A sponsor asks what last year’s event carbon reduction budget abated. Not what you did, what it abated, in tonnes. For most event teams the honest answer is that nobody measured the footprint by category before deciding where the money went.
That is where this work usually starts, and it is not a failure of effort. It is a failure of sequence. Without a measured breakdown of where emissions sit, the only ranking signal left is how visible a lever is.
So budget flows to whatever is easiest to photograph. Recyclable badges get funded. Attendee travel gets a paragraph in the report.
Stakeholders, sponsors and regulators increasingly expect measurable environmental data, not estimates and not pledges. An unranked reduction plan fails that test the same way a pledge does. It states intent, and it cannot demonstrate effect.
This piece sets out a three-axis ranking method: tonnes abated, cost per tonne, degree of control. It covers the category baseline that method depends on, and the published evidence on which levers survive it. Some of the most popular ones do not.
Reduction without a baseline is a pledge with a budget attached
The default behaviour is recognisable. Someone produces a list of things a sustainable event could do: reusable signage, local sourcing, LED rigs, waste diversion, a shuttle service, offsets for the remainder. Every item is treated as roughly equally worth doing, and the work starts wherever it is easiest to start.
That is entirely rational when you have no data. If you cannot say which category holds 40% of your footprint and which holds 2%, visibility is the only signal available. Visible levers tend to be small ones, because the things people see at an event (the badges, the cups, the signage) are rarely where the tonnes are.
The consequence lands at reporting time. You can describe activity in detail and still not demonstrate abatement, which is the position sponsors, clients and auditors have stopped accepting.
The fix is not a better lever list. It is a ranking, and ranking is a data problem before it is a strategy problem. The GHG Protocol Scope 3 Standard already tells reporters to prioritise emission sources using relevance criteria including size and influence. Almost nobody applies that instruction to an event footprint.
What measurement says about the levers teams reach for first
Start with the clearest published counterexample available. It should change how you read every lever list you are handed.
Researchers analysed delegate travel to the American Geophysical Union’s 2019 Fall Meeting: 28,000 attendees, 285 million kilometres travelled, roughly 80,000 tonnes CO2e, about 3 tonnes per attendee. Then they broke the total down by journey. Their finding on short-haul travel is worth quoting exactly:
“only 2% of the AGU 2019 emissions were caused by the 22% of delegates who took flights of less than 1,500 km one way […] Even if all 22% were to use trains, buses or carpools instead of aeroplanes, this would reduce total emissions by only 1%.”
Read that against the lever list. “Encourage rail instead of short-haul flights” appears on almost every sustainable event checklist in circulation. At this event, executed perfectly, for every eligible delegate, it would have moved the total by one percent.
The pattern holds even where rail is genuinely competitive. For a rail-accessible European meeting (the European Geosciences Union’s annual conference in Vienna), the same authors calculate that a full switch from plane to train cuts total travel emissions by at most 10%.
Now the other half of the same dataset. Emissions concentrate. At AGU 2019, 75% of travel emissions came from intercontinental flights over 8,000 km one way, flown by 36% of attendees. The 17% who travelled farthest accounted for 39% of the total on their own.
That shape has a strategic consequence. When a minority of activity carries most of the footprint, effort spread evenly across every category is mismatched to the problem. You are funding the 2% and the 39% at the same rate.
None of this makes short-haul modal shift wrong. It makes it small. Knowing that it is small is what lets you fund it from the right budget line and claim the right thing for it. What you cannot do is discover any of it without a category-level baseline.
Build the category baseline before you name a single lever
The events industry now has a common structure to measure against, which removes the “where do we even start” problem.
The Net Zero Carbon Events initiative is hosted by the Joint Meetings Industry Council and supported by the UNFCCC secretariat. It published its Net Zero Roadmap at COP27 in November 2022, then a Measurement Methodology (1st edition) in December 2023. The roadmap runs through five priority action areas, stated on the initiative’s own site as:
“Venue Energy / Food and Food Waste / Logistics / Smart Production and Waste Management / Travel and Accommodation”
Three transversal workstreams support them: Measurement, Carbon Offsetting and Reporting. The methodology adds emission source categories, measurement tiers, event-level metrics and baseline setting.
Those five areas are a serviceable spine for a first baseline. Size each one for your own event, in tonnes, using recognised emissions factors rather than a spreadsheet of assumptions.
The objection here is always the same: our first baseline will not be accurate enough to rank on. It does not need to be. Tiered measurement exists because data quality improves over cycles, and the decision you are making is coarse. You are separating a 30% category from a 2% category, not arguing about the third decimal place.
Waiting for perfect data is how teams spend a third consecutive year funding the badges.
