Choosing an exhibitor attribution window that matches the real sales cycle

A packaging machinery exhibitor pulled out of a show after eleven consecutive editions. The stated reason, in the exit conversation, was that the last edition produced nine opportunities against a lead count of just over four hundred.

Nine was the figure in their own post-show report, produced thirty days after close because the exhibitor attribution window had been set by the marketing team's reporting cycle. Their median sales cycle from opportunity creation to close was around four months. Thirty days after a show, a machinery buyer who met them on the stand has typically had one follow-up call and has not yet been through procurement, so no opportunity exists to count.

The window was wrong. The decision was real.

Two lags, and most reports collapse them into one

Between a badge scan and a closed deal there are two distinct delays, and treating them as a single number is the source of most bad attribution windows.

The first is the follow-up and qualification lag: the time between the scan and the moment an opportunity record gets created in the exhibitor's CRM. This is a function of how fast the sales team works the list, how long qualification takes, and how quickly the buyer responds. On a B2B floor it commonly runs from a few days to a few months.

The second is the sales cycle proper: the time from opportunity creation to closed won or closed lost. This is a property of the product and the buying process, and the exhibitor almost certainly knows their median for it, because it is on a slide in their sales operations deck.

A window meant to capture pipeline only has to cover the first lag. A window meant to capture revenue has to cover both. Reports that quote one window for both measures are always wrong for at least one of them, and the rule for attaching closed won revenue to a scan is E17's.

What is a thirty day window actually measuring?

A thirty day post-show number is a follow-up speed report. It measures whether the exhibitor's sales team worked the leads, and it does that well.

It measures revenue not at all. For any product with a procurement step, thirty days is inside the qualification lag for most of the file, so the opportunity count at day thirty is dominated by deals that were already in motion before the show. Publishing that number as the exhibition's contribution understates the show and rewards exhibitors who sell simple products.

I would keep the thirty day cut and rename it. Call it follow-up activity, report contacts touched, meetings booked and opportunities created, and say explicitly in the report that it is a measure of the exhibitor's follow-up rather than of the show's output. That framing makes it useful to a stand manager and stops it being quoted as a return figure.

CEIR's 2026 Marketing Spend Decision Report found that sales metrics dominate how exhibitor management evaluates exhibition ROI, with lead volume and post-show closed deals ranking highest as measurement priorities. If closed deals are what leadership looks at, the window over which they are counted determines whether the channel survives the budget review, and thirty days guarantees the channel loses.

How long should an exhibitor attribution window be?

The rule I would use is straightforward and it needs one number from the exhibitor.

Ask for their median sales cycle in days. Set the revenue window at 1.5 to 2 times that figure. For an exhibitor whose median cycle is 120 days, the window lands between 180 and 240 days.

The multiplier exists because the median is a median. Half of their deals take longer than 120 days, and a window set at exactly the median cuts the reporting off in the middle of the distribution, so roughly half the eventual revenue falls outside it. Going to 1.5 times covers a large part of the upper half without extending so far that the attribution becomes meaningless. Going to 2 times is the conservative choice for capital equipment, where the tail is long and heavy.

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, tells exhibitors to set firm post-exhibition reporting dates and to choose them against the length of their own sales cycle and the frequency of the show. That advice is written for an exhibitor chasing their own reps for outcomes, and the same logic decides an organiser's window. The cycle sets the date, and a marketing reporting calendar has no business setting it.

Salesforce's Customizable Campaign Influence has an auto-association time frame in setup that limits when a member-contact relationship counts as influential, and the documentation is clear that this is a configuration choice. Set short, it will silently exclude every exhibitor whose buyers take months, which describes most of a machinery or pharmaceutical floor. The mechanics of that join are E14's subject, and the point here is that the value is a decision somebody made once and it is almost certainly not your exhibitor's cycle.

The cumulative curve on one exhibitor

Take an exhibitor with 412 unique leads from one edition and a 120 day median sales cycle. Count opportunities created with a scanned contact attached, cumulatively, by days since show close. Which leads reach that count at all depends on the matching tiers behind the join, which is E16's.

