A CMO I work with spent €62,000 on a booth last spring. Ten square metres, two days, four people flown in, a coffee machine that got more attention than the product. She came back with 380 badge scans and a genuinely good story about a conversation with a buyer from a company they had been trying to reach for two years.
I asked her three weeks later what had happened to the 380.
The scans were in a CSV. The CSV was in the download folder of the laptop belonging to the events manager, who had gone on holiday the day after the show. Nobody had opened it. The buyer from the two-year target account had received nothing. Not one email.
This is not an unusual story. It is closer to the default. And the reason it keeps happening has almost nothing to do with anyone being lazy.
The 80% number is real, and it is not a motivation problem
The Center for Exhibition Industry Research is the source most people quote here: roughly 80% of trade show leads never get any follow-up contact at all. Other 2026 event surveys land in similar territory, with a large majority of marketers admitting their company does not convert event leads into real opportunities.
Every time that stat gets shared on LinkedIn, the comments fill up with the same diagnosis. Sales is lazy. Reps do not respect marketing's work. Somebody needs to be held accountable.
I think that reading is wrong, and it is wrong in a way that keeps the problem alive.
Go and trace the actual path a badge scan takes at most companies. The scanner app belongs to the show organiser, not to you. Its export lands as a CSV with the organiser's column names, not yours. Someone has to open it, clean it, decide which records are worth importing, map the fields, dedupe against existing contacts, assign owners, and set a campaign source. That is between four and nine hours of unglamorous work owned by whichever person is least likely to have a free afternoon in the week after a trade show, because they were at the trade show.
So the leads sit. Then they sit some more. By the time anyone opens the file, the buyer has forgotten the booth, and the rep who gets handed a fourteen-day-old list of names with no context reasonably concludes it is junk and works their own pipeline instead. Next quarter, the same thing happens, and now the rep has a prior.
Share of trade show leads that receive no follow-up contact at all, according to CEIR. Not because sales refuses. Because the data never reaches them in a usable state while the conversation is still warm.
The fix is not a motivational email to the sales team. It is plumbing. Boring, specific plumbing that you build before the booth ships.
A badge scan is not a lead
Before any of the routing matters, you have to be honest about what you actually collected.
At a decent B2B show, a booth that scans 380 badges has collected something like this: forty people who wanted the tote bag, thirty students and job seekers, twenty competitors doing recon, a hundred and fifty who stopped because the demo screen caught their eye and have no budget or authority, ninety who fit your ICP and had a real but shallow conversation, and maybe twenty-five who described a problem you solve, in their own words, with a timeline attached.
Those groups need completely different treatment. Most teams give them identical treatment, which is the second reason follow-up fails even when it happens.
If you dump all 380 into a "post-event nurture" sequence, three things break at once. Your CRM data quality degrades because you just added a few hundred junk contacts to the database. Your deliverability takes a hit because a good chunk of those addresses were typed into a scanner by someone standing up, holding a coffee, at 4pm on day two. And your reps stop trusting anything tagged with the event, because the first five they called were students.
The tiering has to happen at the booth, in the moment, by the person who had the conversation. Not afterwards from a spreadsheet, where all you have is a job title.
What tiering at the booth actually looks like
Give every person working the stand three buttons in the capture app, and drill them on it in the pre-show briefing so it takes two seconds:
Tier A means the person described a specific problem you solve and there is a reason to talk in the next thirty days. The rep who scanned them adds one line of context in the notes field. One line. What they said, in their words.
Tier B means they fit the ICP and the conversation was real but there is no timeline. Also worth a note, shorter.
Tier C is everything else. Scan it, tag it, and let it go to marketing's general list or nowhere at all.
That single field is the highest-value data point you collect at the entire show, and it costs nothing. It also creates a small useful pressure: a rep who knows they will personally own the Tier A follow-up gets more careful about what they mark Tier A.
The window is 48 hours, and it closes hard
The published numbers on follow-up speed are directionally consistent even when the exact figures differ. Teams that make contact within 24 hours of the conversation convert several times better than teams that wait a week. About 40% of exhibitors who do follow up wait three to five days, which in practice means the buyer has been to eleven more booths and cannot picture your logo.
This is the same physics as speed to lead on inbound forms, except worse, because the decay is tied to a memory rather than an intent signal. An inbound demo request from nine days ago is still a person who wanted a demo. A badge scan from nine days ago is a person who cannot remember which of the forty booths you were.
