Reviews in multiple languages and what they signal

Your review profile is the cheapest research you own. Nobody was surveyed, nobody was incentivized, and nobody had a reason to perform. A customer picked a language and wrote in it. That choice carries information.

The businesses that read this signal make better decisions about staffing, menus, and where to spend. The ones that do not are guessing about their own customers.

What does the language of a review actually tell me?

Three separate things, and they are worth keeping apart.

Who is choosing you. If a third of your reviews arrive in Spanish, you have a Spanish-speaking customer base whether or not you ever marketed to one. That is a fact about your business, not an aspiration.

Who felt comfortable enough to write. A customer writes in the language they associate with the experience. If someone spoke Korean with your staff and had a good visit, the review often comes back in Korean. If your reviews are all English but your dining room is not, something is defaulting to English that maybe should not.

What is missing. The absence of a language is data too. A neighborhood full of Spanish-speaking households and zero Spanish reviews usually means you have never asked in Spanish, not that those customers do not exist.

The national background is real without being a substitute for your own count. In the Census Bureau's December 2023 release of the 2018 to 2022 American Community Survey five-year estimates, 21.7% of the US population age five and older spoke a language other than English at home, and among those speakers, 61.0% of Spanish speakers and 48.2% of Chinese speakers reported speaking English "very well." Fewer than half of Chinese speakers did. That is a description of the country, not of your customers. Your reviews are the description of your customers.

How do I count the split without guessing?

Do reviews in another language help my rankings?

New reviews help. Whether the language of a review helps is not something anybody has established, and you should be suspicious of anyone who tells you otherwise.

What is documented is recency and steadiness. In Whitespark's 2026 Local Search Ranking Factors survey, in which 47 local search practitioners scored 187 factors, quantity of native Google reviews with text ranked ninth at 170 points, recency of reviews ranked eleventh at 164, and sustained influx of reviews over time rather than bursts ranked fourteenth at 154. Darren Shaw, who runs that report, puts the mechanism plainly: "The moment you stop getting new reviews, you're going to see your local rankings start to slip." He also notes that "getting a negative review is actually better than getting no new reviews at all," because the recency signal does not care about sentiment.

That survey is expert opinion rather than test data, and Shaw says so himself. Hold it loosely.

Now the claim to refuse, because it is the one that gets sold into bilingual businesses. The pitch is that you should ask your Korean-speaking customers to write in Korean so you show up for Korean searches, and sometimes that you should ask them to include a dish name while they are at it.

Two problems, and the first is policy. Google's Maps user generated content policy states that when soliciting reviews, merchants "should not require or pressure users to leave ratings or write reviews while on the premises, nor should they request that specific content be included." Asking for a specific language is a request about content.

The second is evidence. Sterling Sky ran the closest thing to a controlled test that exists on review wording. Joy Hawkins used a Christmas tree farm with, in her words, no SEO efforts and no map pack SEO, tested in the middle of summer so no organic reviews would arrive, and had six people leave keyword-containing reviews against a base of 18. Her result, verbatim: "adding keywords to reviews did not help this business' local pack ranking on Maps... Actually, in the first instance, rankings got worse." Her recommendation, also verbatim: "Don't coach your clients to include certain words in their reviews."

That test was about keywords rather than languages, so it does not settle the language question. Nothing settles the language question. No published test establishes that the language of a review moves local rankings in either direction, and I would rather tell you that than pick the answer that sells more work. Ask honestly, in the language of the conversation you just had, and let people write what they want. That is compliant, it is what Google explicitly permits, and it produces the recency that is documented.

How do I ask in a second language without it being awkward?

Ask in the language you have been speaking, at the moment the visit ends, from the person who served them.

That is the entire method. The awkwardness people worry about comes from asking in a language the customer did not choose, or from a card that clearly was not made for them.

Practical version:

1. Print or text the request in both languages. One card, two versions, no decision required from the customer. 2. Have the staff member who actually spoke with them do the asking. 3. Ask at the end of the visit, not three weeks later. 4. Make the link go straight to the review form. Every extra tap costs completions.

Google's own permission is worth knowing, because it draws the line for you. Merchants may "solicit or encourage the posting of content that does represent a genuine experience, without offering incentives to do so or attempting to influence the rating or the contents of the review." Ask everyone, in whichever language fits, offer nothing, script nothing.

Set the pace against your competitors rather than against a target you invented. Shaw's benchmark: "How often are your top competitors getting new reviews? If the answer is twice a month, you should also aim for that, plus one."

If your staff cannot ask in the second language, you have found the constraint. It is a staffing question, and it is usually the reason a business serving two languages has reviews in only one.

Nothing about this changes if you also need the basics fixed. A profile without correct categories, hours, and photos will underperform in every language, and a complete Google Business Profile setup for a restaurant is the floor before any of this is worth doing.

What do I do with reviews I cannot read?

Read them, or hire someone who can. Those are the options.

A machine translation is fine for getting the gist of an incoming review. It is fine for exactly that and nothing else. It is not fine for writing your public response, because a response that reads as automated does more damage than silence.

The response matters more than most owners think, because it is read by future customers rather than by the person who complained. That is true in any language, and the template response that makes a bad review worse is even more obvious when it has clearly been run through a tool.

Do not expect the response to move rankings. Keywords in owner responses to reviews scored 28 out of a possible 235 in the same Whitespark survey, which places it among the factors the report treats as doing essentially nothing. Respond because a stranger is reading it, not because you think Google is.

There is a related trap on the website side. Some businesses paste machine-translated review text onto their own pages as social proof. That produces thin duplicated content in a language nobody wrote, and translated content can work against your rankings rather than for them.

If you are going to collect reviews in a language, commit to reading them. Otherwise you are collecting complaints you will never hear.

What does a healthy multilingual review profile look like?

It roughly matches your dining room or your waiting room.

Not exactly. Some groups review more than others and always will. But if your business runs half in Spanish and your reviews are almost entirely English, the gap is your asking process, not your customers.

Watch the trend rather than the total. New reviews in each language over the last ninety days tells you whether your system is running. The lifetime count tells you what happened years ago.

And notice what the reviews reference. If the Korean reviews all mention one dish and the English reviews all mention the room, you are learning two different things about why people come. That is worth more than a star average.

This is also where reviews connect to everything else you do in a community. Reviews follow relationships, and relationships come from being present rather than from advertising, which is the practical difference between community marketing and running ads. The same goes for coverage in local ethnic press, which still produces real word of mouth in some categories, and community newspapers deserve an honest evaluation rather than a blanket dismissal.

What to do this week

Run the four-number count above. Lifetime split, room split, 90-day split, and the gap. Write them down with today's date.

Make a bilingual review card or text template this week. Same link, both languages, one piece of paper. Say nothing about what the review should contain.

Ask your next twenty customers in the language you served them in, and count what comes back in thirty days.

Check your competitors' review pace and set yours one above it.

Then respond to every review you have from the last month, each in the language it was written in, written by a person.

Questions about your review mix?

Email me at eric@seod.com with the review request you currently send and I will write you the bilingual version, both languages on one card, plus the timing I would use for your business. Korean and Spanish I write myself.

I answer these personally. If your current request is already doing the job, I will say so and tell you where I would put the effort instead.

Or keep reading more on multilingual and community marketing.