This article explains what each KPI measures and how it's calculated, so you always know what your numbers are telling you. You'll find each review KPI first, then the survey KPIs and survey integrity metrics, followed by CompIndex, which compares your KPIs against your competitors.
- Read your KPIs in Report builder with confidence, knowing exactly what each number means.
- Explain your results to your team using the same definitions CXP uses.
- See how you stack up against your competitors with CompIndex.
- Understand your survey results and how far you can trust them.
Review KPIs
- RPS (Reputation Performance Score): The average of all review scores for the filters you've selected, such as date range, location, or segment. It's normalized to a 1–100 scale, so it's easy to compare and read.
- Reviews: The total number of Reviews for the filters you've selected.
- Response Rate: How engaged and responsive your team is with guest feedback. It's the number of Reviews you've responded to, divided by the number of respondable Reviews.
- TrustYou Score: Our own metric, calculated as a weighted average of all Reviews from the past two years. The current and previous month get the same full weight. Earlier months count for gradually less (exponential decay). That way, recent guest feedback carries the most weight while your long-term reputation still counts.
- Popularity Score: How popular each Entity is, based on its Tripadvisor ranking. A higher score means a more popular Entity.
- Source Score: The score shown directly on a review site or online travel agency (OTA), like Google or Booking.com. We collect it from each site every day, for every property in your account.
- Sentiment Score: The percentage of positive mentions out of all mentions for a specific Sentiment Category or Subcategory. CXP's sentiment model detects the mentions, and a single review can include several.
CompIndex
CompIndex compares your performance against your competitors. It's your KPI divided by your competitors' average for the same KPI.
- Above 1: Your KPI is higher than your competitors' average. For example, 1.05 means you're 5% above it.
- Below 1: Your KPI is lower than your competitors' average. For example, 0.96 means you're 4% below it.
CompIndex is available for these KPIs:
- RPS CompIndex
- Reviews CompIndex
- Response Rate CompIndex
- TrustYou Score CompIndex
- Source Score CompIndex
- Sentiment Score CompIndex
Survey KPIs
Survey KPIs are available when your account includes Surveys. Each one uses the survey you pick.
- Survey reviews: The number of reviews guests submitted through the survey.
- Survey response rate: The share of the survey's reviews your team responded to: Responded reviews divided by Responded plus Unresponded reviews. It measures your team's replies, not how many guests answered the survey.
-
Question results: The results of one question. What you see depends on the question type:
- Scale: the average answer, shown on your organization's score scale (for example 0 to 100).
- NPS: the % of promoters (scores 9 and 10) minus the % of detractors (scores 0 to 6).
- Select: how many guests chose each answer.
- Answered: the number of guests who gave an answer. Not answered: the number who skipped the question. Guests who chose "Not applicable" count in neither.
- An answer's % is its share of Answered.
- Conversion rate: Surveys submitted divided by surveys opened in your timeframe. We count the first open and the first submit of each survey. A survey opened just before your timeframe and submitted inside it still counts as submitted, so the rate can go slightly above 100%.
- Email funnel: Follows the survey emails sent in your timeframe through five stages: Email sent, Email delivered, Email opened, Survey opened, and Survey submitted. Each stage shows as a count, a % of the previous stage, or both. Because it follows the emails sent in your timeframe, its Survey opened and Survey submitted numbers can differ from Conversion rate.
Survey Integrity
Integrity metrics use the same labels as the Integrity Status in your Inbox: Authentic, Inconclusive, Suspicious, and Pending.
- Entity Integrity Score: The share of a property's scored survey reviews that pass our integrity check, times 100. A score of 100 means every scored review looks authentic.
- Authentic %, Inconclusive %, Suspicious %: The share of scored survey reviews with each label. The three add up to 100%.
- Pending %: The share of all survey reviews still waiting to be scored. A review stays Pending for its first 31 days, then gets its label automatically.
- Scored reviews: The number of survey reviews that already have a label.
Note: To learn what each label means, see Understanding survey integrity status.
BenchIndex
BenchIndex compares the two sides of a Benchmark: Your set (your own Entities) and the Comparison set (the group you compare them with, which can be your own Entities, competitors, or both). It's the Your set figure divided by the Comparison set figure. You'll see it when your rows are Organization benchmarks or Personal benchmarks.
- Above 1: Your set is ahead. For example, 1.05 means it's 5% above the Comparison set.
- Below 1: Your set is behind. For example, 0.96 means it's 4% below the Comparison set.
BenchIndex is available for these KPIs:
- RPS BenchIndex: Each side's RPS is the average of all its review scores, so Entities with more reviews count for more.
- Reviews BenchIndex: Compares the average number of Reviews per Entity on each side, so a bigger group doesn't win just for being bigger. Only Entities with at least one review in your date range and filters count. Reviews include TrustYou Survey reviews.
- Response Rate BenchIndex: Each side's Response Rate is its responded Reviews divided by its respondable Reviews.
- TrustYou Score BenchIndex: Each side's TrustYou Score is the simple average of each Entity's latest score in your date range.
- Source Score BenchIndex: Each side's Source Score is the simple average of each Entity's latest Source Score for that source, updated within the last year.
- Sentiment Score BenchIndex: Each side's Sentiment Score is its positive mentions divided by all its mentions.
When you break a Benchmark down by a row sub-group:
- Brand: Each brand in Your set is compared with the whole Comparison set (the Totals row figure).
- Any other sub-group (like Country, Source, or Sentiment category): Each row compares its own Your set with its own Comparison set. For example, Your set in Germany with the Comparison set in Germany.
- Totals row: The Your set total divided by the Comparison set total.
Note: BenchIndex is blank when one side has no figure, for example no reviews in your date range. A BenchIndex of 0.00 is a real result: Your set's figure was 0 while the Comparison set's was above 0. Benchmarks in Report builder are available on Multi-Entity accounts that include Benchmarks.