Score every event carbon reduction lever on three axes
With categories sized, every candidate lever can be scored the same way.
Axis 1: tonnes abated
Absolute abatement against the measured baseline, in tonnes, not as a percentage of its own category.
This distinction does most of the work. A 40% cut inside a category holding 2% of your footprint is a 0.8% cut overall, and it will be written up as “we reduced X by 40%”. Both statements are true. Only one is material, and a sponsor’s analyst will notice which.
Scoring on absolute tonnes eliminates most tip-list levers immediately. That is the point. It is a filter, not a scoreboard.
Axis 2: cost per tonne, and the trap inside it
Cost per tonne is the standard ranking used in industrial decarbonisation, usually drawn as a marginal abatement cost curve: each lever gets a bar, the height is cost per tonne abated, the width is how much abatement is available at that cost.
It is a good tool with one failure mode. Ordering purely by cost-effectiveness favours cheap, small levers and defers expensive structural ones. For events, the expensive structural lever is usually the only one large enough to matter.
So sequence the two axes rather than merging them. Absolute abatement decides what is worth doing at all. Cost per tonne decides the order of what survives that filter.
Axis 3: degree of control
This is the axis missing from generic corporate carbon strategy content, and the one an event footprint needs most, because the largest source usually belongs to somebody else.
Sort levers into three bands:
- Levers you decide. Format, host city, event frequency, production specification, supplier selection. No negotiation required, only a decision.
- Levers you negotiate. Venue energy contracts, caterer sourcing, freight consolidation, exhibitor build standards. You set the requirement, a supplier delivers it.
- Levers you can only influence. Individual attendee travel choices, exhibitor behaviour on site, accommodation booked outside your block.
The banding tells you what kind of work each lever needs. A decision lever needs a decision meeting. A negotiation lever needs contract language and a supplier data request. An influence lever needs communication design and realistic expectations attached to it.
This is the standard rather than an invention: size and influence are both named relevance criteria in the GHG Protocol Scope 3 Standard. The three axes apply them to an event.
| Lever | Scores on | Band |
|---|---|---|
| Host city selection | Large tonnage, low direct cost | Decide |
| Event frequency (annual to biennial) | Large tonnage, high commercial cost | Decide |
| Virtual or hub participation for long-haul attendees | Large tonnage, moderate cost | Decide |
| Venue renewable energy contract | Moderate tonnage, contract-dependent cost | Negotiate |
| Freight consolidation and reusable build | Moderate tonnage, often cost-saving | Negotiate |
| Short-haul rail encouragement | Small tonnage, low cost | Influence |
Structure only. The tonnage column stays qualitative because the real numbers are yours to measure, and inventing plausible figures for an illustrative table is the habit this method exists to break.
What the ranking surfaces first: format, location and frequency
Run real levers through the three axes and the same cluster keeps coming out on top. All three are organiser decisions.
Location. Researchers modelled the American Psychiatric Association’s annual meeting by moving the same attendee population to different host cities. Modelled travel emissions varied roughly threefold.
| Host city | Total tCO2e | Per capita tCO2e | Difference vs actual |
|---|---|---|---|
| New York City (actual, 2018) | 19,819 | 1.19 | baseline |
| Chicago | 22,256 | 1.34 | +12% |
| Los Angeles | 31,761 | 1.91 | +60% |
| Honolulu | 52,415 | 3.15 | +164% |
For the 2019 meeting population, Washington DC, Philadelphia, Chicago or New York instead of San Francisco would have cut modelled emissions by 23%. The mechanism the authors identify is transferable: the variation tracked the share of attendees within driving distance of the venue.
Frequency. Moving a meeting from annual to biennial halves its annual travel emissions outright. It is the bluntest lever available, it costs nothing to execute, and it carries the largest commercial consequences. That is exactly why it belongs in a scored ranking rather than a wish list.
Format. The AGU researchers calculate that holding the meeting biennially in Chicago, with about a third of participants (the third responsible for most of the emissions) attending virtually, would cut travel-related emissions by more than 90%. A three-hub model, merging three regional conferences into one simultaneous event across Chicago, Tokyo and Paris, would cut their combined travel emissions by about 80%.
A separate study reaches the same order of magnitude by another route. A fully virtual conference cuts the carbon footprint by 94% and cumulative energy demand by 90%. A hybrid format with carefully selected hubs cuts about 67% while keeping more than half of participants in person.
Which brings up the objection this section exists to answer. The biggest lever is not ours to pull. You do not book your attendees’ flights, so how is their travel your reduction lever?
Because control over an individual journey is not control over the decisions that shape travel demand. Host city, frequency, format and hub structure are all yours. Every figure above comes from changing one of them, not from persuading a single delegate to do anything differently.