So the target is simple. Every Tier A lead has a real, personal message before the show ends or on the morning after. Not a batch send. Not next Tuesday.
That is only achievable if the system is built beforehand, which brings us to the actual build.
Building the plumbing
None of this needs a big platform purchase. I have built it twice this year with tools most teams already pay for.
Capture
Use a capture app that syncs to your CRM natively rather than the show organiser's lead retrieval unit, which is usually rented, usually expensive, and always exports a CSV. Popl, Blinq and Cvent all do native scanning with real-time sync to HubSpot and Salesforce. Momencio does the same. Any of them beats a CSV.
Whatever you pick, test the sync with three fake records a week before the show. Not the day before. Field mapping problems always surface at the worst possible moment, and at 9am on day one nobody is going to debug a custom property.
Enrichment
The email address someone types into a scanner while standing up is unreliable. Run every record through waterfall enrichment overnight so a verified work email, company size, funding stage and headcount are on the record before anyone opens it in the morning. Clay handles this well, and we have written about how to build a waterfall enrichment sequence that does not burn credits on records you will never contact. Enrich Tier A and B. Skip Tier C entirely.
Enrichment also gives you the disqualification pass for free. Anyone at a company under your minimum size, anyone with a competitor domain, anyone whose title contains "student", drops out before a rep ever sees them.
Routing
The rep who had the conversation owns the follow-up. This sounds obvious and almost nobody does it, because the standard routing rules assign by territory or round robin and quietly overwrite the one thing that made the lead warm.
Add an exception: if a scanned_by field is populated, that person is the owner, regardless of territory. Create the task with a twelve-hour due date. If the account already has an owner and a different rep scanned the badge, the task goes to the owner with the scanner's note attached and the scanner cc'd. That handles the awkward case without a meeting about it.
We usually run the whole chain in n8n rather than native CRM workflows, because the enrichment step, the disqualification logic and the task creation need to happen in sequence with a wait in the middle. It is also easier to keep the data inside EU infrastructure that way, which matters for most of the clients we work with. There is more detail in our writeup on n8n for RevOps automation.
Build the follow-up system two weeks before the show, not two days after it.
Every hour of setup you skip before the event costs you roughly a day of delay afterwards, and the delay is what kills the conversion rate. The booth is the expensive part. The plumbing is the cheap part that decides whether the booth was worth it.
What to actually send
Tier A gets a plain email from the rep who spoke to them, sent from a normal mailbox, with no template markers and no graphics. It references the specific thing the person said. That is the whole trick, and it is why the one-line note field matters more than any other field you collect.
Something like: "You mentioned your team is running three separate quoting spreadsheets and finance re-keys everything at month end. That was the thing I kept thinking about after. Worth 20 minutes next week to show you how two other companies fixed it?"
No deck. No "it was great connecting at booth H24". They met forty people. Your booth number means nothing to them; the spreadsheet problem is what they will recognise.
Tier B gets a short sequence, three or four touches over two weeks, tied to the theme of the show rather than to your product. If the whole conference was about AI agents in support, the sequence talks about that. Our follow-up sequence guide covers the structure in more depth.
Tier C gets added to the newsletter, if they opted in, and otherwise gets nothing. Resisting the urge to email Tier C is a real discipline and it protects the domain you need for Tier A.
One more thing that consistently outperforms email: if the show gave you a strong Tier A conversation and the account matters, have someone else from your company reach out too. A second thread from a founder or a technical lead lands differently from a fourth email in the same chain. This is standard multithreading practice and events are the easiest place to start one, because you have a legitimate reason to be in touch.
Your attribution window is throwing away most of the value
Here is where the finance conversation usually goes wrong.
Events take a large share of B2B marketing budget. Gartner's 2026 cross-industry figures put events and field marketing at around 16% of the B2B marketing budget, and CEIR reports that exhibiting alone accounts for a much larger share of exhibitor marketing spend in some segments. Published cost per lead for trade shows ranges from about $112 to $811 depending on the source and the definition, which tells you mostly that everyone counts differently.
Then the CFO asks what the show produced, and marketing runs a report with the default 30-day attribution window. In a business with a six to nine month sales cycle, a 30-day window is measuring the wrong thing entirely. It captures the meetings booked in the fortnight after the show and misses every deal that closed in Q1 with a first touch from a June conference.
