The Event Roster Playbook: From Speaker List to Ranked Accounts

The conference speaker and exhibitor list just dropped, and you have a decision to make before you commit budget and calendar to the trip: which accounts to chase, and who to try to meet. This guide is for founders selling their own product, account executives, SDR leads, and partnerships teams. It turns a public roster into a ranked account plan, each account carrying a named contact and a specific reason to meet, sorted by fit and reachability, before the event begins.

Most sales advice about events lives at one of two extremes: post-show badge-scan cleanup, or generic "how to network" motivation. Neither covers the pre-trip job of converting a public speaker, sponsor, and exhibitor roster into an ordered account plan. That conversion is a procedure, not a mindset, and the order matters because the deadline is not the event. It is the day senior calendars lock, which arrives weeks earlier.

Why the roster is worth mining before you travel

A published roster is a list of companies that have already spent money to be in the room, which is intent you cannot get from a cold database. Companies that pay $5,000 to $50,000 or more for a booth have budget and are investing in growth, and that is a buying signal a cold list will never show you.

The three roster types each carry a distinct signal, and you mine them for different things. Speaker lists give you named senior decision-makers. Exhibitor lists give you company profiles and a proxy for budget. Sponsor lists tell you who has the biggest budgets, ranked by tier. Treat them as three separate feeds, not one blended attendee list.

The event's own website is the primary source and the only one you should trust. Exhibitor directories, speaker and panel rosters, and sponsor pages are almost always public and safe to mine. What you should not touch is an "attendee registration" list sold by a third-party vendor: those are frequently scams or violate the event's terms of service. Everything in this playbook runs on public rosters.

The mechanism behind that gap is the reason this whole job is pre-trip work. The four to six week rule is not a preference; it is dictated by when buyers lock their schedules.

The signal fields each roster type exposes

Capture the right fields from each roster and you can score accounts without ever guessing. The table below maps each roster type to its primary signal and the public fields worth pulling into your sheet.

Two fields do more work than the rest. The session topic on the speaker list is the cheapest way to make an outreach touch specific, because it hands you a reason to meet that the person cannot dismiss as spray-and-pray. The tier on the sponsor page is the cleanest budget proxy you will find, because it encodes a committed dollar amount rather than a maybe.

Sponsor tiers map to explicit dollar ranges. In one real conference example, Platinum was attained by sponsorships totalling $17,000 or more, Gold ranged from $9,000 to $16,999, and Silver ranged from $3,000 to $8,999. A tier label is therefore a far stronger signal than a bare exhibitor listing, because a booth entry alone can be a cheap directory placement with no real spend behind it.

The timing window is set by buyer calendars, not by data

Start pre-event outreach four to six weeks before the conference, land your first touch five to six weeks out, and land your last touch five to seven days before the show. Exhibitor lists usually publish two to three months in advance, so the data is ready long before you need it. The constraint is not availability. It is the buyer's calendar.

Senior buyer calendars lock three to four weeks before major conferences. If your first touch arrives after that, the meeting slots you wanted are already filled. This is why the on-site networking app, which typically opens about a week before the event, is too late for the people worth meeting. By the time it opens, senior buyers have locked their schedules.

Sources disagree on how wide the window should be, and the disagreement is worth understanding rather than averaging away.

The compressed two-to-four-week windows are defensible for junior or operational contacts whose calendars stay flexible. But for the VP-and-above buyers you are chasing off a speaker roster, the wider window wins, because their calendars are exactly the ones that lock earliest. When in doubt, start at five to six weeks and treat the last touch as a hard cutoff five to seven days out, after which you switch to on-site tactics.

The procedure, in order

Run these eight steps in sequence. The order is the point: scoring before de-duplication wastes effort on accounts you cannot own, and picking named contacts before scoring means you research people at accounts that never make the list.

Steps two and three are where most of the value and most of the risk sit. Before you load anything into the CRM, deduplicate against existing records to avoid creating duplicate contacts, assign account ownership by territory rules, and tag contacts with the signal type that triggered their inclusion. Skip this and a rep and an SDR end up working what looks like two prospects when it is the same person, splitting ownership and inflating pipeline.

De-duplication done right is not cosmetic. One company reduced its CRM account count from 650,000 to 180,000 through de-duplication, roughly a 72 percent cut, and reduced disqualified leads by 20 percent as a result. Trade-show leads are a known duplicate source precisely because the same company shows up under minor name variants across accounts.

Scoring: separate ICP fit from roster signal, then combine

Score fit and signal separately, then add them, so you can see whether an account is on your list because it fits or only because it bought a booth. A cheap listing at a company that will never buy from you is not worth a Tier 1 touch, no matter how prominent its logo is.

Fit is your usual judgement: score every account on fit, access, timing, and expansion room. No single validated weighting model exists in the published literature, so do not chase a magic formula. Score on fit plus signal tier and move.

The signal layer adds points for four things, in rough order of reliability:

• Sponsor tier. The strongest signal, because it is a dollar threshold. Platinum outranks Gold outranks Silver, and each maps to committed spend.
• Booth size. A proxy for company scale. A large booth means more than a listing-only entry, so segment by booth size, not mere presence.
• Speaking slot. Signals seniority, because conference speakers are typically senior leaders at VP level and above.
• On-topic session. If a speaker's session topic maps to your product's problem, that account is both budgeted and timely.

