Automating Hotel Reconciliation: PMS, OTAs, Cards and Banks
What to automate in hotel reconciliation, how matching rules and exception queues work, and the controls that keep automation accurate.
Quick answer: Automate the matching, not the judgment. Let software match PMS transactions to card processor settlements, OTA payouts and invoices, and bank deposits, using clear rules and tolerances. People then work only the exceptions, with approvals, audit trails and regular sampling of the auto-matched items so the automation itself stays accurate.
What does hotel reconciliation involve?
Hotel reconciliation means proving that what the PMS says was earned matches what was billed, what was paid and what reached the bank. It spans several data sources that were never designed to agree with each other.
- PMS: reservations, folios, payments posted, and daily revenue.
- Card processor: authorizations, settlements, fees, refunds and chargebacks.
- OTAs: commission invoices, payouts net of commission, and virtual card payments.
- Bank: the deposits and withdrawals that actually happened.
- Accounting system: where it all needs to land correctly by month-end.
Done manually, this work is usually sampled rather than complete, and it often slips to the end of the month. See our month-end close guide for how it fits the wider close.
Which reconciliation tasks should you automate first?
Start with high-volume matches that follow clear rules. They produce the biggest time savings and the most measurable results.
- Card settlements to bank deposits: match processor batches, net of fees, to bank lines.
- PMS card payments to processor settlements: confirm every posted payment actually settled, and every settlement belongs to a folio.
- OTA commission invoices to stays: check each commissioned reservation against what actually happened (stayed, cancelled, no-show, shortened).
- OTA virtual cards: confirm every virtual card was charged in full before it expired.
- OTA payouts to reservations: for merchant-model bookings, match payouts to the stays they cover.
These five areas share a pattern: large volumes, structured data on both sides, and a clear definition of what a correct match looks like. They also carry deadlines. A virtual card that is not charged before it expires, or a commission dispute raised after the OTA window closes, may not be recoverable, so automation here protects revenue as well as saving time.
Leave items that need interpretation, such as contract disputes or unusual group billing, in a human workflow, supported by the data automation gathers.
How do matching rules and tolerances work?
A matching rule tells the system which fields must agree, and a tolerance tells it how close is close enough. Good rules combine several fields, such as confirmation number, card last four digits, date window and amount.
For example (illustrative numbers only): a processor settles a batch of $50,000 in gross card sales. Fees of 2.5% are $1,250, so the bank receives $48,750. A rule that expects $50,000 at the bank will fail every day. A rule that expects gross sales minus the fee schedule matches cleanly, and any difference beyond a small tolerance becomes an exception.
Keep tolerances tight and documented. A generous tolerance makes the match rate look good while hiding exactly the errors you want to find.
What is an exception-based workflow?
An exception-based workflow means staff only touch transactions the system could not match. Everything else is matched, logged and available for review.
For example (illustrative numbers only): a hotel has 2,000 card transactions in a month. If 96% match automatically, that is 1,920 matched and 80 exceptions. Reviewing 80 items carefully is realistic. Reviewing 2,000 is not, which is why manual processes usually sample.
A good exception queue:
- Groups exceptions by type (missing deposit, amount difference, duplicate, unmatched OTA charge).
- Shows the source records side by side, so the reviewer does not have to log into four systems.
- Assigns an owner and a due date to each item.
- Records the resolution and who approved it.
- Tracks aging, so nothing sits open past the close or past a dispute deadline.
What controls should an automated reconciliation have?
Automation should make controls stronger, not bypass them. The same principles in a financial controls checklist apply.
- No silent write-offs: the system should never adjust or write off a difference without a named approver.
- Segregation of duties: the person who changes matching rules should not also approve exceptions.
- Change log for rules: record every change to rules and tolerances, with a reason.
- Sampling of auto-matches: review a sample of "clean" matches each month to catch rules that are wrong.
- Read-only data access: reconciliation tools should read from systems, not write back to them.
- Deadline tracking: virtual card expirations, chargeback response windows and OTA dispute windows should drive priority.
What results should you expect from automating reconciliation?
Expect the first months to surface more exceptions, not fewer, because the process is checking every transaction for the first time. Some of those exceptions are historical recoveries: commissions that should be credited, cards that can still be charged, and fees that were never collected.
Over time, as rules are refined and root causes are fixed, exception volume should fall and close should get faster. Measure match rate, exceptions per month, average age of open exceptions, and dollars recovered.
How does reconciliation connect to OTA revenue leakage?
OTA activity is where many hotels see the largest reconciliation gaps, because commissions are billed on the OTA's view of a reservation, not the hotel's. Stays that were cancelled, no-showed, shortened or fraudulent can still be commissioned. x·quic OTA Commission 360° matches OTA invoices to contracted terms and audits every reservation 72 hours after check-out, clawing back commissions that should not have been charged. Read OTA commission overbilling and the true cost of OTA bookings for more.
Frequently asked questions
Can reconciliation be fully automated?
Matching can be largely automated, but exceptions need people. Disputes, contract interpretation and approvals should stay with a named person.
How often should a hotel reconcile card and OTA activity?
Daily for card settlements and deposits, and continuously or at least weekly for OTA activity and virtual cards, since expirations and dispute windows do not wait for month-end.
Do we need to replace our PMS or accounting system to automate reconciliation?
Usually not. Most reconciliation tools read data from existing systems. See our FAQ for how x·quic uses read-only access.
What if we manage multiple hotels?
Standardize matching rules and exception categories across properties, then review results in one place. Our management companies page explains portfolio-level views.
Find the commissions you never owed.
Your free 1-year Profit Audit matches every OTA commission line to the stay that actually happened, on your own data. No cost, no commitment, nothing to install.
