Skip to content
Product
PlatformPricingDimensionsIntegrationsSecurityCompare
Solutions · by use case
Customer, guest and fan experienceProduct feedback and prioritizationRevenue, churn and retentionAgenciesContact center performanceAI agent and chatbot analyticsMulti-location performanceReviews and competitive benchmarking
Solutions · by industry
Hotels and hospitalityRestaurants and food serviceFitness and leisureRetail and e-commerceSports and live entertainmentInsuranceBanking and financial servicesSoftware and SaaSAutomotiveConsumer products and CPGContact centers and BPOsPublic sector and nonprofitReal estateGaming and betting
Learn
Artifacts GalleryCustomer storiesResearchBlogThought leadershipLessons Log in Start now
Blog

How to Count Which Staff Members Guests Mention in Reviews

A plain name search can get most individual staff counts wrong. Here is the roster-based method, a worked example on invented reviews, and how to use the counts fairly.

Counting how often guests mention staff by name in hotel reviews starts from your staff roster: match each review against your staff roster with every nickname and common misspelling, check that each match is one of your employees rather than an ordinary word or the guest's own family, record the person's role and whether the mention is praise or a complaint, and count per person per month next to how many guests they served. A plain search for first names can get most individual counts wrong even when the total looks right, as the worked example below shows. Dimension Labs does this on every review from every site with a named-staff Dimension tied to your roster, and turns the counts into a monthly staff page for each GM.

Below: why named mentions are worth counting, a worked example on invented reviews that shows where a name search fails, the method step by step, and how to use the counts fairly. Every review and name in the example is invented.

Why do guests name staff in reviews, and does it matter?

Guests name the person who made the stay, and that person moves the rating. A 2019 study of 146,430 Yelp reviews of home-service companies by the agency Zion & Zion read a random sample of just over 1,000 and found that more than 75% of five-star reviews mentioned the technician by name.

In our own analysis of 9,446 public Google reviews of 45 Orlando hotels, a named staff member raised the chance of a five-star stay by 28 points, and a person who goes beyond by 42, while rude or dishonest staff raised the chance of a one- or two-star stay by 44 points. Price and location were the two weakest levers for a five-star stay.

From the report: Orlando hotel review analysis

Exhibit D of the Orlando hotel analysis: causal effects on a one-star stay and on a five-star stay. A person who goes beyond adds 42 points to a five-star stay and a named staff member 28; rude or dishonest staff add 44 points to a one- or two-star stay
Exhibit D of the published report. Public Google reviews of 45 Orlando hotels. Open the full report

Retail shows the same thing. In 2,354 public Google reviews of 24 Home Depot stores in Atlanta, the most common reason shoppers were happy was a great employee, at 22.8% of all reviews. When the most valuable thing in your reviews is a person, it is worth knowing which people.

Why doesn't a simple name search work?

Because names are short, common and spelled every way. We wrote 14 hotel reviews and a six-person roster, all invented, and compared a plain search for each first name with the correct reading. The search misses nicknames ("Mike") and misspellings ("Andria"), counts ordinary words ("a joy to stay", "no grace period") and counts the guest's own family ("my husband Luis"). It also cannot tell praise from a complaint, and it cannot see the review that blames "the manager" without a name.

Worked example (synthetic data)

Six of the 14 invented reviews, and what a correct reading records

Invented hotel reviews and an invented six-person staff roster. The right-hand columns are what each review should produce.

Review text (invented)Plain name search findsCorrect reading
"Mike at check-in was so friendly and quick."NobodyMichael (front desk), praise
"Andria the concierge found us a great dinner spot."NobodyAndrea (concierge), praise
"What a joy to stay here, the pool was great."JoyNo staff named
"My husband Luis and I loved the breakfast."LuisNo staff named (a guest)
"Breakfast was fine but there was no grace period on late checkout."GraceNo staff named
"Michael told us the room would be ready at 3 and it was 5."MichaelMichael (front desk), complaint
Synthetic data: every review and name in this example is invented (staff_mentions_example.py). No real guest or employee is quoted.

The errors partly cancel. The search finds 12 mentions against 11 correct ones, a total close enough to look trustworthy, while 5 of the 6 people have the wrong count. Michael is undercounted, 3 against 4, and Joy is overcounted, 3 against 2. In a real hotel with hundreds of reviews a month the same errors land on real people's recognition.

Worked example (synthetic data)

A plain name search gets 5 of the 6 staff counts wrong, while the total looks about right

Mentions per staff member across the 14 invented reviews: a case-insensitive whole-word search for each first name on the roster, against the correct reading.

