Business Intelligence - Data Processing Policy
Last updated: 10 August 2026
This policy describes how Lighthouse Business Intelligence products process hotel performance data and what customers can expect with respect to cross-source data variance, timeliness, and processing reliability.
Data Sources
Lighthouse Business Intelligence products integrate three distinct data sets sourced from a hotel’s property management system. Each is governed by its own rules and serves a different purpose within the platform.Summary Data
Aggregated daily KPIs such as occupancy, Average Daily Rate (ADR), capacity, and group blocks. Summary data represents the hotel’s overall operational picture as reported by the PMS. It may include values, such as group blocks, complimentary stays, or redemptions, that are not tied to individual reservations.
Reservation Data
Granular, reservation-level records including revenue, room number, stay dates, and rate codes. Reservation data reflects individual bookings and may be subject to customer-configured exclusion rules (for example, filtering out day-use rooms).
Transaction Data
Individual financial charges and credits posted to a hotel folio, sourced from the property management system's posting ledger. Transaction data captures revenue at the posting level; including room charges, food and beverage, spa, parking, and other ancillary revenue, as well as adjustments, voids, and payments. Unlike Reservation data, which reflects activity at the booking level, Transaction data represents the complete revenue footprint of a stay and underpins Total Revenue reporting within Lighthouse Business Intelligence products.
Cross-Source Data Variance
Because Summary, Reservation and Transaction Data are distinct sources governed by different rules, they will rarely match exactly. This is expected behaviour, it is not an indicator of a data error. The sources should always be directionally aligned, but differences in scope and configuration mean an exact match is neither expected nor required.
Common reasons for variance between data sources include:
Summary data includes values (such as group blocks, complimentary stays, or redemptions) that are not tied to individual reservations and therefore do not appear in reservation data.
Customer-configured exclusion filters (for example, excluding day-use rooms) apply to Reservation data but cannot be applied to Summary data.
Timing differences in how the source system records and closes transactions.
Rounding rules applied differently across data types within the PMS.
Incorrect or non-standard PMS practices (e.g. deleting reservations without cancelling them)
Interpreting variance
A variance of up to 5% between Summary and Reservation data is within normal operating parameters and is typically attributable to one or more of the reasons listed above. Both data sources accurately reflect the information supplied by the source system; they represent different views of hotel activity, not conflicting measurements of the same metric.
For clarity, two measures apply under this Policy: (i) accuracy, meaning the degree to which data in Lighthouse Business Intelligence matches the data in the source system at the time of extraction; and (ii) cross-source variance, meaning differences between the Summary, Reservation and Transaction data sets within the platform. The 5% threshold applies to both measures. Variances caused by changes made in the source system after extraction (for example refunds, adjustments or redemption revenue) are expected variances, are not errors, and will not be accepted as support issues.
A variance above 5% may indicate a configuration issue, a change in source-system behaviour, or a data processing problem worth investigating. Customers who observe a sustained or unexplained variance above this threshold should contact Lighthouse Support.
Variance involving Transaction Data
Transaction-level data, which underpins Total Revenue, is typically aligned within a 5% variance of the corresponding summary report. As with Summary and Reservation data, the two views are not conflicting measurements of the same metric; they reflect different aggregations of source-system activity, each governed by its own rules.
Common reasons for variance involving Transaction data include:
Adjustments made directly to a transaction after ingestion are not reflected.
Split transactions appear as duplicates.
Different transaction category groupings in our category mapping compared to user integration source category.
A variance above 5% may indicate a configuration issue, a change in source-system behaviour, or a data processing problem worth investigating. Customers who observe a sustained or unexplained variance above this threshold should contact Lighthouse Support.
Data Timeliness
Timeliness refers to how quickly data is available within Lighthouse BI products after it has been generated by the source system.
Under normal operating conditions, Lighthouse aims to make processed data available within a few hours of receipt from the customer’s PMS or integration layer. The exact window depends on the delivery schedule configured for that integration.
Customers should be aware that actual availability may be affected by:
The schedule on which the source system exports or transmits data.
The availability and performance of third-party integration middleware.
Temporary disruptions to upstream systems outside of Lighthouse’s control.
If a customer expects data to be available by a specific time and it is not, they should check whether their source system completed its export before contacting Lighthouse Support.
Data is generally available in the morning local time at each hotel. A guaranteed delivery time is not feasible due to the factors listed above. Customers who do not see expected data by noon and who have not received an in-app or other notification of a delay, are welcome to contact Lighthouse Support after first confirming that their source system has completed its scheduled export.
Processing Reliability
Reliability refers to the consistency with which data is successfully ingested and processed. Lighthouse designs its data pipelines to process all incoming data feeds without interruption under normal conditions.
In the event of a processing failure (caused by malformed data, upstream unavailability, or an internal pipeline error) Lighthouse:
Relies on automated monitoring to detect failures.
Will investigate and resolve the root cause as quickly as reasonably practicable.
Will re-process affected data once the issue is resolved, so that historical reports remain accurate.
Customers who notice a gap in their data, such as a missing day or an unexpectedly stale report, are encouraged to contact Lighthouse Support, who can confirm whether a processing event is under investigation.
Resolution Times
Lighthouse will make reasonable efforts to investigate and resolve data processing issues as promptly as possible. However, Lighthouse cannot guarantee specific resolution timeframes. The time required to resolve an issue is often influenced by factors outside of Lighthouse’s direct control, including:
The responsiveness and availability of third-party integration providers or middleware vendors that sit between the hotel’s PMS and Lighthouse.
The source system’s ability to re-export or retransmit affected data.
The complexity or scope of the underlying issue, which may require coordinated investigation across multiple systems or vendors.
Where a delay is attributable to a third-party provider, Lighthouse will communicate this to the customer and provide updates as information becomes available. Lighthouse is not accountable for delays caused by the failure or unavailability of third-party systems.
Troubleshooting
Customers who observe an unexpected value, variance or gap in their data should:
Identify whether the report is drawing on Summary data or Reservation data.
If comparing values across both sources, assess whether the variance is within the 5% threshold. If so, it is likely attributable to one of the structural differences described in the Cross-Source Data Variance section of this Policy.
Check whether your source system completed its scheduled data export for the relevant period.
If the variance exceeds 5%, data appears to be missing, or the issue persists, contact Lighthouse Support with the relevant date range and report details.
.