A 90+ driver program is not “focus more on 90+.” It’s a defined way of running the work.

It starts with protected capacity, so 90+ doesn’t disappear the moment something urgent happens. It includes simple rules that separate accounts you can move now from accounts that are blocked upstream. It creates clear ownership for anything stuck. It standardizes escalation so results don’t depend on who happens to be working the account that day. And it uses a small set of operational metrics that reveal leakage early, before it shows up as an ugly month-end surprise.

The goal is straightforward: stop time leakage first. When repeated touches and rework go down, collections become more predictable, and revenue stops slipping away quietly.

Step 1: Protect capacity so 90+ work can’t be borrowed

If time for 90+ is optional, it will be taken every week.

There are a few workable ways to protect it. Some teams assign a dedicated owner or small group that owns the 90+ queue and becomes skilled at payer patterns, escalation paths, and resolution steps. Others use a rotating model, where someone owns 90+ coverage for a week or two at a time, which builds cross-training while still protecting the work. Another approach is time-blocking, such as reserving certain mornings each week for 90+ follow-up with a clear rule that the block cannot be used to catch up on current AR.

Whichever method you choose, the point is the same. Protected capacity is the foundation. Without it, everything else turns into a plan that only works on calm days.

Step 2: Segment 90+ by collectability state, not just by age

One reason 90+ becomes inefficient is that it gets treated as a single category. In reality, aged accounts usually fall into different states, and each state needs a different approach.

Some accounts are ready to collect. They have what you need, and a clear next step is available now. These should move quickly and consistently, with the goal of closure, not activity.

Other accounts are blocked. They cannot move until something specific is resolved upstream, such as missing documentation, a coding clarification, authorization proof, medical records, or a payer-specific requirement. Blocked accounts need a firm rule: they must have a named owner, a due date, and a defined barrier. If “blocked” becomes the place accounts go to wait, that’s where leakage forms.

A third group requires escalation. These are accounts with payer complexity or disputes that need a defined path, such as peer-to-peer follow-up, supervisor-level payer contact, appeals, contractual review, or an internal leadership decision. The key here is consistency. Escalation should be a standard process, not an improvisation.

Finally, some accounts are low value or low probability. That doesn’t mean they are ignored. It means the organization makes an intentional decision about how much time to invest, especially when balances are small and time cost is high, or when repeated attempts fail without a new pathway. This is not giving up. It’s protecting time so higher-value accounts can be collected.

Segmentation creates clarity. It makes the work more predictable, and it reduces wasted effort.

Step 3: Define ownership so touches actually move accounts forward

Revenue leakage often looks like repeated touches without progress. The same calls happen again. Notes don’t clarify what changed. Accounts return to the queue in the same state they left.

A strong 90+ program fixes this by standardizing what every touch must produce. After each touch, there should be a clear next action, a named owner, a deadline, and a definition of done. There should also be an escalation trigger, meaning what happens if the deadline passes.

That one standard shifts the posture of the work. It turns “busy follow-up” into forward movement.

Step 4: Use metrics that expose leakage early

Many teams look at aging totals and feel stuck because totals lag behind reality. By the time totals show the problem clearly, the leakage has been forming for weeks.

A 90+ driver program uses a small set of operational measures that surface friction sooner. Touches per account without resolution is one of the most revealing. If touches increase, rework is rising. Another is time-to-next-step, which shows how long accounts sit after activity before anything meaningful happens. The percentage of 90+ accounts in blocked status is also important. If blocked keeps growing, upstream workflow issues are driving aging.

You also want to track aging velocity, which shows how quickly accounts are entering 90+. And finally, resolution cycle time by segment matters because ready-to-collect should close fastest, escalation should follow a consistent path, and blocked should have owners and deadlines.

These metrics keep the team focused on the system, not just the totals.

Step 5: Create a weekly rhythm that keeps 90+ in the driver seat

The goal is not to add meetings. The goal is consistent barrier removal.

A lightweight weekly rhythm usually works best. A 30–45 minute driver review can be enough to look at segment counts, identify the most common blockers, and assign owners and deadlines. From there, a barrier-removal loop routes blocked accounts to the right upstream owners, whether that’s coding, documentation, clinical teams, eligibility, or another internal partner. For payer patterns, the team captures repeat requirements and builds a standard response path instead of rediscovering it every time.

Then comes the piece most teams skip: upstream fixes. Choose one or two repeat causes each week and assign a fix with a deadline. That might be updating a template, adding a documentation checklist, changing a handoff standard, or capturing payer rules in a usable way.

This is where long-term leakage actually stops, because the organization stops recreating the same 90+ problems.

Step 6: Decide what to automate and what must stay human

Automation doesn’t replace judgment. It removes friction.

In a 90+ driver model, automation is best used for routing worklists by segment, reminders and SLA timing for blocked items, escalation triggers when deadlines pass, payer-specific follow-up templates, documentation completeness flags, and dashboards that show what is blocked, why it’s blocked, and who owns it.

Human work remains essential for payer disputes and negotiation, clinical or documentation interpretation, escalation conversations, cross-department decisions, and exception handling. The goal is to protect human attention for what truly requires it.