In B2B consulting, most companies measure event success by how full the room looks. We did too, until the room stayed full and the pipeline stayed flat. That's when we found the KPI quietly ruining almost every workshop we ran: attendance. We replaced it with a single rule about who's allowed in the room, and it turned four workshops a year into our best-performing lead channel.
Why most B2B companies get events wrong
Most B2B companies treat events as a sponsorship line item. You pay for a booth or a slot on someone else's agenda, your team collects badge scans all day, and by Friday those scans get relabelled as leads in a spreadsheet somewhere. Nobody asks whether a single real conversation happened in that room. The only people who do ask a hard question sit at the top, and it's the opposite extreme: how many new deals did this close? That's just as unhelpful as counting badge scans, because deals take months to close, and blaming a single event for not producing one by next quarter misses the point in the other direction.
What a badge scan actually is
A badge scan means someone walked past your booth with an ID that has a barcode on it. That's the entire data point. We did have a system for it, of course, we logged every scan and tagged the leads. But out of roughly 400 people scanned at a typical event, maybe 20 turned into an actual conversation. The rest got a follow-up email and we waited for a reply. Almost nobody ever wrote back.
There's a second problem specific to consulting: at a Microsoft Fabric or Google event, how many of the people in that room are the actual decision makers you need to be talking to? In our experience, they're either not there at all, or they show up for the opening keynote and disappear before anything substantive happens.
The KPI problem: attendance instead of composition
The metric almost every B2B marketing team defaults to is attendance: how many people registered, how many showed up, how full the room looked in the photo. It's not that most B2B teams let anyone through the door, plenty of them do put thought into approving who's on the list. The real gap sits around the room, not on it: almost nobody runs a proper prebriefing before the event or a structured follow-up after it. We fell into exactly that trap at Adastra Czechia before we changed what we measured, and the room always looked good. The pipeline didn't.
There's a distinction worth naming here: some KPIs just count what already happened, and some KPIs decide what your team does next. Attendance only ever does the first. You can hit that number by inviting whoever will say yes fastest, and it tells you nothing until the quarter is already over. Composition does the second, because the person building the guest list has to make a different call on every single invite to hit it. That's the whole reason one metric changes behaviour and the other one never does.
The insight behind the 30% rule
What we were optimising for before
For a long time, our workshop KPI was seat count. Did we get to 80 people? Did the room feel full? We never stopped tracking total audience, that number is still on the dashboard today. What changed is which number outranks the other: composition became goal number one, and total headcount followed behind it, not the other way round. Before that shift, every planning conversation revolved around driving the seat count up, so the team invited whoever was fastest to say yes - mostly people we already worked with, because they're the easiest audience to fill a room with.
The moment we realised the metric was producing the wrong behaviour
The room kept filling up and the pipeline stayed flat. That's usually the moment you're supposed to notice something's off, and eventually we did: we were measuring whether we could fill a room, not whether the room contained anyone worth talking to. Account managers and salespeople will always invite their own clients, that's natural and it hasn't changed. What we added on top is a quota that belongs to marketing alone: actively hunting down the 30% of the room that has to be new faces. A full workshop of existing clients makes for a nice afternoon. On its own, it isn't demand generation.
Why composition matters more than attendance in consulting
In consulting specifically, the value of an event isn't just the headcount, it's who's in the room and what they're hearing. That said, I won't overstate it: workshops full of our own clients still produce real upsell and cross-sell opportunities, we see that pipeline directly, tied back to the event. But new-logo pipeline is a different problem, and that's where non-client attendance is what actually predicts whether a workshop opens a door that wasn't open before. For new business, attendance alone was never going to get us there.
How the 30% rule works in practice
The rule
Thirty percent of every registered list has to be non-clients. In practice, everyone who fits our ICP gets invited, clients and prospects alike; the 30% isn't a cap on outreach, it's a floor on who ends up in the room. And the funnel behind it is steep: it typically takes something like 600 invitations to land 100 registrations, and 20 to 30 of those 100 need to be non-clients for the rule to hold.
It's also never come at a client's expense. Account managers were never filling an 80-person room with their own clients alone, there's always capacity left over. Adding the 30% from strangers doesn't crowd anyone out, it just uses room that would otherwise sit empty. The only condition on who fills it is that they meet our ICP.
When it's enforced
The check happens before the invitations are sent, not after the event, when it's too late to do anything about it - that timing is the entire mechanism, because checking composition after the fact just gives you a report on a mistake you can no longer fix. In practice, that meant building a real target list instead of one mass send: out of roughly 600 invites, something like 150 went to existing clients and 450 to prospects, weighted that heavily because a cold prospect converts at a fraction of the rate a warm client does. Reaching those 450 prospects meant working through them person by person, direct outreach, rather than firing off one blanket email to everyone at once.
