Most hotel tech companies find out a property is unhappy when it sends notice. Looking back, the signs were there months earlier: logins thinned out, the champion left, support tickets stopped because nobody was trying any more. A customer health score is a way of seeing those signs while there is still time to act.

Generic health score templates borrowed from other software markets tend to fail in hospitality. They weight logins heavily, ignore seasonality and miss the events that matter most at a hotel, such as a new general manager or a change of owner. The framework below is built around how properties actually use and leave software. It has four parts, and you can run it in a spreadsheet before you buy anything.

The four parts

The weights are a starting point, not a rule. Adjust them once you can compare the score with which accounts actually churned.

Part 1: Usage

Usage is the most measurable part, and the one most often measured badly. Raw login counts reward the property where one enthusiastic person does everything and penalise the one that uses the product exactly as intended once a week.

• Breadth: how many named staff used the product in the last 30 days, against how many you trained.

• Core action: whether the one action that delivers the value happened at the expected frequency, such as rates published, rooms released or messages answered.

• Trend: usage this period against the same period last year, or against the property's own last quarter, to allow for season.

If you have not yet defined the core action for your product, do that first. How to instrument activation in a hotel tech product covers how to choose and track it.

Part 2: Outcome

A property can use the product every day and still leave because it did not get the result it was promised. The outcome part ties the score back to the reason the hotel bought.

• Record the property's buying reason and success measure at the sale, in the CRM, in the customer's words.

• Agree a baseline at onboarding and check the measure at 90 days and every quarter after.

• Score the account on whether the measure is improving, flat or getting worse, using the property's own figures.

• Note whether the general manager or owner has seen the result, because a result nobody senior knows about does not protect the renewal.

Part 3: Relationship

• Is there a named champion at the property, and have they responded in the last 30 days?

• Has anyone above the champion, such as the general manager or the owner's representative, been in contact this quarter?

• Is the tone of support tickets neutral or deteriorating?

• Has the property attended training, a review call or a user event recently?

• Have they acted as a reference, or declined when asked?

A single champion is a risk in itself. Hotel staff move often, and a relationship that lives in one person leaves when they do.

Part 4: Property context

This is the part generic templates leave out, and in hospitality it is often the strongest predictor of all. These are events at the hotel that put any vendor's contract at risk, regardless of how well the product is performing.

• New general manager in the last six months.

• Change of ownership or management company, announced or rumoured.

• Turnover in the role that uses the product, which is common at the front desk and covered in the front desk turnover problem nobody prices in.

• A competitor in conversation with the property, which you will often only hear about from the champion.

• Financial pressure, such as a renovation, a weak season or a cost review across the group.

Scoring and bands

Score each part from 0 to 10, apply the weights and add them up to a score out of 100. Keep the scoring rules written down so two people would score the same account the same way. Then group accounts into three bands, each with a clear action.

Note the override in the last row. A new general manager or a change of owner puts an account in the at-risk band whatever its score, because both reset the relationship and usually trigger a review of every system the hotel pays for.

Calibrating the score

1. Score every current account once, by hand.

2. Score the accounts that churned in the last year as they looked six months before they left.

3. Check whether the churned accounts would have shown as watch or at risk. If not, look at which signals were missing and adjust the weights.

4. Review the bands each quarter for the first year, then twice a year.

The score should feed directly into renewals. Accounts in the watch and at-risk bands are where the renewal is decided, often months before the contract date, which is the starting point of building a renewal playbook for hotel software.

The takeaway

A useful health score for hotel properties measures breadth of usage rather than logins, ties the account back to the result it bought, checks that the relationship is wider than one person, and treats changes at the hotel as the warning signs they are. Start it in a spreadsheet, calibrate it against the accounts you lost, and make every band trigger an action.

If retention is where your revenue leaks, the 12-point leak audit takes about twenty minutes and shows how churn compares with demand, conversion and activation on your own numbers.

Frequently Asked Questions