Set the event attribution window to 180 days. Track event influence as a flag on the opportunity rather than as a single-touch source, so a deal can be marked as event-influenced without event marketing having to claim all of the credit. Then report two numbers side by side: pipeline sourced by the event, and pipeline influenced by it. The influenced number is usually three to five times the sourced number and it is the honest one.
The field marketing benchmark people quote in 2026 is a 5x to 10x pipeline-to-cost ratio within the year. That is achievable, but only if the window is long enough to see it. If your model is single-touch and 30 days, you will conclude events do not work, cut the budget, and be wrong. We go deeper on why this happens in revenue attribution.
The four numbers worth reporting
Skip the scan count. It measures foot traffic and stand position, not much else.
Report the Tier A count, which measures whether your team had real conversations. Report the contact rate within 48 hours, which measures whether the plumbing works. Report meetings held from the event, which is the first honest signal. And report event-influenced pipeline at 180 days, which is the one the board actually needs.
If the Tier A count is low, the problem is who you sent and how they were briefed. If Tier A is healthy but the 48-hour contact rate is low, the problem is the system described above. Those two failure modes get confused constantly and they have completely different fixes.
Where this breaks in practice
Three things go wrong even when the design is right.
The first is that nobody owns the event as a revenue project. Marketing owns the booth, sales owns the pipeline, and the handover in between belongs to no one. Name a single owner for the follow-up, in writing, before the show.
The second is that the pre-show meeting booking gets skipped. The highest-return activity around any conference is reaching out to target accounts two to three weeks beforehand to book coffee. Walk-up scans are lottery tickets. Booked meetings are pipeline. A team that books fifteen meetings in advance will beat a team that scans 500 badges, every time, and it costs nothing extra.
The third is that the CRM has no place to put any of this. No campaign object, no event field on the opportunity, no scanned_by property. So the data arrives and immediately becomes unqueryable. Twenty minutes of schema work solves it permanently, and it is the same twenty minutes that most teams keep postponing.
Spending real money on events and not sure what came back?
We build the capture, enrichment and routing layer before your next show, and the attribution model that tells you whether the last three were worth it.
Book a 30-minute audit →FAQ
How fast should we follow up on trade show leads?
Tier A leads should get a personal message within 24 hours, ideally before the show closes. Everything else within 48 hours. The decay is steep because badge scans depend on the buyer remembering a conversation, and after three or four days at a busy conference that memory is gone. Waiting until the following week, which around 40% of exhibitors do, is the single most common reason event leads never convert.
Are trade show leads worth the cost per lead?
Judged on cost per lead alone, usually not. Published figures put trade show CPL anywhere from about $112 to over $800, which is far above most digital channels. The case for events rests on conversation quality and on multi-touch influence over a long window, not on volume. Measured with a 30-day single-touch model, almost every trade show looks like a failure. Measured on 180-day influenced pipeline, good shows clear a 5x to 10x ratio against cost.
Should we import every badge scan into the CRM?
No. Import Tier A and Tier B, and hold Tier C somewhere else or discard it. Bulk-importing every scan adds junk records, damages reporting accuracy, and trains your reps to distrust anything tagged with an event. A cleaner database with 90 real contacts beats a polluted one with 380.
What is the best lead capture app for trade shows?
Any app with native real-time sync to your CRM beats the organiser's rented lead retrieval unit, which typically exports a CSV. Popl, Blinq, Cvent and Momencio all sync directly to HubSpot and Salesforce. The choice matters less than testing the field mapping a full week before the show with a few dummy records.
How do we prove event ROI to a CFO?
Show two numbers with a 180-day window: pipeline sourced by the event and pipeline influenced by it. Sourced means the event was the first touch on the account. Influenced means an event touch appears anywhere in the buying cycle. Present both, state the window, and state the cost including travel and staff time. CFOs push back on event spend mostly because the numbers they have been shown were measured over 30 days and looked terrible for reasons that had nothing to do with the event.
Getting this built
The uncomfortable part of all this is that the expensive decision has already been made by the time anyone thinks about follow-up. The booth is booked, the flights are paid, the stand is being built. What is left is a few hours of configuration that determines whether any of that money comes back.
We do this work as part of CRM and RevOps builds and go-to-market systems, usually with the automation layer sitting between the capture app and the CRM. If you have a show coming up in the next quarter and no plan for what happens to the scans, get in touch and we will map it out.