Events are usually one of the biggest lines in a B2B marketing budget, and the hardest to defend. The costs are painfully visible: stand, travel, tickets, a week of the team's time. The returns are scattered across months of conversations, follow-ups and deals that close long after everyone has forgotten which conference started them.
So events live under permanent suspicion, and the suspicion is usually a measurement problem rather than a performance problem. This guide is a practical way to measure what a conference actually produces.
Why Events Look Bad on Dashboards
Everything valuable at an event happens offline. A fifteen-minute conversation with the right person can be worth more than a quarter of ad spend, and it produces exactly zero recorded touchpoints. When that person visits your website a fortnight later from a Google search, your attribution report credits the search.
The result is a systematic bias. Click-based reporting makes events look like a cost centre and makes the search ad that harvested the event's demand look like a star. It is the same trap we described in why your attribution dashboard is lying to you, at its most expensive.
The value of an event arrives in the weeks and months after it ends: the follow-up meeting, the introduction, the buyer who remembered your stand when the budget appeared. Measure the ripple, not just the week.
Measure the Whole Timeline
The practical fix is to decide what you will measure before you go, then keep measuring for longer than feels natural afterwards.
Two details make this work in practice. First, tag every account you spoke to in your CRM with the event name, so months later you can answer "what did that conference produce?" with a list instead of a shrug. Second, make the survey question part of the system. When a new lead writes "saw you at a conference in the spring", that is the event follow-through your click reporting will never show. The mechanics of asking well are in our HDYHAU survey guide.
Judging an Event Honestly
Here is a deliberately simple worked example, with invented numbers, to show the shape of an honest judgement. Suppose a conference costs twenty thousand in total. Over the following six months, the accounts your team actually spoke to produce a handful of opportunities and one closed deal worth forty thousand, and three later survey answers name the event. On the week-one dashboard, the conference looked like a dead loss. On the six-month view, it roughly doubled its money and seeded pipeline that is still moving.
The numbers are made up, but the lesson is not: the verdict on an event depends almost entirely on when you pass it and what evidence you allow. Decide both before you book the stand. Compare like with like, too. An event is a demand creation activity, and it should be judged against other creation work over creation timescales, not against a retargeting campaign over a fortnight. That distinction is the subject of demand creation versus demand capture.
Decide the Scorecard First
Most event ROI arguments are really arguments about evidence, held after the money is spent. Agree the scorecard before you go: the accounts, the conversations, the follow-up window, the survey signal.
Then let the event be judged on the game it is actually playing.
Frequently Asked Questions
How do you calculate the ROI of a trade show?
Add up the full cost, including tickets, stand, travel and the time of everyone attending. Then track the pipeline and revenue from the accounts you actually spoke to, over months rather than weeks. Add the slower evidence: survey answers naming the event and any lift in branded search afterwards. A single week rarely pays back within the same quarter, so judge events on the horizon they actually work on.
Why does event marketing look bad in attribution reports?
Because the influence happens offline. A good conversation at a stand produces no click. When that person visits your site two weeks later from a search, the attribution report credits the search. Events end up looking expensive and unproductive in click-based reports precisely because their best work leaves no click trail.
Are badge scans a good measure of event success?
They measure attendance, not interest. A scan tells you someone walked past your stand and let you scan them. The better unit of measurement is a real conversation: someone who discussed their situation with your team and agreed a next step. Fifty genuine conversations usually beat five hundred scans.
How long after an event should you measure results?
Longer than feels natural. B2B buying cycles run months, and an event conversation often resurfaces when the budget or the trigger arrives. Keep the event tagged against accounts in your CRM and keep reading your survey answers, because "we met you at a conference" can arrive a year later. Closing the books after four weeks writes off most of what the event will produce.
James Kevan is the co-founder of First Signals, which catches the long tail of event influence in survey answers, months after the badge scanners have been packed away.
Related guides: Measure Podcast Advertising ROI · How to Measure Word of Mouth · HDYHAU Surveys for B2B
© 2026 James Kevan / firstsignals.ai. Share freely with attribution.