The digital marketing performance metrics a hotel should report

The monthly marketing report arrives as a deck. Impressions are up, click-through rate is steady, sessions are at a record. Then the general manager asks whether any of it made the hotel money, and the room goes quiet.

The trouble is rarely too few metrics. It is that the five which answer that question are missing, or buried on slide twenty. Revenue by channel, direct share, blended cost per acquisition, cost of distribution, and return on ad spend with brand and non-brand kept apart. Those are the digital marketing performance metrics worth reporting. The rest are diagnostics that explain why those five moved.

That is how we report in the Growth Marketing practice, and it is the reason a report should start with what the booking engine took rather than what the ad platforms saw.

Why most reports miss

Each platform reports on itself. Google Ads counts the bookings it believes it drove. Meta counts the ones it believes it drove. The metasearch account counts its own. Add them up and the hotel often appears to have taken more bookings than it did, because the same guest clicked three ads before booking once.

None of those platforms is lying. Each is measuring by its own rules, with its own attribution window, and each has a reason to claim the booking. It goes wrong when the report is built from their totals and nobody checks them against the system that actually took the money.

The booking engine and the PMS are the source of truth. Every other number is somebody's estimate.

Getting reliable figures out of those systems and into analytics is a tracking job, and our guide to setting up GA4 key events for a hotel covers it. This piece is about what to do with the numbers once you have them.

The five numbers that matter

Revenue by channel

Start with room revenue from the booking engine and PMS, split by where the booking came from. Direct website, phone and email, each OTA, metasearch, wholesale, groups and corporate. Show it against target and against the same month last year.

This is the number owners and revenue managers already understand, so it is the one that earns the marketing team a hearing. It also shows the trade-off that a single total hides. A good month on the OTAs and a flat month on direct is a different business from the reverse.

Direct share

Direct share is the proportion of room revenue booked through channels you own. Measure it against the business you could realistically win, not the whole pie. Contracted corporate rates, groups and wholesale were never going to come from an OTA, so including them makes the direct number look better or worse than the marketing deserves.

The useful version is direct against OTA, for leisure business only. That is the split marketing can move, and it is the one our guide to boosting direct bookings is built around.

Blended cost per acquisition

Cost per acquisition is what it cost to win a direct booking. Take everything you spend on the direct channel in the month (paid search, paid social, metasearch, agency fees, booking engine transaction fees) and divide it by the number of direct bookings. Expressed as a percentage of direct revenue, it sits next to OTA commission and can be compared with it.

Blended matters. A single campaign's CPA tells you about that campaign. The blended figure tells you what the direct channel costs as a whole, including the brand, the content and the email that never get credited with a booking of their own.

We have to have an overarching cost per acquisition that everyone's bought into. We shouldn't be too microscopic and granular with our measurement.
Harry Fielder, CEO, Umi

Agree the target CPA with revenue and finance before the year starts. Once there is a number the business has accepted, "it is working, can we spend more?" becomes a simple conversation instead of an argument about budgets.

Cost of distribution

Cost of distribution is everything it costs to sell a room, across every channel, as a share of room revenue. OTA commission and programme uplifts, metasearch, GDS and channel manager fees, payment charges, and the direct CPA above.

It is the number that stops direct and OTA being argued about in separate rooms. If direct bookings cost 9% to win and the OTAs cost 18%, shifting share is worth money, and you can say how much. What OTAs really cost goes into the parts of the OTA bill that never reach the invoice.

Return on ad spend, brand and non-brand apart

Return on ad spend (ROAS) is the booking revenue a campaign is credited with divided by what it cost. It is useful, and it is easily flattered.

Brand search, where guests type your hotel's name, will always show a high ROAS because most of those guests had already chosen you. Prospecting campaigns aimed at people who have never heard of you will show a much lower one, and that can be fine. Blend them and a weak prospecting account hides behind a strong brand campaign. Report them separately, and judge brand on what it protects rather than what it returns, which is the case for a brand protection campaign.

Where the rest belongs

The metrics that fill most reports still matter. They are the reasons the five numbers moved, so they belong underneath them.

• Traffic and sessions. Useful for spotting a fall in demand or a tracking break. Not a result, and not a target.
• Impressions and impression share. They tell you whether you were seen, especially on your own name. They do not tell you whether you were chosen.
• Click-through rate and cost per click. Signs of whether the ads and the bids are healthy. A cheaper click that books less is not progress.
• Engagement, followers and reach. Fair measures of awareness work, judged against awareness goals, never presented as revenue.
• Email opens and clicks. Apple's Mail Privacy Protection inflates opens, so report revenue per send and rebookings instead.

If a metric cannot be traced to one of the five numbers, ask why it is in the report at all.

Reading them honestly

Last month will change

Google Ads credits a booking to the day of the click, not the day of the booking. A guest who clicks in March for a stay in August and books in April counts towards March. Hotels have long lead times, so last month's figures keep rising for weeks after the report goes out. Google's conversions by conversion time columns show bookings on the day they happened, which is closer to how the booking engine sees them.

Compare like with like

Month on month is the least useful comparison in hospitality. Easter moves, events come and go, and a hotel's year has its own shape. Compare with the same month last year and with target, and say what changed in the market.

Treat ROAS as a symptom

When ROAS drops, the ads are the first thing blamed and often the last thing at fault. Before changing the account, look at how the OTAs did over the same period. If Booking.com and Expedia fell too, the market moved. If they held and direct fell, look at the rate, the offer and the booking journey before the bids.

Building the monthly report

A report that leads with the commercial numbers is shorter, and it is read further.

1. One page first. The five numbers, each against target and the same month last year, with one line on why each moved.
2. Each number from its own system. Revenue and direct share from the booking engine and PMS, costs from the invoices, ROAS from the ad platforms checked against both.
3. Diagnostics after. Channel detail, traffic, search terms and creative, in the order that explains the first page.
4. One decision. End with what changes next month and which of the five it is meant to move.

Writing the report is not the difficult bit. Joining booking engine, PMS and ad data so the first page fills itself in, rather than someone copying figures between systems, is the job Mio does.

Then put one question to whoever runs your marketing. What did direct bookings cost to win last month, as a percentage of their value?

If the answer arrives in clicks and impressions, the report is measuring the marketing. It should be measuring the hotel.