The scan to unique lead ratio and what a healthy number looks like

An exhibitor sales director emailed the show team ten days after close. His stand had 1,940 scans and 1,050 unique badges, and the question was whether that was good.

Nobody in the show team could answer it, because nobody had computed the same figure for any other stand. They knew his lead count. They did not know where it sat.

The scan to unique lead ratio between those two numbers is the cheapest diagnostic an organiser owns. It needs no CRM, no survey, no exhibitor cooperation and no additional instrumentation. It is one division per exhibitor over data the lead retrieval system already produces, and it tells you something the lead count cannot: how the scanning kit was actually used.

One division, computed per exhibitor

For each exhibitor, count total scan events and count distinct badge ids, then divide the first by the second.

His stand: 1,940 divided by 1,050 is 1.85. Every unique visitor was scanned, on average, 1.85 times.

Do it for all 214 exhibitors at the show and you get a distribution. On a floor I would expect the median to sit somewhere between 1.1 and 1.3, with the tenth percentile close to 1.02 and the ninetieth somewhere near 1.7. Those are the shapes to look for, and your own show will have its own.

Whether you compute the numerator before or after suppressing near duplicate scans changes the number materially, and you have to state which you did. The windowed rule that makes that decision is E11's subject. For the rest of this post, assume the numerator is post-suppression scan events, because a ratio computed on raw rows is mostly measuring how many devices timed out.

What does a ratio near 1.0 tell you?

A stand at 1.03 scanned 620 badges and captured 640 scan events. Almost every visitor was read exactly once.

The comfortable reading is that this stand runs a disciplined single pass capture: greet, qualify, scan, move on. That reading is often correct for a small stand with two staff and a short show.

The uncomfortable reading is that the stand scanned everyone once and never spoke to anybody twice, which on a three day show with a serious product demo is a sign that the stand is harvesting rather than selling. A visitor who books a follow-up demo, comes back for it, and gets scanned again by the engineer running it is a better lead than one who was read at the aisle and never seen again. A ratio pinned at 1.0 across a four day event means that pattern is not happening.

The way to separate the two readings is to look at whether the repeat scans that do exist are spread across days, which is E13's subject, and to look at how many separate scanning users the stand had active.

What does a ratio above 1.6 usually mean?

At 1.85, the arithmetic starts to strain.

Suppose that exhibitor had four staff on the stand in the average hour across 24 exhibiting hours. That is 96 staff hours. Against 1,940 scan events, the stand captured 20.2 scans per staff hour, one every three minutes, sustained across three days including the dead hour after lunch on the last day. Scan rate as a staffing measure is E31's, and it is being used here only as a sanity check.

The CEIR Industry Insight Report on improving lead quality and sales conversion, written by Jefferson Davis of Competitive Edge and published by CEIR in 2019, sets out an Exhibit Interaction Capacity Formula for the opposite purpose, working out how many leads a stand should expect. It multiplies exhibiting hours by staff on duty by a target number of interactions per staffer per hour. His worked example uses 24 hours, four staff and three interactions per staffer per hour, giving 288 interactions, and then applies the 39 per cent qualification rate that CEIR's 2015 study on Exhibitor ROI and Performance Metric Practices found among exhibitors who qualify leads, for a goal of 112 qualified leads.

Run that in reverse and it becomes a plausibility ceiling. Three interactions per staffer per hour is a working target for a conversation with qualification in it. Twenty scans per staff hour is nearly seven times that. Either the interactions are around 90 seconds long, in which case they are badge grabs, or the scanner is doing something other than recording conversations.

In practice, above 1.6 the usual causes are an unattended kiosk at a prize draw, a demo theatre reader firing on everyone who sits down, or a staff member re-scanning at each stage of a multi-station stand so the CRM shows a journey. All three are legitimate stand designs. All three make the lead count and the ratio mean something different from the stand next door, and none of them will be mentioned in the exhibitor's own report to their board.

Publish the median so the number reads in context

A single exhibitor cannot interpret 1.85. The same exhibitor looking at 1.85 against a published show median of 1.21 can interpret it in about four seconds.

So put the median on the exhibitor report. One line, computed across every exhibitor at that edition of that show, with the count of exhibitors it was computed from. Not an average, because a handful of kiosk-heavy stands will drag a mean upward and the median will not move.

The 2025 Channel Insights Report from Explori and UFI, produced with support from SISO, draws on more than 3,000 events surveyed between 2017 and 2025 and tracks four exhibitor measures: overall satisfaction, likelihood of return, net promoter score and importance. Exhibitor net promoter score across that set has shifted upward by 27 points comparing post-pandemic events against pre-pandemic ones. One of the report's findings is that exhibitors want organisers to provide the tools to measure and track business outcomes. A show median next to an exhibitor's own figure is about as cheap a version of that as exists, and it costs one extra column in the report template.

Comparing an exhibitor to the other exhibitors at the same show, in the same halls, on the same days, beats comparing them to any figure published across the sector, and benchmarking within a show makes that argument properly in E23.

What to do with each end of the distribution

The point of computing this for 214 stands is that it sorts your account managers' week for them.

The stands below about 1.05 with a large unique count are running high volume single pass capture, which is a separate question from how lead volume itself is spread across the floor, covered in E21. Ask whether they wanted volume, because plenty of exhibitors did and are perfectly happy. If they wanted qualified conversations, the conversation is about stand layout and staffing.

The stands above 1.6 need a note in the report explaining what the ratio means, because otherwise their internal report divides spend by scan count and produces a cost per lead that is too flattering by a factor of nearly two. That correction is better coming from you before the renewal conversation than from their finance team after it.

The stands sitting at exactly 1.00 with a suspiciously round unique count are usually not scanning at all and have uploaded a list from somewhere else, which is worth knowing separately.

Where this stops

The ratio is a measure of capture behaviour and it will never be a measure of lead quality.

A stand at 1.85 with an unattended kiosk and a stand at 1.85 with a genuine three station demo journey produce the same number, and no amount of arithmetic on the scan table separates them. The only thing that does is knowing what was on the stand, which means either an exhibitor services record of what kit they ordered or a walk of the floor with a notepad. Most organisers have the first and almost none of them join it to the scan data, which makes it the cheapest upgrade available to your exhibitor analytics.

The ratio also degrades badly at small volumes. An exhibitor with 14 unique badges and 19 scans has a ratio of 1.36, and that number carries essentially no information, because two extra kiosk reads would move it to 1.5. I would suppress the ratio entirely below about 50 unique badges and print the raw counts instead. Publishing a precise-looking figure computed from fourteen visitors is the kind of thing that gets quoted back at you.

And the ratio says nothing at all about the exhibitors who scanned nothing, because they have no denominator. That population is usually larger than anyone expects and it is a different measurement problem, covered in E3.

Take the last edition of your largest show, compute total scans and distinct badges per exhibitor, and sort by the ratio. The stands in the top twenty are the ones whose lead counts you should not compare to anybody else's without a footnote, and you will probably recognise most of them by name before you check what kit they ordered.