Attendee lifetime value is the present value of what one attendee contributes across several
editions of your event: registration, ancillary spend and, for an association, a share of dues,
less the cost to serve them, weighted by the chance they return each year. Over five years it
is larger than the contribution from the first badge, and how much larger depends mostly on
your return rate. It also sets the most you should pay to acquire a first-time attendee.

Take a hypothetical $190 acquisition cost against an $895 badge. It looks comfortably profitable, and that
reading is wrong in both directions: it ignores the four later years of registration, workshop
tickets, and dues that follow from the first attendance, and it ignores that many first-timers
will not come back.

## What does lifetime value mean for an event rather than a subscription?

Subscription lifetime value (LTV) is straightforward because the relationship is continuous and
cancellation is observable. Event attendance is neither. Someone who skips a year has not
churned and may be back when the location suits them or their employer approves travel. The
relationship is episodic, so the right model is a repeat-purchase model with a per-year return
probability.

You cannot infer that probability from who cancelled, because nobody cancels. You measure it as
the share of a fixed cohort that shows up again, which requires identifying the same person in
two different years. Counting who came back is its own job, covered in the guide to
[attendee retention](https://eventiq.io/md/blog/attendee-retention). This page starts where that one stops: it takes
the return rate as an input and puts a price on it.

Freeman's 2025 end-of-year trends report puts the industry average for attendee
retention at 30% to 35% year over year
([Freeman](https://www.freeman.com/wp-content/uploads/sites/4/2026/01/Freeman-EOY-Report.pdf)).
Freeman supplies services to the market it measures and does not publish the sample or method
behind that figure, so read it as an order of magnitude and use your own rate in the model.

## What goes into the revenue side?

Three components. The third applies only to membership organizations.

**Annual contribution per attendee (year t) = Registration revenue, net of discounts + Ancillary revenue + Attributable dues − Variable cost to serve**

Variable cost to serve is food and beverage, materials, the badge, and onsite service, the same
per-head figure your [event budget](https://eventiq.io/md/blog/event-budget-template) uses.

Ancillary revenue (workshops, ticketed events, certification, on-site purchases) is usually
understated because it sits in different systems from registration: a certification fee in the
membership system, a workshop ticket in the registration platform, a dinner in a separate
ticketing tool. If you cannot assemble it per person, use a per-head cohort average and label
it as one.

Attributable dues need a defensible share of the dues line, never the whole of it. A member who
attends is not a member because they attend, and assigning 100% of dues to the event overstates
the case badly. Assign only the renewal-probability difference between attendees and a
comparison group, or a fixed conservative share agreed with finance in advance. State it in the
report, because it is the assumption most likely to be challenged. The guide to
[member retention and event attendance](https://eventiq.io/md/blog/member-retention-event-attendance) covers the
causal question behind it.

## How do the survival and discount pieces work?

LTV sums the annual contribution across five years, weighted by the probability the person is
still attending and discounted back to today's money.

**Survival probability to year t = Return rate ^ (t − 1)**

**Present value of year t = (Annual contribution × Survival probability to year t) ÷ (1 + Discount rate) ^ (t − 1)**

**Attendee LTV (5 years) = Sum of the present values of years 1 to 5**

The constant return rate is a simplification. People who come back a second time tend to be
more likely to come back a third, and where that holds in your data a single rate understates
the value of the people who stay. If you have the history to measure first-timers, second-year
attendees, and long-tenured attendees separately, use three rates. If not, use one and say so.

## What discount rate should you use, and does it matter?

Use your organization's own cost of capital, or in a non-profit the rate finance already uses
for multi-year commitments. Do not invent one or borrow a corporate figure.

The rate matters less than people expect over five years at a moderate return rate. Most of
the present value sits in years 1 and 2, because survival decays faster than discounting does.
In the example below, moving the discount rate from 6% to 10% lowers Cohort A's LTV from $1,217
to $1,188, about 2%, and Cohort B's from $2,101 to $2,013, about 4%. Moving Cohort A's return
rate from 45% to 50% at an 8% discount rate raises its LTV from $1,202 to $1,294, about 8%.
Measure the return rate accurately and spend less time on the discount rate.

