The lead was in HubSpot before the visitor left the stand

ISE Barcelona, 2026. Our biggest show of the year. A visitor wraps up a conversation at the stand and walks off into the hall. Before they have gone far, their contact record is already in HubSpot. Event source attached. Brand tagged. Campaign attribution in place. Nobody has touched a spreadsheet.

That moment is the whole story. The rest of this piece is how we got there, and why I think marketing leaders should be building this layer of their event operation rather than renting it.

How trade show leads usually die

You know the standard setup. You rent the venue's lead retrieval scanners. Your team taps badges for three days. Sometime the following week, someone exports a CSV from the vendor portal and the cleaning begins.

By then the damage is done. Nobody remembers which conversation belonged to which brand. Follow-up goes out late and generic. Attribution becomes a guess, then a shrug. The leads technically exist. The value mostly doesn't.

This is not a tooling complaint. It is a structural one. The most expensive marketing activity of your year runs its data through infrastructure you do not own, on a timeline you do not control, in a format that strips out the one thing that made each lead valuable. The context.

The brief that made renting impossible

ISE Barcelona was our group's biggest event of the year. Four brands and a partner programme, represented on one stand.

Think about what that means for a badge tap. A venue scanner does not know which brand the conversation was about. It cannot route a contact to the right follow-up. It records that a human stood near another human for a moment. That is not a lead. That is proximity.

A pile of anonymous badge taps in a vendor portal was never going to carry four brands' worth of context out of that hall. So I stopped treating the vendor's system as a given.

What I built instead

A custom lead capture system, built by me, running live on the stand.

The flow is short. Scan the visitor's badge QR code. If a code is not an option, photograph the badge or a business card and let OCR pull the details. The contact then syncs straight into HubSpot in real time, with the event source, the brand context of the conversation and the campaign attribution already attached. No vendor portal in the middle. No export step. No waiting.

I kept the capture flow deliberately simple, because the people using it were busy working a stand, not testing software. A few seconds per lead, then back to the conversation. That constraint shaped the build more than any technical preference. Stand staff at hour seven of day two will not fight an interface, and they should not have to.

What changed during the show

Leads landed in the CRM the moment they were captured. So follow-up could begin while the show was still running, not a week after everyone had flown home. The brand tagging meant each contact routed to the right follow-up rather than a one-size-fits-nobody email. And attribution survived contact with reality. We knew exactly which contacts came from the event, because the system wrote it down at the moment of capture instead of reconstructing it afterwards.

The outcome I value most is quieter. Sales opened HubSpot during the show and found fresh, complete records waiting for them. No chasing, no handover meeting, no spreadsheet archaeology. That is what marketing infrastructure is for. It makes the next person's job start further along.

Build the layer, rent the venue

Here is the argument I actually want to make, beyond one show.

Events recur. The city changes, the stand changes, the capture problem does not. That makes the capture layer exactly the kind of thing that rewards building once and owning. Rent it and you pay every time, in fees and in lost context. Build it and every show after the first inherits the system. Ours cost nothing in vendor fees.

Ownership also changes whose rules apply. When the pipeline runs badge to CRM through your own build, your fields, your routing and your timing win. When it runs through a vendor portal, you get whatever the export gives you, whenever it chooses to give it.

And no, this does not mean every marketing leader must write code. I happen to. The transferable question is narrower. Who owns the path from conversation to CRM record at your events? If the honest answer is the venue's scanner vendor, you have outsourced the most valuable moment in your events calendar to a company whose incentive is the rental fee.

Questions worth asking before your next show

Where does a lead live in the hour after it is captured? If the answer is a third-party portal, decide now what context dies there.

Can follow-up start before the stand is packed down? If the process is built around an export, the answer is no, and everyone has quietly accepted it.

If several brands or product lines share the stand, how does a captured contact carry the conversation it came from? A name without context routes nowhere.

What does lead retrieval rental cost you across a year of shows, and what would a capture layer you keep cost against that?

None of these questions need a technical answer. They need a leader who treats events as infrastructure rather than theatre. The badge scanner was never the point. The point is that the most expensive three days of your marketing year deserve a pipeline you own.