• day 30: 9
• day 60: 19
• day 90: 27
• day 120: 34
• day 180: 41
• day 240: 47
• day 365: 50

The thirty day report showed 9. At 180 days the figure is 41, which is 4.6 times larger. At 240 days it is 47. Between day 240 and day 365 only three more arrive, so the curve is essentially flat past eight months.

Two ratios are worth stating on the report. The day 30 count captures 9 of the eventual 50, or 18 per cent of what will be visible a year out. The day 180 count captures 41 of 50, or 82 per cent. That second number is the argument for the window: it gets you most of the truth roughly six months earlier than waiting for the curve to flatten.

The shape also tells you where the two lags separate. The steep section from day 30 to day 120 is mostly qualification lag, opportunities appearing as the sales team works through the list. The slow section after day 180 is mostly long-cycle deals whose first contact was at the show and whose procurement took two quarters. If your curve is still climbing steeply at day 240, the window is too short and the exhibitor's stated median sales cycle is probably optimistic.

One window for the show, a different one per exhibitor

There is a genuine tension here and it is worth resolving explicitly.

Comparability within a show needs one window applied to everybody. If the software exhibitor is measured at 90 days and the machinery exhibitor at 240, their opportunity counts cannot sit in the same table, and any percentile or category ranking built on them is meaningless.

Accuracy for the individual exhibitor needs a window matched to their cycle. Those two requirements cannot both be satisfied by one number.

My resolution: the organiser publishes one show-wide window, chosen from the median sales cycle across the categories that dominate the floor, and uses it for every comparative figure in the exhibitor report. Alongside that, the organiser ships the lead export and the method so the exhibitor can run their own window inside their own CRM. The comparative numbers are consistent, the exhibitor's own number is accurate, and the report says which is which. When each of those figures gets published is a scheduling decision of its own, which E18 takes.

The window has to be fixed and published before the show. A window selected after seeing the data is a free parameter, and anyone who has watched a marketing team pick a lookback that makes the quarter look good knows how that ends.

The cost of making the window longer

Long windows are not free, and the cost is contamination.

Every additional month in the window admits more opportunities that would have existed without the exhibition. A buyer scanned in March who opens an opportunity in November has probably had six other touches with that supplier in between, and calling the November opportunity show-influenced is a stretch that gets harder the further out you go.

There is a rough test for where the window stops earning its keep. Compare the opportunity creation rate among scanned contacts to the rate among a comparable set of CRM contacts who were not scanned, month by month after the show. Early on the scanned group runs far ahead. At some point the two rates converge, and past that point the window is mostly counting business as usual. That crossover, where you can compute it, is a better window boundary than any multiple of the sales cycle. Most exhibitors cannot compute it, which is why the multiplier rule exists.

Where this stops

The rule depends on the exhibitor telling you a median sales cycle, and the number they give you is often not a median.

Sales operations teams frequently quote an average, which on a right-skewed distribution runs well above the median and would push a 120 day median exhibitor to a 400 day window. Some quote a target rather than an observation. Some quote the cycle for their enterprise segment when the show brings them mid-market buyers with a much shorter one. You have no way to audit any of that from your side of the join.

The deeper limit is that the window is a blunt instrument applied to a continuous process. An opportunity created on day 179 counts in full and one created on day 181 counts not at all, which is arbitrary at the boundary and will be visible to anyone who looks at the daily counts. Decay weighting is the obvious alternative and I would avoid it in an exhibitor report, because a stand manager can check a date against a cutoff and cannot check a weighted credit, and a measure nobody can check is a measure nobody trusts.

Ask your five largest exhibitors for one number each: their median days from opportunity creation to close. Take the median of those five, multiply by 1.5, and compare it to the window your current post-show reporting actually uses. If the gap is more than a month, your exhibitor analytics has been telling exhibitors the wrong thing about your show.