Who owns the KPI
The person building the invitation list owns the number, tracked as an average across every workshop they run, not graded pass or fail on a single guest list. That's what makes it their actual job rather than a formality: deciding who gets invited is the metric they're judged on over time, not a box to check once.
What happens when the list doesn't hit 30%
The reality is less binary than a single list passing or failing. Nobody is expected to hit exactly 30% on every event; what matters is the average over time, and the person owning the KPI knows they need to keep bringing in ICP-fit strangers to hold it. What actually holds people accountable is the debrief we run after every workshop, going through in detail what worked and what should change before the next invite list goes out. One thing that debrief taught us early: non-clients say yes far more often when an existing client is the one speaking and sharing their own experience, and when the landing page has something genuinely worth reading on it. Get those two right and the 30% takes care of itself.
What changed when we flipped the metric
How the team started thinking differently about invitations
Once the 30% rule was in place, invitations stopped being an administrative task and became a targeting exercise. We never ran this through a LinkedIn or Google campaign; you don't control who actually shows up on the other end of an ad click, even with an approval step in place. Instead, we kept our own running database of people we wanted at an event one day, filtered it by topic and industry for each specific workshop, and reached out manually. Slower, but it meant the emails landed in an inbox instead of a spam folder.
From "did we fill the room?" to "who's actually in it?"
That's the shift in one sentence. Nobody on the team talks about hitting 80 attendees anymore. The conversation now is whether the three or four companies we most want to open doors with are actually going to be sitting in that room. What changed the most, though, is what happens after: a debrief with sales where we go through every attendee, one by one, and decide who moves to SQL for a salesperson to follow up directly, and who stays an MQL for marketing to keep nurturing.
The downstream effect on conversation quality and pipeline
Better composition produced better conversations, and better conversations are what turn into pipeline. That's not a coincidence. It's the direct, traceable effect of changing one number on one KPI, reinforced on both ends: a pre-event briefing where we go through the registrant list company by company and assign a specific salesperson to each person attending, and the post-event debrief that follows.
The results - what own events produce
The numbers
We run four of these workshops a year, 80-plus attendees each. Five of the ten new logos we've attributed to marketing this fiscal year have a workshop somewhere in their history, though it's rarely the only touchpoint. Most of those accounts also came in through LinkedIn campaigns or reached out directly through the website, so crediting the workshop alone would overstate it. What the workshop usually adds is the conversation that turns an anonymous account into a qualified one.
Own events vs. LinkedIn vs. third-party events
Each channel does something different. LinkedIn campaigns are where we generate volume and run the top of the funnel at scale. Third-party events are where we show up because the industry expects us to, and mostly they produce badge scans. Our own workshops are where the actual conversations happen, because we control the room and who's standing at the front of it, not just the guest list.
Why "own events" is consistently the highest-quality channel
Third-party events produce badge scans. Our own events produce real conversations, and in consulting, that's the only kind that eventually becomes a contract. We didn't get here by getting better at filling seats. We got here because the KPI forced the right question from the start.
What you need to run your own events well
The format that works
The format that actually works: our clients presenting their own real results to a room that includes prospects who haven't heard it before. We don't hand them a deck and a podium, though. I interview them on stage instead of letting them run a straight presentation, because a client walking through a technical project solo can get dry fast, and a technical, dry presentation is exactly when people start checking their phones. Interviewing keeps it conversational, and the floor stays open for questions the entire time, not just at the end; I'd rather someone interrupt with a question the moment it occurs to them than sit on it.
Why the event must not feel like a vendor pitch
The second a workshop feels like a pitch, the prospects in the room switch off - they've already sat through that presentation from five other vendors. What keeps people listening is specificity: a real project, real numbers, real constraints someone actually had to work around. It also helps that it's a peer talking, not us. When someone from the same industry describes a problem they've actually had, the audience trusts it in a way they'd never trust the same point coming from a vendor, and the language matches how that audience actually talks about their own work.
Minimum viable setup
The setup itself isn't complicated: a room, and a guest list discipline strict enough that someone's job depends on it, plus a follow-up process that catches conversations before they evaporate by Monday. What isn't minimal at all is the content. Getting a single workshop ready properly took us three to four months: building the material, shaping the landing page, and lining up speakers who needed that much lead time to commit.
If your own events aren't producing the pipeline they should, I'm happy to talk through what changed ours.