Member versus non member registration pricing and what the gap buys you

The membership committee meets in September and somebody proposes widening the gap. Members currently pay 495 and non-members 695. If the non-member rate went to 795, the argument runs, more people would join, because joining costs 340 and would then save them 300 on the badge.

It is a reasonable-sounding argument, and nobody in the room can say whether member versus non member registration pricing has ever produced a single membership at this show. The show has run a member rate for nineteen years. The number has never been measured, because measuring it needs the registration file and the membership file joined on a date, and those two systems are owned by different directors.

Two things the gap is sold as, and they conflict

Internally the member rate gets justified two ways, usually in the same meeting.

It is a member benefit. People pay dues, and one of the things dues buy is a cheaper badge. Under this reading the discount is funded by membership revenue, it is supposed to go to existing members, and giving it to people who were already members is the entire point.

It is also described as a membership acquisition device. The gap is wide enough that attending the show becomes a reason to join, so the registration campaign feeds the membership funnel.

Those two readings imply opposite measurements. Under the first, discount given to existing members is the benefit working. Under the second, it is waste. An organisation that has not chosen between them will read whichever measurement flatters the decision it already made, and both numbers are available in the same file.

Choose one, write it down, and report against it. My preference is to call the gap a member benefit and to stop crediting it with membership growth, because the arithmetic below rarely survives the second reading.

What does a published gap look like?

Two large shows publish member and non-member pricing, and the structures are worth comparing before you set your own.

The NACS Show's 2026 rate card holds the gap constant in absolute money. A buyer on a full conference registration pays 355 US dollars at the member early bird rate against 605 for a non-member, 505 against 755 in the pre-show window, and 655 against 905 on site. The gap is 250 at every step. The one day registration runs the same 250 gap: 255 against 505, then 405 against 655, then 555 against 805.

Holding the gap constant in money means it shrinks as a proportion. A non-member registering early pays 70.4 per cent more than a member, since 605 over 355. A non-member registering on site pays 38.2 per cent more, since 905 over 655. So the incentive to join is strongest for the person with the most time to act on it, which is defensible design: joining an association takes paperwork and the on-site registrant has none of that runway. How far apart the three date windows on that card should sit is a separate design question and A13's.

The same rate card treats a different population entirely differently. A non-exhibiting supplier pays 1,195 at the member early bird rate against 2,495 for a non-member, a gap of 1,300 and a premium of 108.8 per cent. That gap is not a membership incentive in the same sense. It is a barrier priced to make a supplier think hard about walking the floor without buying space.

CONEXPO-CON/AGG's 2026 policy uses a 200 dollar gap on the same category: a non-exhibiting vendor from a member company of AEM, NRMCA or NSSGA pays 599 against 799 for a non-member company. Note that this gap runs on corporate membership rather than individual membership, which changes the whole measurement, since the person registering may have no idea whether their employer is a member.

Counting the registrations the gap actually moved

Here is the join, and it needs one field most organisations already store: the date a membership record was created or renewed.

Take the closing registration file, 9,000 registrations, of which 5,400 paid the member rate. For each of those 5,400, look up the membership transaction date and flag any registration where a membership transaction falls within 30 days before the registration was created.

Suppose 400 registrations carry that flag. Those 400 are the whole population where the gap could plausibly have caused the membership.

Now split the 400 by what the person was before. Suppose 260 of them were members at the prior edition, so the transaction inside the window was a renewal that was going to happen anyway and merely happened to land near the registration date. The remaining 140 are first-time joins.

Out of 9,000 registrations and 5,400 member-rate badges, the gap produced at most 140 memberships, and even that is generous, since a person who joins an association in the same month they register for its show may have been joining for the journal, the certification or the local chapter.

The 260 matter as their own line. At a 200 gap they received 52,000 of discount for behaviour that was already established, which is the same shape of finding as the early bird cannibalisation measurement in A14 and worth reporting the same way.

What did each membership cost?

Put the whole discount against the 140.

All 5,400 member-rate registrations received a 200 discount against the non-member price, so the gap gave away 5,400 times 200, which is 1,080,000.

Against that, the 140 new joins produced membership revenue of 140 times 340, which is 47,600 in the first year.

