The Account Management Function

I built an account management program from a starting point a lot of owner-led businesses share: strong sales, solid delivery, and nothing structured in between. Accounts stuck around simply because nothing was broken. But there was no real relationship and little care if any for the team or service.

As a result, only 25% of landed accounts were still active in any meaningful capacity, underutilization was catastrophic, and feedback was impossible to gather, turning product/service improvement into guesswork.

This is buying a cold beer and getting only two sips because there's a gaping hole in the bottom of the mug - where'd all my beer go?

What was actually missing

1. No defined cadence for checking in with existing accounts. Some got attention weekly, others went months without a real conversation, and it depended entirely on who happened to be paying attention.

2. No signals tracked for at-risk accounts. A usage drop, a pattern of complaints, or slower replies to emails all meant something, but nothing was watching for them.

3. No clear owner for a client's relationship once the sale closed. Sales handed it off, and from there it belonged to whoever picked up the phone.

4. No system for spotting expansion opportunities. The only trigger for a real conversation was a client threatening to leave.

5. Retention data that couldn't explain why an account actually left or stayed. Churn got noticed. The reason behind it rarely got captured anywhere useful.

If this sounds familiar: name why your last two lost accounts actually left, without guessing. If you can't point to a real reason, retention in your business is running on hope instead of a system.

What building the function actually requires

1. Define the ICP by attributes, not just a label. Look at your different customer types and personas, then look specifically at your best accounts and figure out what attributes actually make them your best. Revenue size is part of it, but the real signal is usually in behavior: how they use the service, how they communicate, what they value.

2. Map the phases an account moves through. Onboarding, active, renewal-risk, expansion. "Check in" means something different at each phase, and treating them all the same is why nothing gets caught early. I like to think of it in terms of projects -> when a deal is in motion, its a project that deserves all the coordination a normal ops project would.

3. Define the data points needed to know where an account sits in that phase. Usage patterns, engagement, complaint history, contract timing. These are the actual inputs, more concrete than a general impression of how things are going. This is the key to unlocking which customers are your best because this is where they tie into your service outputs -> what are they buying from you or costing you.

4. Build messaging and nurture sequences tied to phase and attributes. A new account needs a different message than a five-year account facing an upcoming renewal. A single quarterly newsletter can't do both jobs. Different buyer personas also come into play here -> your marketing materials should target the audience more like a sniper and less like a shotgun blast wherever possible, and this is the data that helps you pull that off.

5. Centralize all of it in one hub. The CRM becomes the place where account, delivery, and marketing people work off the same picture, instead of three different spreadsheets and whoever remembers the last conversation. If you want to leverage technology, you've got to build your "digital brain". If you want AI to play with your data, you've got to have clean data and processes to keep it maintained by the people creating the outputs.

   
   

The actual goal

Retention and expansion become a designed outcomes instead of a byproduct of who happens to hold the relationship that week. Making that happen means making the phases, the signals, and the messaging explicit enough that anyone on the team can execute the same playbook, without needing to add headcount to pull it off.

Six months in

Once this is running, the differences are visible. Every account has an owner and a phase. At-risk signals get caught and addressed before a churn conversation becomes necessary. Expansion conversations start from data instead of a lucky guess about timing. Reporting can actually answer why an account left or stayed, instead of just recording that it happened.

The lesson underneath it

Retention without a system is luck wearing a job title. It works until the person holding it all together leaves, or the business grows past what one person can track in their head.

If you're wondering whether your retention is designed or just lucky so far, that's exactly what the Growth Diagnostic is designed to uncover.