You can't calculate conference ROI on contacts you never followed up.

The math most people do on the flight home

Somewhere over the midwest, you open your notes app and do the arithmetic. Ticket: $2,400. Flight + hotel + meals: $1,800. Three days away from the desk. You talked to maybe 60 people, got cards or scans from 35, had real conversations with 12. You divide $4,200 by 12 and call it $350 per meaningful contact. Reasonable, you think. The conference paid for itself if two of those become customers.

But that math treats every contact as a closed loop. It assumes the 12 conversations will be followed up, remembered correctly, and acted on with enough speed and context that the other person still connects you to the conversation you had. That assumption is doing most of the work in your calculation, and it's almost never examined.

The missing variable: execution rate

Standard conference ROI calculations treat follow-up as binary, you either send one or you don't. The real variable is execution rate: what fraction of contacts you intended to follow up with got a follow-up that was (a) sent within a useful window, (b) specific enough to recall the conversation, and (c) appropriate to where the relationship actually was.

Call that fraction E. Most people, if they're honest, put E somewhere between 0.3 and 0.6 for a high-volume event. Which means the actual addressable contact pool isn't 12, it's 4 to 7. And the real cost-per-actionable-contact is closer to $600 to $1,050, not $350.

This isn't a productivity lecture. It's a measurement problem. If E doesn't appear in your ROI calculation, you're not calculating ROI. You're calculating aspiration.

Why the standard framing breaks down

The dominant post-event ROI frame focuses on cost inputs: registration, travel, time-away. Some more sophisticated versions add pipeline contribution, deals that close and trace back to conference contacts. Both are legitimate, but both share a flaw: they treat the contact list as a fixed asset the moment the event ends.

It isn't. The contact list is a decaying asset. The decay is fast in the first 72 hours and persistent after that. Context, which table, which conversation, what they said they were working on, goes first. Then the name-to-face mapping. Then even the basic category of relationship (prospect vs. peer vs. press vs. vendor) starts blurring when you're staring at a list of 35 names five days later with no notes.

The ROI calculation most people do on the flight home is measuring the theoretical value of the contacts they made. The actual ROI is determined by what they do with those contacts in the week that follows. Those are different numbers, and conflating them is why people keep attending conferences they probably shouldn't and skipping ones they probably should.

What a real ROI calculation looks like

A more honest post-event calculation has four inputs and one output.

Inputs: total event cost (ticket + travel + lodging + time, valued at your day rate), number of contacts made, your execution rate E (be honest), and your average pipeline contribution per converted contact based on prior events.

Output: expected pipeline contribution, adjusted for E.

The formula: (Contacts × E × avg. pipeline contribution per contact) minus total event cost = adjusted ROI.

If you don't have a prior-event pipeline contribution figure yet, use a conservative estimate from your deal history, the first-contact-to-closed-won rate for relationships that started at a live event, multiplied by your average deal size. That number exists somewhere in your CRM if you've been doing this for more than a year.

Running this calculation before the event, using projected contacts and a realistic E, is also how you decide whether to go. Most people skip this step entirely and make the decision on gut feel and the conference's own marketing materials.

A scenario: the 80-card problem

You leave a three-day industry conference with 80 business cards and scans. You had real conversations, the kind where you remember the specifics, with maybe 20 of them. The other 60 are people you chatted with at a booth, at dinner, at the badge pickup line.

Day one back in the office, you send follow-ups to 9 people. You meant to send 15 but ran into a backlog. Day three, you do 4 more. Day five, you look at the remaining stack and genuinely can't remember which Sarah K. was the procurement lead and which one was the conference organizer's assistant. You don't send those.

Your execution rate for this event: 13 of 20 targeted contacts, so E = 0.65. Better than average, actually. Now apply the formula: if your average first-meeting-to-pipeline contribution is $8,000 and your first-contact-to-qualified rate is 20%, each followed-up contact is worth an expected $1,600 in pipeline. Thirteen contacts times $1,600 = $20,800 in expected pipeline against a $5,000 event cost. That's a clear positive.

But the seven you didn't follow up? Those weren't free. You paid $350 each for them. The real cost of not building a follow-up system isn't missed productivity, it's concrete expected pipeline that never entered the funnel.

What changes if you actually track this

If you run this calculation consistently across events, a few things shift.

First, your conference selection improves. You stop going to the well-attended events that produce mostly low-context contacts and start weighting toward the smaller, denser events where your E rate is historically higher, because conversations were longer and follow-ups were easier to write.

Second, you start measuring the right thing during the event, not just after it. The question isn't how many cards you collected. It's whether you captured enough context per contact to actually execute the follow-up. A stack of cards with no notes is a lottery ticket. A smaller stack with real context is a workable pipeline.

Third, you get a number you can actually use to defend or kill a conference in your planning cycle. 'I always go to X' is not a business argument. A consistent negative adjusted ROI over three years of X is.

The capture problem sitting underneath the math

Every part of the ROI calculation above depends on execution rate, and execution rate depends on context capture. You can't write a specific follow-up to someone you can't remember. And in a two-day conference with 80 contacts, memory alone isn't a system.

The apps built for this space mostly solve for scanning speed. That's a feature, not the product. The product is whether you can walk back into the office on Monday with enough context on each contact to actually execute, who they were, what they said, what the obvious next step is. Scanning a card in three seconds doesn't give you that. A quick voice note, a context tag, a follow-up draft while the conversation is still warm, that's what closes the gap between the contact list you have and the follow-up execution rate you need.

Met's Event Mode is built around this gap: capture the contact, add context in the moment, queue the follow-up before you walk to the next booth. The scan takes seconds. The context layer is what makes the ROI math work later.

What to track before you go next time

One concrete step before your next event: pull the last conference you attended and run the adjusted ROI calculation backward. Estimate your E, estimate your pipeline contribution per contact, compute the adjusted number. If you can't estimate E because you didn't track it, that's the finding, and the thing to fix before the next one.

Bring a number to the 'should we go' decision, not a vibe. The events that look expensive on the way in are often the ones that pay out. The ones that feel like a win because of the energy in the room are sometimes the ones where your E rate collapses the moment you're back at your desk.