## Why is a first-time attendee worth more than the badge they bought?

Because the badge is year 1 of 5, and the first attendance creates the possibility of the other
four. A single-year view sets the acquisition ceiling at a fraction of one registration's
contribution. The five-year view applies the same fraction to a larger number, and how much
larger depends on your return rate.

None of this argues for unlimited acquisition spending. The point is narrower: the ceiling is
being set by the wrong number. Organizations that budget acquisition against a single badge
underspend on first-time attendance, which is the only input to the repeat-attendee population.

## How do you handle survivorship without flattering yourself?

Attendee-value figures usually fail the same way: the denominator moves. If you take the people who attended this year and ask how many
attended last year, you are measuring survivors, because everyone in that sample already came
back. Fix the cohort at the start and follow it forward.
| Year | Attended | Cohort survival | Year-over-year return rate |
| --- | --- | --- | --- |
| 1 | 600 | 100.0% | denominator |
| 2 | 270 | 45.0% | 45.0% |
| 3 | 141 | 23.5% | 52.2% |
| 4 | 82 | 13.7% | 58.2% |
| 5 | 49 | 8.2% | 59.8% |
**Cohort survival = Attended in year t ÷ Attended in year 1**

**Year-over-year return rate = Attended in year t ÷ Attended in year (t − 1)**

Report both columns. They answer different questions and are routinely confused.

Three further traps. Attendees who change employers and re-register with a different email look
like new attendees and depress the return rate. A host-city change can suppress one year's
attendance for reasons unrelated to loyalty, so a curve built across a full rotation of cities
is more trustworthy than a three-year one. And a cohort from an unusually strong year shows a
worse survival curve than a typical one, so pick a representative year and name it.

There is also a hard practical limit. Following one person across five years means matching
them across years and usually across systems, and most organizations cannot do it reliably. The
fallbacks, in descending order of quality:

- a stable person identifier issued at first registration;
- membership ID, where the attendee is a member;
- email-based matching with a documented error rate;
- self-reported first-time status at registration, which is the weakest.

The last is a single checkbox and worth adding today. Where the method is weak, label the
figures approximate and name the method.

## What acquisition ceiling does LTV imply?

A share of LTV, never all of it: you need the margin, and the model is uncertain.

**Acquisition ceiling per new attendee = Attendee LTV × Target payback share**

**Payback period (years) = Years until the accumulated discounted annual contribution exceeds the acquisition cost**

Set the payback share with finance. Allowing acquisition spending up to a third of LTV with
payback inside two years is one workable posture. It is a policy choice with no benchmark
behind it, and the right number depends on your cash position and how much you trust your
return rate. Publish the ceiling as a rule and the payback as a monitor, so marketing can spend
up to a line without asking each time.

## Example: two hypothetical cohorts over five years

Take two cohorts at an association annual meeting. All figures are hypothetical and are not
benchmarks.
| Line | Cohort A: non-member attendee | Cohort B: attending member |
| --- | --- | --- |
| Registration revenue, net | $780 | $690 |
| Ancillary revenue | $120 | $150 |
| Attributable dues | $0 | $195 |
| Variable cost to serve | −$190 | −$190 |
| Annual contribution | $710 | $845 |
| Return rate per year | 45% | 68% |
| Discount rate | 8% | 8% || Year | Discount factor | Cohort A survival | Cohort A present value | Cohort B survival | Cohort B present value |
| --- | --- | --- | --- | --- | --- |
| 1 | 1.0000 | 1.0000 | $710 | 1.0000 | $845 |
| 2 | 1.0800 | 0.4500 | $296 | 0.6800 | $532 |
| 3 | 1.1664 | 0.2025 | $123 | 0.4624 | $335 |
| 4 | 1.2597 | 0.0911 | $51 | 0.3144 | $211 |
| 5 | 1.3605 | 0.0410 | $21 | 0.2138 | $133 |
| Five-year LTV |  |  | $1,202 |  | $2,056 |
Rows are rounded to the dollar and totals are computed before rounding, so Cohort A's rows add
to $1,201.

