Hotel KPI list: which metric answers which decision?

I have rarely seen a hotel report short of metrics; I have seen many drowning in them. The PMS produces a forty-column end-of-day report, accounting adds as many again, and by month end there is a hundred-row table nobody knows how to read. Plenty of metrics, few decisions.

This is not a KPI glossary — the definitions live in the glossary. Here I work backwards: a hotel makes five recurring decisions, each decision has three or four metrics of its own, and each metric has a correct reading frequency. Build the list that way and the report gets shorter while decisions get faster.

Not a KPI list — a decision map

The "key" in Key Performance Indicator is the part most hotels forget. Everything that can be measured gets measured, so nothing is key. The right question is not "what can we measure" but "which decision are we making, and which number would change it".

Five decisions recur in every hotel: the price for tomorrow, capacity and length-of-stay rules for the coming period, which channel to sell through, profitability, and the team. The table below gives the metrics that feed each decision and how often they are read.

The frequency column is the most important one. Seeing a daily metric at month end means acting late; reading a monthly metric every day means reacting to noise.

The five decisions, one by one

1. The pricing decision: forward-looking metrics

Prices are set for the future, not the past, so every metric in this group measures dates that have not arrived yet. OTB (on the books) tells you how many room nights you hold for a date right now; pickup tells you how many were added to that date since yesterday. Read together they show which dates demand is heading for and how fast; the pricing decision is your response to that speed.

Rate shopping gives the competitive set's price for the same date; RGI (Revenue Generation Index) gives your RevPAR as a ratio of the comp set's RevPAR. Above 100 you are taking more than your fair share of the market. Watching competitor rates without watching RGI is looking at the price list and never at the result.

2. The capacity and length-of-stay decision

Occupancy is the most familiar metric in this group and the most misleading. Made a target on its own, the team hits it by cutting rate. The correct reading is occupancy alongside length of stay: the same 80% occupancy means a completely different cost base and channel structure at an average stay of 1.4 nights than at 3.2.

Lead time (how far ahead bookings are made) tells you when to switch on minimum-stay and closed-to-arrival rules; the no-show rate tells you where to set the overbooking limit. These four are read weekly — daily movement is noise, and a monthly reading is too late to change a rule inside the season.

3. The channel decision: net, not gross

Most hotels track channel share (what percentage of bookings come from OTAs, direct, agents); few track net ADR. Net ADR is the rate left for the room after commission. A 3,000 TL OTA booking at 18% commission nets 2,460 TL; a 2,700 TL direct booking is worth more. A hotel deciding on gross ADR grows its OTAs; a hotel deciding on net ADR grows its direct channel.

Cost of distribution is the whole of that calculation: commission, channel manager fees, payment costs and the advertising spend behind direct sales, as a share of room revenue. It is measured monthly; channel strategy is not something to change every week.

4. The profitability decision: the ladder from revenue to profit

RevPAR shows room revenue, TRevPAR total revenue, and GOPPAR the gross operating profit per available room after costs. The three form a ladder; a hotel that only looks at the first rung cannot understand why revenue grows while profit does not. CPOR (cost per occupied room) gives the floor rate, and flow-through tells you what share of each extra unit of revenue reached profit.

Flow-through is a practical cross-check: if revenue rose by 100 and profit by only 20, the other 80 went to cost or commission. To find out which, go back to the channel group.

5. The team decision

In most hotels the performance of the reservations and front desk team is felt rather than measured. Four things can be measured: enquiry-to-booking conversion, response time to enquiries, upsell revenue per person, and guest review score. This group is read weekly, because training and feedback work on a weekly rhythm; "conversion fell" at month end is a finding nobody can act on.

Example: one month, five readings

The figures below are fictional and do not belong to any real property. Take Sample Hotel: 120 rooms, city centre, open all year. One reading per decision from a single month-end report:

The five readings are linked: the low flow-through in the profitability row points at the OTA weight in the channel row; the channel decision will move the BAR band in the pricing row. This is exactly the moment a KPI map earns its keep — the metrics are read as a chain of decisions, not one at a time.

The four most common mistakes

1. Treating everything measurable as a KPI

A forty-metric report highlights nothing. Write the decision each metric feeds next to it; any metric without one leaves the report. In most hotels this cuts the list to a third and nobody misses the rest.

2. Making occupancy the target

A team given an occupancy target hits it by cutting rate. The target should be RevPAR or RGI; occupancy is one of its components, not the target itself.

3. Mixing daily and monthly metrics

Changing price on daily RevPAR is reacting to noise; changing price on monthly pickup is acting late. Each metric's frequency is in the table; split the report the same way — the morning report and the month-end report should not be the same document.

4. Changing definitions between reports

If ADR includes VAT in one report and excludes it in another, or occupancy is calculated on total rooms in one and sellable rooms in another, the metrics can no longer be cross-checked. Each definition is written once and stays the same everywhere. The denominator debate in the RevPAR article is the best-known example.

How to set it up in your own hotel

1. 01Write down the five decisions. If your recurring decisions differ (a resort's season-opening decision, a city hotel's corporate contract decision), build the list from your own — it is a map, not a template.
2. 02Put at most four metrics under each decision and define each in one sentence: what is the denominator, are taxes included, which period.
3. 03Give each metric a frequency and split the report in two: a daily morning report (pricing group) and a monthly management report (channel, profitability). Team and capacity sit on a separate weekly page.
4. 04Give each metric three references: same period last year, budget, competitive set. A metric without a reference is just a number; it does not say whether it is good or bad.
5. 05Come back after three months and check which metrics changed no decision at all. Those leave. The list is a living document, not something built once and left.

Related concepts

• RevPAR, ADR and occupancy — rate and volume in a single number; see the RevPAR article for the denominator debate.
• Pickup and OTB — the two daily inputs of the pricing decision; the subject of the next article.
• GOPPAR and flow-through — the last rungs of the ladder from revenue to profit.
• Cost of distribution and net ADR — reading the channel decision net rather than gross.
• RGI, MPI, ARI — position against the competitive set; defined in the glossary.

Summary

1. 01A KPI list is a decision map: five decisions, three or four metrics each.
2. 02Every metric has a reading frequency; the morning report and the month-end report are different documents.
3. 03Occupancy is a component, not a target; the target is RevPAR or RGI.
4. 04The channel decision is made on net ADR and cost of distribution, not gross rate.
5. 05Definitions are written once and kept identical across reports.
6. 06Revisit the list quarterly: a metric that changes no decision leaves the list.