One honesty check before you bank any of it: virtual is not zero. That same life-cycle assessment counts the information and communication technology footprint of virtual participation as a real category, and finds that improving its energy efficiency reduces a virtual conference’s footprint further. Booking online participation as free abatement is overclaiming, which is the failure this whole method exists to prevent.
Where the ranking has to bend
Tonnage is not the only constraint, and a plan that pretends otherwise will not survive its first meeting with the commercial team.
The researchers behind both studies raise the limits themselves. The AGU authors note that a biennial, US-based meeting “would exclude many scientists based outside the United States from attending in person, potentially resulting in a two-tier conference system.” The APA authors are explicit that meetings matter for socialisation, networking and learning, and that they are a significant source of income.
Treat equity, revenue and attendance as ranking qualifiers, not footnotes. A lever that tops the tonnage column and destroys the commercial case is not a defensible recommendation. Neither is a plan that quietly drops it without saying so.
The practical move is documentation. Record the trade-off next to the score: what the lever would abate, what it would cost commercially, and the decision taken. A rejected lever with a recorded reason is a defensible position under scrutiny. A rejected lever that never appears in the report looks like it was never considered.
Offsets are the residual, not a lever
Keep offsetting outside the event carbon reduction ranking entirely.
The established sequence is avoid, reduce, substitute, then compensate what remains. Scoring offsets alongside real levers invites the exact substitution that sequence exists to prevent, because compensation will almost always look cheaper per tonne than structural change.
This is now a reporting risk rather than a philosophical preference. In the UK, the Digital Markets, Competition and Consumers Act 2024 came fully into force in April 2025, letting the Competition and Markets Authority enforce consumer law directly and fine misleading environmental claims up to 10% of global turnover.
In the EU, Directive 2024/825 on empowering consumers for the green transition entered into force in March 2024 and applies from September 2026. It restricts generic environmental claims, including claims of climate neutrality that rest on emissions offsetting. Compensation-led claims are precisely the ones drawing scrutiny.
The industry’s own structure reflects the same logic. Net Zero Carbon Events treats Carbon Offsetting as a transversal workstream alongside Measurement and Reporting, not as one of the five priority action areas where abatement happens.
Benchmark against yourself, then rebaseline
A single event’s numbers tell you where you are. Only comparison tells you whether a lever worked.
Three comparisons are worth building into the reporting cycle:
- The same event, year over year. The cleanest signal you will get, because the confounding variables are fewest.
- Comparable events inside your own portfolio. Useful for agencies and multi-event organisers, and the most realistic route to a defensible internal benchmark.
- Intensity per attendee day, not totals. Without it, a drop in headcount reads as abatement and a growth year reads as failure.
Two limits are worth stating plainly, because an event carbon reduction plan that overstates its own evidence base is not audit-ready.
First, there is no open, current, event-industry benchmark dataset with published category shares by event type. Internal benchmarking is therefore stronger than external benchmarking today, and any external figure you are handed deserves a source check before it reaches a stakeholder report.
Second, the published evidence here comes from large international conferences, where attendee travel dominates. A regional trade show, a local-audience festival or a heavily built exhibition looks different, with production, freight and venue energy weighing far more. The ranking method transfers. The category shares do not. Measure your own.
Then close the loop. Re-measure after the event, update the baseline, rank again. Each cycle should be built on better data than the last, which is the only version of this that improves rather than repeats.
What a defensible event carbon reduction plan looks like
Hold your current plan against four items:
- A category-level baseline, measured with recognised emissions factors, not assembled from assumptions.
- Every candidate lever scored on tonnes abated, cost per tonne and degree of control.
- Decisions recorded with their trade-offs, including the levers you rejected and why.
- A post-event re-measurement that updates the baseline for the next cycle.
That is the difference between telling a sponsor what you did and showing them what it abated. The first describes effort. The second is a number somebody can audit.
Producing it at event scale is what EventZero is built for. It measures across travel, accommodation, catering, production and logistics, with category-level carbon accounting and benchmarking designed for the event footprint rather than adapted from generic corporate carbon tools.
The levers were never the hard part
Every event team already has the lever list. What almost none of them have is the measured category baseline that turns that list into a ranked plan, with tonnes attached to each line and a recorded reason for the order.
Without it, event carbon reduction budget follows visibility, and the report at the end describes activity rather than abatement. With it, you can tell a sponsor which lever you funded, what it abated, what you rejected and why.
A reduction plan that cannot show which lever abated what is a claim about effort. Build the baseline first, rank second, spend third.
Book Your Carbon Strategy Session to build the category-level baseline your ranking depends on.
