Read the Weather, Not the News

Every month, a multi-site healthcare platform gets a news report. The P&L closes, EBITDA lands, collections post, and somebody builds a deck explaining why the numbers came in where they did. The news report is accurate, but it arrives late. By the time a margin miss shows up, the cause is usually six to eight weeks old, and the people who could have prevented it have already moved on to the next fire.

I spend most of my time on a different report. I call it the weather report: the small, daily signals that tell you what the news will say next month.

What lives in the weather report

Leading indicators are simple counts that your front line already produces, usually in a scheduling system or a phone system nobody looks at above the site level.

• Open shifts for the next 14 days. When open shifts climb, overtime, agency spend, and missed visits follow. Staffing is the first domino in almost every margin problem I have worked on.
• Call answer rate. An abandoned call today is an empty appointment slot next month. At one PE-backed specialty group, we raised the call answer rate 70%, and the schedule filled behind it.
• Authorization and intake backlog. In ABA, every week a family waits for an authorization or an assessment is a week of care and revenue that never comes back. Reworking authorization workflows site by site moved families into treatment 30 days sooner.
• Days from visit to claim. If charges sit, cash will not show up when the forecast says it will. On a New York City contract portfolio, we cut billing lag from 6 months to 14 days by fixing the handoffs between operations and revenue cycle with the CFO.
• Escalations per site. Count them, and count who resolved them. If every escalation lands on an executive, the site leaders do not yet have the tools or authority they need.

Why leaders default to the news

The news report is comfortable. Finance produces it, the board expects it, and it settles arguments. Leading indicators are messier. They live in five different systems, they move every day, and they point at people instead of at line items. It is easier to explain a bad month than to admit you could have seen it coming in week two.

The other reason is ownership. Lagging numbers belong to the executive team. Leading numbers belong to site leaders and supervisors, which means somebody has to teach them what to watch and give them permission to act.

How to start reading the weather

1. Pick about five signals. Choose the ones that predict your biggest cost and your biggest revenue risk. For most multi-site platforms that means staffing, access, authorizations, charge lag, and escalations.
2. Give each signal one owner. One person is responsible for watching it and acting on it.
3. Review them weekly, in the same meeting, in the same order. I use EOS Level 10 meetings because the scorecard comes first and each issue leaves the meeting with an owner.
4. Set a watch line and a storm line. Agree ahead of time what level triggers attention and what level triggers action. That removes the debate about whether a number is bad.
5. Connect the weather to the news every month. Show the board how last month's leading indicators predicted this month's results. Over time, that builds the board's confidence in the weekly numbers.

What it looks like when it works

At a 16-site operation running 24/7 with more than 2,200 employees, staffing fill sat at 65%. The monthly numbers said we had a problem. The daily shift data told us exactly where, on which shift, and in which service line. We built coverage around that view, and fill reached 105% in 90 days.

None of that required new software. It required looking at the right numbers early and training a team that knew what to do when they moved.