How to calculate funnel conversion rate for your bookings

A funnel conversion rate is the share of people who took the next step, out of the people who could have. There are two versions: the whole funnel's rate, which divides by everyone who started, and one step's rate, which divides by the people who reached the step before it.

The example here is a physiotherapy clinic that takes bookings online. People land on a treatment page (back pain, sports injuries), go to the booking page, pick a time, and confirm. The clinic runs a monthly email to past patients, a listing on a local health directory and the odd short video, and wants to know which of those fills the diary. All the figures below are illustrative.

The two numbers called conversion rate

The overall rate answers: of everyone who started, how many finished? For the clinic, of everyone who landed on a treatment page, how many confirmed a booking?

A step's rate answers a narrower question: of the people who reached this step, how many went on to the next? Of the people who picked a time, how many confirmed?

Mix them up and the later steps look terrible. Divide the confirmations by everyone who ever landed on a treatment page and the confirmation step seems to be failing, when in fact most of those people never got near it. Both numbers are correct. They answer different questions, and only the step rate tells you whether a particular step is working.

Worked example: the whole funnel

Overall conversion rate = (people who finished ÷ people who started) × 100

Say this happened at the clinic in one month:

Overall rate: 90 ÷ 2,000 × 100 = 4.5%. Of every hundred people who read about a treatment, between four and five booked. That is the number to quote when someone asks how the website is doing at filling the diary.

Worked example: one step at a time

Step conversion rate = (people who reached this step ÷ people who reached the step before it) × 100

Using the same month:

The last step converts at 64.3% even though only 4.5% of the funnel gets there. That is not a contradiction. The overall rate carries everyone lost earlier; the step rate counts only people still in the running.

Read the step rates for where to look.

Here the weak point is the booking page itself: 72% of the people who open it leave without picking a time. That is worth a look before anyone writes another treatment page. Perhaps the next free slot is three weeks away, or the page asks people to create an account first. Reading a funnel drop-off goes through how to work out which.

The counting mistake: people who skip the start

The formulas assume everyone at a step walked there from the step before. Often they did not.

Plenty of the clinic's patients never see a treatment page. They tap "Book" on the clinic's search listing, or follow a link in the email to past patients, and land straight on the booking page. If they are counted as if they started at the top, the booking page looks busier than the treatment pages fed it, and every step rate after it is off.

The fix is to keep two counts apart:

• "How well does our path work?" Count only people who started at the first step. You are measuring the path, so only people who took it belong in it
• "How many bookings did we get?" Count everyone. A booking is a booking however the person arrived

Most analytics tools can split a funnel by entry page. If yours cannot, at least look at how many people land directly on the booking page, so you know how much of the number is people who skipped the start.

The rate that decides what to do next

The step rates tell you where the website leaks. The rate that tells you which marketing to do more of is the overall rate split by where people came from.

Read like this, the email is small but excellent, the directory is steady, and the videos bring plenty of visitors who mostly do not book. That changes the next month's plan: send the email more often, keep the directory, and either change what the videos ask people to do or spend those hours elsewhere.

Tag your links so the split is possible.

UTM tags are small labels on the end of a link, such as utm_source=newsletter, that tell your analytics where a visit came from. Add them to every link you control: the email, the directory listing, the link under each video. Without them, most of this table lands in "everything else". Free and low-cost tools all read them; Plausible vs Fathom compares two simple ones.

Small numbers move on their own

Getting the arithmetic right does not make a small number meaningful. If 12 people picked a time this week and eight confirmed, the rate is 66.7%. If one more had confirmed, it would be 75%. Nothing about the clinic changed; one person's choice moved it by eight points.

Three habits help:

1. Widen the date range before you narrow the funnel. Three months of a small funnel beats one week of it
2. Say the counts out loud. "Nine bookings from the directory, up from five" is clearer than "conversion up 80%"
3. Decide monthly, not daily. Looking every morning means reacting to noise

Frequently asked questions

Two formulas, then one decision

The overall rate divides by everyone who started. A step's rate divides by the people who reached the step before it. Keep people who skipped the start out of the path, and do not read a verdict into a rate built from a dozen people.

Then split the overall rate by source, because that is the version that tells you what to do next. Mole Marketer's weekly initiative pitches are built on the same idea: every initiative gets its own UTM tags, the included Funnel Analytics counts what each one brought in, and that feeds the next pitch. Your team still decides what to run.