Most referral teams report activity: visits made, lunches delivered, calls logged. Activity is easy to count, but it doesn't tell you whether any of it is working. A marketer can make forty visits in a month and win nothing, or make ten and double a key relationship.
The number that connects effort to results is referral share.
What referral share means
For any source, referral share is the portion of the patients they refer out, for the services you provide, that come to you. Here's an illustrative example: if a hospital discharges 80 patients to home health in a quarter and 20 of them come to your agency, your share of that hospital is 25%.
Raw referral counts can't separate two very different situations:
- The hospital is sending fewer patients to everyone, because of seasonality, a unit closure or a census dip.
- The hospital is sending the same number of patients but fewer to you, because a competitor got there first.
Referral share tells you which one is happening. The first is a market condition. The second is a problem your team can fix.
How to estimate it
You need two numbers for each source: how many patients they refer out, and how many came to you.
Your side comes from your own records. Count referrals by source from your intake or care software, and keep the source names consistent so the counts add up.
Their side is harder. Claims-based market data can estimate how many patients a hospital or physician sends to post-acute care. Treat those estimates with care: they're largely based on Medicare fee-for-service claims, they usually run months behind, and private-pay clients don't appear in them at all. For a private-pay home care agency, the most honest denominator is often what the partner tells you about their volume, or a trend you watch over time.
The estimate doesn't have to be perfect. It has to be consistent from quarter to quarter, so that changes mean something.
How to use it
Rank opportunities. A large source where you have a small share is usually a bigger opportunity than a small source you already dominate. Sort your list by the referrals you don't get, not the ones you do.
Spot problems early. If a partner's share drops two quarters in a row while their total volume holds steady, someone else is winning that relationship. That's the moment for a visit, not six months later.
Judge outreach by outcome. When a marketer works a source, look at whether share moved in the following months. Over time you'll learn which kinds of outreach (an in-service, a lunch, a quick update on a shared patient) actually change behavior with which kinds of partners.
Pair it with two other numbers
Share on its own can mislead, so read it alongside:
- Referral-to-admission rate: how many referrals from a source become clients. A source can send a lot and convert poorly if the referrals don't fit your services or your payer mix.
- Two-way balance: for facilities, how many patients you send them versus how many they send you. Partnerships that only run one direction tend not to last.
Report it the same way every time
Pick a period (quarter over quarter is usually steadier than month over month), keep source names consistent, and show the trend rather than a single number. A simple table works: source, referrals to you, estimated share, change versus last period. The point isn't a perfect dashboard. It's a weekly conversation about where to spend the team's time.
Wellby is built around this measure: it shows your share of each source's referrals, how it moves quarter over quarter, and which outreach came before the change.