At a 33% payback share, the five-year view allows $397 to acquire a Cohort A attendee and $678
for Cohort B. A single-year view, 33% of the year 1 contribution, allows $234 and $279. The
ratio between the two views is the ratio of LTV to the first year's contribution. For Cohort A
that is $1,202 ÷ $710 = 1.69, about 69% more. For Cohort B it is $2,056 ÷ $845 = 2.43, more
than double.

The second reading is the retention argument, and it is the more valuable one. Cohort B is
worth $854 more than Cohort A at present value, and most of that gap comes from the return
rate. Keep Cohort A's $710 contribution and raise only the return rate to 68%, and LTV is
$1,727, which accounts for $525 of the $854.

Now run Cohort A at a 55% return rate in place of 45%, holding everything else constant. Its
LTV rises to $1,397, a gain of about $195 per person. Ten points of return rate on a cohort of
600 first-timers is worth around $117,000 of present value on these hypothetical numbers.

## How does this reframe the retention argument?

It turns retention from a soft measure into a capital allocation question. "We should look
after first-timers" is a sentiment. "Ten points on the return rate is worth $117,000 in present
value against a program that costs $40,000" is a proposal, shown here on hypothetical figures.
The LTV model is useful mainly because it makes the second sentence possible.

It also changes what counts as a retention lever. Anything that raises the return rate
compounds across the whole curve: a first-timer orientation, a second-year discount, a chapter
connection, a track built for early-career attendees. Anything that raises this year's yield
without touching the return rate moves only year 1.

## What to do this quarter

- Add a first-time-attendee checkbox to registration, and issue a stable person identifier at
  first registration.
- Build one cohort survival table on a representative starting year with the denominator
  fixed, and report both cohort survival and year-over-year return.
- Assemble ancillary revenue per head, even as a cohort average, and note which systems it
  came from.
- Agree the attributable-dues share with finance in writing before you calculate anything.
- Calculate LTV and the acquisition ceiling for two or three segments, not one blended figure.
- Put the ten-point return-rate calculation in front of whoever approves the
  first-time-attendee budget.

## Common questions

### Five years, or some other horizon?

Five is a reasonable default, and a horizon longer than your strategic plan invites arguments.
If your return rate is high, model seven years and check whether the answer moves enough to
matter. In the example, seven years adds about 1% to Cohort A's LTV and about 7% to Cohort B's.

### We cannot match people across years at all. Is this pointless?

No, but label it. Calculate LTV by segment using self-reported first-time status and a return
rate estimated from whatever matching you can do, present it as an estimate with a stated
method, and start capturing a stable identifier now.

### Should we include sponsorship revenue in attendee LTV?

Not per attendee. Sponsorship is sold against the audience as a whole, and pushing it down to
an individual requires an allocation that will not survive scrutiny. Keep it as a separate
audience-level stream, priced the way the guide to
[event sponsorship packages](https://eventiq.io/md/blog/event-sponsorship-packages) sets out.

## Where EventIQ fits

EventIQ replaces nothing. It connects on top of the platforms you already run: event platforms
(Cvent, Zoom, Swapcard), CRM (Salesforce, HubSpot, GoHighLevel), and marketing (Google Ads, Meta
Ads, LinkedIn Ads, Mailchimp, Google Analytics). Platforms with an API outside that list are
connected on request.

EventIQ does not calculate attendee lifetime value, and it does not build the cohort survival
table. Both are yours to run. The product holds some of the inputs.

Registrations from Cvent carry their date, ticket type, and price where the platform provides
them, together with an attendance mark. Attendees are matched to CRM contacts by exact email,
and the result stays on each record: matched, unmatched, or no email. Event costs in the
product are budget figures your team enters.

Two limits. There is no connection to an association management system, so dues and membership
status come out of your membership system by hand, which [associations](https://eventiq.io/md/for/associations)
should know before a demo. And exact email is a strict key: a person who re-registers with a
new address does not match the CRM contact held under the old one.

[Book a demo](https://eventiq.io/#early-access) to see registration records with ticket type and price,
attendance, and the email match result on a sample event, in a 20-minute demo.

---

HTML version: https://eventiq.io/blog/attendee-lifetime-value