legible. It sits one inbox at a time, in no order, and nobody can look across it on a Monday and say which producer relationships need attention this week. That's the actual gap, and it's a different gap from the one the CRM was sold to fill.
The shared submissions mailbox is usually older than the CRM, incidentally, and still working. Four people can see it. Nobody owns it.
Every line here came out of metadata your team generated by doing the work. Nobody logged anything to produce it.
An honest test for whether you need one
Three questions. If any of them is a yes, buy the CRM for those people and don't let anyone talk you out of it.
- Do you run a genuine outbound motion, with people whose whole job is opening producer relationships from a prospect list?
- Do you have distribution headcount who don't underwrite, and who are managed on activity and stage progression?
- Are you running stage-based projects that move through gates over months, like carrier appointments, programme launches or an acquisition pipeline?
If they're all no, and at the MGAs we've looked at they usually are, then the CRM is solving a problem you don't have while the one you do have sits in the mail.
We don't know what share of MGAs those three questions would disqualify. Our sample is the ones who agree to talk to a vendor, which isn't a neutral sample, and we've been on the vendor side of the table for long enough that it probably colours the reading.
Something to check this week
Open the CRM and sort producer records by last modified. Look at where the dates stop.
Then take the three records at the top, the most recently touched ones, and ask whether those are your three most commercially active relationships this month or just the three somebody happened to update. Five minutes, and you'll know which of the two you're paying for.
For transparency, reading that record out of Microsoft 365 email metadata and turning it into a weekly email is what we build at BindSignal. The argument works without us. However you get there, a standing review across inboxes, an export or something purpose-built, the principle is the same. The intelligence should fall out of the work rather than being a second job stacked on top of it.
Your underwriters aren't salespeople, and the answer was never to make them into salespeople. It was to stop buying software that needs them to be.