How to measure meetings and events spend

This is a method guide. Meetings spend management explains why the category is hard to classify; this page is the procedure for finding the money — the sequence of extracts, decisions and reconciliations that turns scattered transactions into one number you are willing to defend.

The exercise is mostly clerical. There is no analytical technique hiding here: there is a boundary decision, several data extracts and a reconciliation. And the output is not a precise figure. It is a defensible figure with a documented scope, which is a different and more useful thing.

It is the most valuable work a new programme can do, because everything downstream — policy thresholds, supplier consolidation, the business case — is calculated against it.

The instinct is to ask finance to run a report. It is reasonable and it does not work, for structural reasons rather than because anybody has done their job badly.

A general ledger is organised around how the organisation is managed, not around what activity the money funded. A training day appears under learning and development because that is whose budget paid for it, not because anyone decided it was not a meeting. The code answers a question about accountability; you are asking about activity.

Three things compound it. Suppliers serve several categories at once — a hotel group appears in individual travel claims and in departmental bookings with nothing distinguishing the two. A single event generates costs across several suppliers and months with no shared reference joining them. And a material share of the category never reaches the purchase ledger at all, because it was paid on a card or claimed on expenses.

This is the step people skip, and skipping it is why so many baselines collapse under challenge. If the scope is not written down before the analysis, every awkward finding becomes an argument about whether it should have been included — and that argument is unwinnable, because there was never an agreed rule.

There is no universally correct boundary, only a correct one for your organisation, arrived at by deciding and recording. Work through the following and write an answer to each.

Meetings expenditure is distributed across systems that were never designed to talk to each other. Expect to use at least four of the following.

A single method produces a number. Two methods produce a number and a measure of how much confidence it deserves.

Start from the total and work inwards. Take the full purchase ledger, sort suppliers by annual value, and work down identifying anything plausibly meetings-related — venues, hotels, caterers, AV and production companies, agencies, transport operators, training providers. Classify each supplier’s full year against your scope statement, then add the card and expense extracts.

The strength is coverage: nothing is missed because you did not know it existed. The weakness is precision — a hotel invoice might be a conference or four separate overnight stays, and the ledger cannot tell you which.

Start from the events and work outwards. Build a list of every meeting and event in the period from whatever exists — the events team’s records, room booking systems, the training calendar, marketing’s activity plan — then cost each one from contracts, invoices and the people who ran it.

The strength is accuracy and context: each figure attaches to a real event with a purpose, a department and a supplier. The weakness is coverage — you only find events somebody recorded, and the ones nobody mentioned are precisely the ones the programme exists to catch. It is slower, and it produces something reusable, because the event list becomes the first draft of your programme record.

The gap between them is the most informative output of the whole exercise, and it resolves into three categories. In top-down but not bottom-up is activity nobody centrally knows about — the invisible portion of the category, and direct evidence for a single route in. In bottom-up but not top-down usually means costs sitting in a source you have not extracted; chase each one. In both at different values is either partial capture or double counting, and each needs resolving individually rather than averaging.

You do not need the two to converge. You need to explain the difference. “£X from ledger analysis, £Y from event reconstruction, and here is what accounts for the gap” is considerably stronger than a single figure with no working.

Sometimes a source cannot be extracted in the time available. Sampling is a legitimate response, provided it is done openly: take a defined subset, analyse it completely, extrapolate, and say precisely what you did.

A sampled baseline with a clearly stated method is more credible than a complete-looking one assembled from whatever was easy to get. The objection is almost never to sampling — it is to undisclosed sampling.

Meetings is a long-lead category, which creates three legitimate answers to “what did we spend”. Actual is what was invoiced and paid in the period; it reconciles to the ledger and describes the past. Committed is contracted value — what you are obliged to pay whether or not money has moved. A conference contracted in November for the following June is a commitment from signature, and sits in no ledger until the first deposit. Forecast is expected expenditure on activity planned but not yet contracted: least precise, often most useful, because it is the only one still open to influence.

Lead with actual, because finance can verify it. Report committed alongside it. The gap between them is the most commercially interesting part of the picture and disappears entirely if you merge the two — event spend reporting covers presenting all three on an ongoing basis.

Every signed venue contract carries a contingent liability: the amount payable if the event does not happen. Most organisations have never aggregated it, and it is entirely calculable if you have the contracts. For each contracted but undelivered event, record the contracted value, the cancellation scale and the dates at which each step applies. The current point on that scale gives a percentage payable today; sum across the forward book for your exposure.

Do it as part of the baseline rather than as a separate project, because you are already handling the contracts. It produces a number genuinely new to most organisations, it needs no benchmark to be persuasive, and it moves the conversation from procurement to risk — often the conversation that gets a programme funded. The terms are explained on venue procurement, the programme view on meetings risk management.

VAT causes a disproportionate number of arguments, because different systems present it differently and nobody notices until two figures fail to match. The purchase ledger usually gives net values. Card statements and expense claims are typically gross. Agency invoices may separate fees from disbursements. Add these together without normalising and your total is neither net nor gross.

Pick one, state which, and convert everything to it. Net is usually more meaningful where VAT is recoverable, because it reflects cost to the business and matches how budgets are set. Gross is the figure people actually feel where it is not — parts of the public sector, and organisations with partial exemption. The choice matters less than the consistency.

Two traps. Overseas venues may carry non-recoverable foreign VAT, so a net figure understates real cost — flag those separately. And expense claims often include VAT never reclaimed because no valid receipt was obtained, which makes the “net” value theoretical. Ask finance which apply before you convert rather than afterwards.

A single total answers almost no useful question. Keep the structure shallow: three or four dimensions, each with a small fixed list of values, will support every question a programme needs to ask in its first two years. Deep taxonomies feel rigorous and get filled in inconsistently, which destroys the comparison you built them for.

Define each value in a sentence and keep the definitions with the baseline. Data quality collapses in year two not through carelessness but because “conference” meant something different the second time — the same discipline described on meetings data and reporting, applied retrospectively.

Your baseline will be challenged, and the challenge will rarely be analytical. It will come from someone whose department looks worse than expected, or who has been quoting a smaller figure for two years. Documentation is what lets you answer without redoing the work.

Write a short methodology note stating the period and basis, the scope, every source queried, the classification rules, everything estimated or sampled and how, the double-counting checks, the VAT basis, and the known gaps.

That last item matters more than it sounds. Volunteering what you could not see removes the most effective line of attack available to anyone who dislikes the conclusion. And use round numbers — precision the data does not support invites someone to find the item that is wrong and dismiss the rest.

A baseline is a measurement taken on a stated basis, so it decays. A full rebuild annually, on the same scope and basis, suits most organisations. Between rebuilds the programme record should be capturing activity as it happens, which replaces most of the reconstruction work above. The first baseline is archaeology; the second should be an extract.

Two rules make the refresh useful. Change the scope only deliberately, and when you do, restate the prior year on the new basis so the trend stays readable. And keep the old methodology notes, because “why is this year’s number higher” is usually answered by a scope change rather than by activity growth.

Expect the measured figure to rise in the first year or two of a programme. That is normally better capture rather than more spending, and saying so in advance is easier than explaining it afterwards — see meetings spend visibility and meetings and events KPIs.