Cost per membership generated is 1,080,000 divided by 140, which is 7,714. Nobody would sign off a membership acquisition programme at 7,714 a head.

That number is the argument for calling the gap a member benefit rather than an acquisition device. Under the benefit reading the 1,080,000 is not a cost at all: it is a service delivered to 5,400 dues-paying members, at 200 each, funded by dues they already paid. That is a coherent position and it is the one most associations are actually operating.

What it forbids is the September meeting. If the gap is a benefit, widening it to drive joins is a category error, and the honest way to grow membership is a membership offer rather than a registration price.

There is a second-year question worth adding to the report. Of the 140 who joined to register, how many renewed a year later without the show in front of them? If the answer is 38 per cent, the programme produced 53 durable members and the cost per durable member is over 20,000. If the answer is 70 per cent, the case is quite different. Most associations can compute this in an afternoon and very few have.

Turnout and repeat rate across the two rates

The gap should also be visible in behaviour after registration, and it is, though the reading is treacherous.

Compute turnout for the two populations separately, using badge collection or first entry scan for both, and holding the same rule for both populations in the way a paid and free comparison has to, which is A16's point. Members will almost certainly come out higher. Compute repeat rate as well, using a stable definition: CEIR's 2016 Attendee Retention Insights series found the most common industry definition of retention to be attendance at two of the last four editions, and found 77 per cent of trade show organisers tracking attendee retention at all. Use that definition and hold it steady between editions.

Members will very likely come out ahead on repeat rate too. What you cannot do is read either gap as caused by membership. A person who pays dues to a professional body has already declared that this field is central to their career, and that declaration predicts both attendance and repeat attendance on its own. Membership works here as a marker of that disposition, and no cut of the file separates the marker from the thing it marks.

The useful version of the comparison is narrower. Take the 140 who joined in order to register, and compare their turnout and their repeat rate against non-members who registered in the same window at the same rate card position. Those two groups are much closer in disposition, and the difference between them is a fairer estimate of what the membership relationship adds. The sample will be small and the estimate noisy. It is still the only version of the comparison that is about the membership rather than about the person.

Sizing the gap

If you are keeping the gap as a benefit, the size question becomes simpler: it should be large enough that members notice and small enough that non-members still register.

The failure mode is a gap that has crept up over years of raising the non-member rate to protect the member rate. At a 200 gap on a 495 member badge, the non-member is paying 40.4 per cent more. Push the non-member rate to 795 against the same 495 and they pay 60.6 per cent more, and the population you are pushing is your new-audience pipeline, since a first-time attendee is almost by definition a non-member.

An association show that widens its non-member gap is taxing exactly the registrations its growth depends on. That is the trade to argue about in September, and it is a different argument from the one the committee thought it was having.

One adjacent decision belongs in the same meeting. Board members, chapter officers and committee volunteers are usually carrying free badges rather than member-rate ones, and those sit under an entitlement with an owner and a cap if A17's document exists, or nowhere at all if it does not. Counting them as member-rate registrations overstates what the gap delivered.

Where this stops

The measurement rests on the membership transaction date, and that field is unreliable in ways that will not be obvious until you look. Renewals processed in batches, memberships backdated to a fiscal year start, corporate memberships where individual records are created by an administrator months after the company joined, and lapsed members reinstated with their original join date all corrupt the 30 day window. Before trusting the 400, plot the distribution of membership transaction dates across the year and look for the spikes that indicate batch processing.

Corporate membership breaks the method more fundamentally, as in the CONEXPO structure above. When eligibility flows from the employer, the individual registrant made no decision at all, the 30 day window is meaningless, and the gap cannot have moved anyone. For those shows the equivalent question is whether the corporate membership was sold on the registration saving, which is a sales conversation rather than a data one.

The 30 day window is also a choice. Widen it to 90 days and the flagged population grows and the causal claim weakens. Report the number at both windows so a reader can see how sensitive it is.

Join your last edition's registration file to the membership file on person, which is the sort of join any attendee analytics layer has to support before this question can be asked at all, and pull one column: the most recent membership transaction date. Count the member-rate registrations where that date falls within 30 days before the registration date, then split that count into first-time joins and renewals. Those two numbers are what your gap bought, and everything else the gap did was a benefit delivered to people who had already joined.