The reason the on-topic step is load-bearing is that relevance gates everything downstream. 73 percent of B2B buyers avoid vendors who send irrelevant outreach, and the session-topic field is the cheapest way to make each touch specific. A named contact with a vague reason to meet converts no better than a cold email.

Finding the named contact and the reason to meet

For each Tier 1 and Tier 2 account, you need one specific person and one specific reason to meet, and the roster gives you the person for speakers but rarely for exhibitors. This is where a scored account list becomes a meeting plan, and where partnerships teams hit their real bottleneck.

The named-contact step is harder for some titles than others because the underlying supply is thin. In Refolk's index of professional profiles, there are 1,435 US "Head of Partnerships" contacts against 34,217 "Director of Business Development" contacts, a 23.8x gap. For partnerships teams, the account step is easy and the person step is the constraint, so budget more time here.

That table also warns against copying a playbook across markets. In Refolk's index, there are 8,908 US "VP Sales" contacts against 140 in the UK, a 63.6x gap. Some of that is index composition rather than pure market size, so treat it as an index-supply figure, not a census. But the direction holds: a US VP-targeting sequence cannot be pasted onto a UK roster where that exact title surfaces far less often, so localise titles before you build the contact list.

When the roster names a company but not the person, or names a person whose title may be stale, you need to go from an account and a title to a real, current human. Searching each company's public activity by hand is the slow part of step seven, and it is where a plain-English people search earns its place.

Once you have a name, write the reason to meet before you write the email. Anchor it to something specific: their session, a product they just shipped, an expansion into your territory. The reason is the message.

Subject: your [SESSION TOPIC] session at [EVENT]

Hi [FIRST NAME],

Saw you're speaking on [SESSION TOPIC] at [EVENT]. We work with [SIMILAR COMPANY TYPE] on exactly that problem, and [ONE SPECIFIC, RELEVANT FACT ABOUT YOUR PRODUCT OR RESULT].

I'll be at [EVENT] and would value 15 minutes on [SPECIFIC QUESTION TIED TO THEIR SESSION]. Worth grabbing a slot before calendars fill?

[YOUR NAME]

Finding the right people at the right accounts, across public LinkedIn, GitHub, and the open web, is exactly what Refolk is for: you ask in plain English for the roles and companies on your roster, and get named contacts back.

How this goes wrong: the failure modes

Most pre-conference plans fail on a small set of predictable errors. Each one has a tell, and each one has a check. The table names the signal, what it falsely proves, and how to catch it.

• Booth-as-budget false positive. A cheap listing-only booth is not the $5,000 to $50,000 spend signal. The tell is a company in the exhibitor directory with no booth number or a tiny footprint. Check actual booth size or tier before you score it as budget.
• Roster name equals attendee. A logo on the sponsor page proves company intent, not that your buyer is in the room. The tell is an account you score high with no named person confirmed. Verify with the person's own public activity, such as a post mentioning the event.
• Stale contact after scrape. A contact who was VP of Sales when added may have moved companies three months later with no flag. The tell is a title pulled weeks ago and never rechecked. Re-verify title before the first touch.
• Duplicate-driven double outreach. A rep and an SDR work the same person as two prospects because of name variants. The tell is two open records for one company. Run lead-to-account matching before assigning ownership.
• Weak warm intro. A generic "we're both at X" line converts no better than a sharp cold email. The tell is a reason to meet you could paste into any account. Make the reason specific to their session or their work.
• Purchased attendee list. Third-party attendee lists are frequently scams or violate the event's terms, and if the vendor cannot document a lawful basis, you inherit that risk. Stick to public rosters.
• Starting too late. Waiting for the networking app means senior calendars are already locked. The app opens about a week out; by then the slots are gone.

The two failure modes worth extra weight are the stale contact and the duplicate. Both are silent: nothing in your sheet flags them, and both waste your scarcest resource, which is the touches you get before the calendar locks.

The roster tells you which companies have budget. It does not tell you who to email, and it lies about who is coming.

What "done" looks like, and how to keep it current

You are done when a 400-plus company roster has become a tiered list where every Tier 1 and Tier 2 account has one named, verified contact and one specific reason to meet, and the sequence is live on calendars before you travel. The best exhibitors pre-identify their top 50 target accounts from the roster, and for a standing named-account program the working list runs 50 to 150 companies.

Run this checklist before you call the plan finished.

Keeping the plan current is mostly about re-verifying titles and attendance as the event approaches, because the scrape ages. Re-check any contact pulled more than two or three weeks before the first touch, since a VP can move companies with no flag in your sheet. And if the event publishes a second wave of speakers or exhibitors, which many do closer to the show, run the whole procedure again on the new names rather than bolting them onto the old list. The discipline that makes this work is the same each time: score before you assign, verify before you touch, and reach the buyer before the calendar closes.

To carry this into the next event, save your combined scoring rule and your tier definitions as a reusable template, so the only per-event work is the scrape, the CRM match, and the named-contact hunt. That is the difference between a one-off scramble and a repeatable pre-conference pipeline motion.