Forward bookings as a metric only works if the comparison date is fixed

The slide says forward bookings are up 13 per cent. It is the second bullet in the trading section of a board pack, and it is the number the chair asks about, because it is the only figure on the page that says anything about next year.

The question that follows is always the same. Up 13 per cent against what.

Forward bookings as a metric is one of the few genuinely useful early signals an event business has, and it is also the easiest one to break by accident. Nobody has to falsify anything. Two people pull the same query six days apart, both are honest, and the answers differ by four percentage points.

What does forward bookings actually count?

Start with the thing nobody writes down. There are at least four defensible bases for a forward booking figure, and an organiser typically uses two of them in the same building.

A signed contract is one basis. An invoice raised is another, and it will be smaller, because contracts sit unbilled until the deposit schedule catches up. Cash received is smaller again. A sales pipeline weighted by probability is larger than all three and is a different kind of object entirely, because it contains no commitment from anyone.

None of these is wrong. What breaks the metric is mixing them, which happens most often when the sales system and the finance system disagree about what counts as won and the report pulls from whichever one answers faster.

The standards give no help here. Forward bookings is a voluntary measure with an issuer defined basis, and it lives outside the accounts because the revenue has not been earned. The nearest audited anchor is the contract liability balance, and that only covers the part already invoiced or collected, which is the part you already knew about.

Public reporting shows the same gap

Informa PLC's 2026 half year results, published on 30 July 2026, told the market that "Group revenue of $4.5bn+ committed or visible through subscriptions, forward bookings and contracts, representing around 80% of 2026 targeted revenues and pacing ahead of last year on a like-for-like basis".

That is a serious disclosure with a serious qualifier attached. The phrase doing the work is "pacing ahead of last year on a like-for-like basis", because it concedes that the comparison needs a stated basis before the pacing claim means anything.

Worth reading alongside it: the glossary of alternative performance measures at the back of that same statement, running from page 55, defines adjusted results, adjusted operating margin, free cash flow, net debt, underlying revenue and underlying adjusted operating profit. Forward bookings appears once, in the highlights on page one, and is not among the defined terms.

The European Securities and Markets Authority's guidelines on alternative performance measures, published on 5 October 2015 as ESMA/2015/1415en, are direct about that gap. Paragraph 21 says issuers "should disclose the definitions of all APMs used, in a clear and readable way". Paragraph 41 says "The definition and calculation of an APM should be consistent over time". Paragraph 37 requires comparatives for the corresponding previous periods.

I read those three together as the whole specification for an internal forward bookings report, whether or not anything is going to a market. Define it, keep the definition still, and show the comparative on the same footing. The bridging work that the same guidelines demand of measures which do derive from the accounts, such as adjusted operating profit, is a separate discipline and sits with the reconciliation an alternative performance measure needs.

Why does the comparison date matter so much?

Because event sales are not spread evenly through the year, and the last third of the selling window carries a disproportionate share of the value.

Take a show that opens on 14 September this year. Last year it opened on 8 September. Both editions sell the same way, with a slow start, a step up after the floorplan release, and a rush in the final quarter of the window.

Strike the comparison at a fixed 90 days before doors open. This year that is 16 June, and contracted space revenue stands at 6.10 million. Last year, 90 days out was 10 June, and the figure was 5.40 million. Growth is 6.10 divided by 5.40, which is 1.1296, so 13.0 per cent.

Now strike it on the calendar date instead. Pull both years as at 16 June. This year is unchanged at 6.10 million. Last year, 16 June was 84 days out, and by then bookings had reached 5.62 million. Growth is 6.10 divided by 5.62, which is 1.0854, so 8.5 per cent.

Same two editions, same source data, same query, and the answer moves by 4.5 percentage points depending on which of two reasonable dates somebody chose.

The mechanism is visible in the middle of that arithmetic. Between day 90 and day 84 last year, contracted value went from 5.40 million to 5.62 million, which is 220,000 across six days, roughly 36,700 a day. Handing the prior year six extra selling days hands it 220,000 of free growth, and the flattering direction is whichever one the person building the slide did not intend.

Which date should you fix?

Days to open, every time, and I would go further than that.

Fix a small number of gates rather than a continuous curve, because a continuous curve invites the reader to pick a favourable point on it. Something like 180, 120, 90, 60 and 30 days before doors open covers the useful shape of the window without giving anyone a choice.

Report the gate, the two absolute values and the percentage, in that order, so a reader who wants to check the division can. On a show that moves within its month every year, this is the only comparison that holds. On a show that has moved between seasons, no gate saves you, and the honest report says so and switches to a full edition to edition comparison after the fact.

There is a second date to fix, which teams miss more often than the first. The as-at date of the data extract needs to match the gate. A pipeline pulled on Monday morning for a gate that fell on Saturday has two days of extra bookings in it, which on the arithmetic above is another 73,000 of drift.

What the definition needs to say

Five things, and it fits on half a page.

The basis. Signed contract, invoice raised, or cash received. Pick one for the headline and show the others as supporting lines if people want them.

The scope. Space, sponsorship, delegate fees, ancillary services. A forward bookings figure that quietly includes stand build revenue in one year and not the other is a portfolio mix problem wearing a growth costume.

Cancellation treatment. Whether a contract cancelled after the gate is stripped out of the historical figure retrospectively, or left in. Both are defensible. Only one can be true in a given report, and restating history without saying so is how a metric loses its audience.

The gate. Days to open, listed explicitly.

Currency. Whether prior year values are restated at current rates. For a portfolio selling in several currencies this decision is worth more than most of the others combined, and the restatement mechanics belong with constant currency growth rather than here.

Where this stops

Forward bookings tells you about contracted value at a point in the window. It says nothing about what happens after the gate, and the second half of the window is where the interesting variance lives.

A show can be 13 per cent ahead at 90 days and finish flat, because the prior year had an unusually strong final six weeks, or because this year pulled bookings forward with an early payment discount that borrowed from a period it will now be compared against next year. The metric is a lead indicator with a known blind spot, and the blind spot grows the earlier the gate.

It also flatters a portfolio that has added editions. A group figure that rises because there are now twelve shows instead of eleven is telling you about the portfolio's shape, and separating that out is the work of holding a fixed set of editions across both periods. Bookings for contracted future work also have a formal cousin in the accounts, and the transaction price allocated to unsatisfied performance obligations is the disclosure that governs it once contracts run beyond a year.

The last limit is the one that matters most in practice. A forward bookings series is only as good as the oldest edition in it. If your sales system was replaced two years ago and the historical extract is a partial migration, your three year trend has a step in it that has nothing to do with the shows, and you will not find it by looking at the percentages.

The first step this week

Pull your largest show's contracted space value at exactly 90 days before doors open for the last three editions, using the door date of each edition rather than a calendar date, and write the three numbers on one line with the three door dates next to them. Then pull the same three at the calendar date the current year's gate falls on. If those two sets of growth rates differ by more than a point, your reporting has been quietly choosing a comparison for you, and fixing the gate is a one afternoon change to a saved query. The wider finance reporting stack can wait until that one line is right.