Measurement foundations and trustworthy sport data · Lesson 2

Reconciling ticket, scan and capacity ledgers

11 min read

Lesson objectives

  • Reconcile valid entitlements, scans and re-entries and present unresolved records as a conditional range.
Reading tools

Last time we separated admission entitlements from people who entered. Now you must close a match report, and the ticketing platform, gate scanners and seating plan provide different figures. Do not average them. Build a reconciliation bridge that shows how each raw ledger becomes the measure you intend to publish. A residual after that bridge is an exception to investigate, not a number to hide in a convenient category.

Request a defined ticket balance: tickets issued, then tickets voided or refunded, and tickets still valid at the report cut-off. Keep complimentary allocations separate from paid ones. Then inspect the gate record: a successful scan is not necessarily a newly admitted person. Remove documented re-entries from the unique-entrant count, and keep rejected scans outside successful admissions. Finally, take operating capacity from that event's seating plan after excluding unavailable seats. Attach a numerator, denominator and extraction time to every percentage. A late update made after one system's cut-off can create an explainable mismatch.

A sales manager may ask whether 1,100 ticket holders stayed away. That figure assumes every unresolved scan is a distinct admission, which remains unproved. The provisional difference between 10,000 valid entitlements and unique entrants ranges from 1,100 to 1,140. It does not identify who stayed away: invitations, transfers and synchronization errors need their own checks. A precise no-show rate from a preliminary report would confuse accurate arithmetic with incomplete evidence. State how the range affects the decision rather than manufacturing precision.

Use read-only working copies and list the fields needed for matching. You will normally need the ticket identifier, event identifier and scan result. The device number helps investigate the fault, but identifies equipment rather than a person. Inspect examples of apparent duplicates before adopting a deletion rule. The same pass may have a rejected attempt followed by legitimate admission, or an authorized re-entry. Keep an exclusion reason for each removed row in a separate table. A colleague can then inspect the record and its treatment. Removing every repeated pass identifier could erase a legitimate movement between event areas when the system records such movements.

Set a practical reporting cut-off that lets the team act in time. Define what makes a report preliminary and what is required before calling it reconciled. Your team might require all device files and resolution of exceptions material to the decision. Process owners choose these conditions for their workflow; they are not a universal reporting deadline. Record the remaining exception, its size, owner and expected resolution date. When management needs a decision earlier, show the count bounds and what each would imply. If the resource recommendation remains the same at both bounds, a justified provisional recommendation may be possible while the unresolved reconciliation remains visible.

Check validity over time as well. A pass cancelled in the latest file may have been valid when scanned and cancelled during a later adjustment. Examine status history rather than current status alone. Ask the system owner to explain the sequence and supporting record. Keep financial treatment in its own ledger while attendance reflects supported admission. When a pass is transferred or reissued, retain the relationship between old and new identifiers and inspect it before counting people.

Lesson summary

Reconciling voids, refunds and repeat scans separates valid tickets from admissions. Unresolved records create a conditional range and block a falsely final count.