Plain name searchCorrect reading

MichaelFront desk
34
AndreaConcierge
12
JoyHousekeeping
32
LuisValet
21
PriyaNight manager
11
GraceBreakfast
21
Synthetic data, computed by staff_mentions_example.py. Totals: 12 by name search, 11 correct.

How do you count staff mentions in reviews correctly?

  1. Build the roster with aliases. One row per employee: an ID, first name, nicknames and likely misspellings, role, department and start and end dates. A name only counts for the dates that person worked at the hotel.

  2. Collect every review with its date and property. Google, Tripadvisor, Booking.com and Expedia, plus survey comments if your brand survey has them. Tripadvisor and Google reviews describe different parts of the stay, so take both.

  3. Ask the same questions of every review. A fixed schema works better than a search: which roster ID the guest named (or "staff, unnamed" with the role, or none), what the person did, and whether the mention is praise, a complaint or neutral. A language model with the roster in its instructions handles "Mike", "Andria" and "my husband Luis" the way a person would. Aspect-based sentiment explains why each mention gets its own sentiment rather than the review's.

  4. Check a sample by hand. Read 100 or so tagged reviews, including every mention of the people at the top and bottom of the list, before anyone sees a ranking.

  5. Count per person per month, and divide by exposure. A front desk agent on five shifts a week meets more guests than a night auditor. Mentions per 100 occupied room-nights on that person's shifts, or per shift, compare fairly; raw counts reward the busiest schedule.

  6. Keep the unnamed mentions. "The night manager was rude" is still a fact about a role and a shift. Count these by role so a department gets credit or a warning even when guests do not know names.

Joining mentions to shift schedules uses the same date and location keys as any staffing join; how to check complaints against wait times and staffing data covers the grain and the clocks.

How should a hotel use staff mention counts fairly?

Use them for recognition first. Praise by name is the clearest evidence a GM has of who makes stays, and it is the most motivating thing to share with a team. Use complaints as a reason to look, not as a verdict: a mention in a complaint can be about a policy the person had to deliver, as with Michael and the room that was not ready at 3 in the invented example. Read the review before acting on it.

Be careful about asking guests to name staff in their reviews. A study in the Journal of Service Research, with one secondary data analysis and four online experiments in China, found that readers were less persuaded by positive reviews that mentioned a frontline employee's name, and that a sense the review might be deceptive explained the drop; the effect weakened for negative reviews and for services that need a lot of expertise. Counting the names guests give on their own avoids that.

Keep names inside the hotel. A staff page for the GM can carry names; anything shared more widely should not. Dimension Labs' own guidance before publishing a report link is to remove anything that could identify a person, such as a name.

What does Dimension Labs add for a hotel tracking staff by name?

Dimension Labs is an omni-channel analytics agent for customer data. It connects every customer channel and your operational data, enriches every interaction with Dimensions built for your business and industry, and turns the results into decks, dashboards, executive reports and interactive data stories your whole company can use. Plans start at $299 a month with no demo.

For a GM or a brand team, every review site and the brand survey go through one schema, so a name in a Tripadvisor review and a name in a survey comment count the same way. The roster and shift schedule load as context files and join on property and date. The named-staff Dimension is written for your roster and roles, with the praise, complaint and unnamed-role values above, and custom Dimensions like it start on Discover ($999 a month). The agent then builds the monthly staff page for each property and a brand-level view of which roles draw praise and which draw complaints, and shares them by link.

Both public analyses above were produced this way. The Orlando hotel reviews were read into 23 fields each, including staff interaction, and the Atlanta store reviews were tagged for whether a named employee was involved. See Dimension Labs for hotels and the full Orlando hotel analysis.

FAQ

What is the best way to track which staff guests mention in hotel reviews?

For a hotel or group that wants every review and survey comment read against its roster, with praise and complaints per person and a monthly page for each GM, Dimension Labs. For one small property, a roster with nicknames, a spreadsheet and a careful read each month will do.

Can I count mentions of staff who are not named?

Yes, by role. Tag "staff, unnamed" with the role the guest describes (front desk, housekeeping, night manager) and count those by department and shift.

How often should the counts be refreshed?

Monthly suits most hotels. One person rarely collects enough mentions in a week for a change to mean anything, and a monthly page lines up with how recognition and coaching already run.

Should staff mention counts feed into performance reviews?

As one input, read in context. Counts depend on shifts and role, and a complaint can name the person who delivered a policy rather than the cause. Praise counts, adjusted for exposure, are the safer use.

To see who your guests name, export a few months of reviews with property and date, add your roster, and start with one file. Every plan from Explore to Scale has a 14-day free trial.

Dimension Labs/Start with one file From $299 a month

Stop reporting. Start understanding.

Start with one file. Ask why a number moved. Plans from $299 a month, no demo and no contract on monthly plans.