Skip to main content

MULTILINGUAL & COMMUNITY MARKETING · September 2026 · ~11 min read

Marketing to first generation versus second generation customers

They often find you through completely different doors and judge you on different things. One may arrive through a personal recommendation and decide based on whether the product is right. The other may arrive through search and reviews and decide before ever walking in. You need both doors open.

A warning before any of this is useful. First generation and second generation are rough shorthand, and they leak badly. Someone who arrived at five is not the same as someone who arrived at forty. Plenty of people born here speak the language at home and prefer it. Plenty who immigrated as adults do all their searching in English.

Treat what follows as a hypothesis to check against your own customers, not a description of anybody.

01

Is there actually evidence for this split?

Partly, and being precise about which part matters more than the framework.

There is real primary evidence for a language gradient by age. The Census Bureau's December 2023 release of the 2018 to 2022 American Community Survey five-year estimates reports that among people who spoke a language other than English at home, 61.0% of Spanish speakers spoke English "very well." Broken out by age, that figure was 58.3% among Spanish speakers aged 18 to 64 and 41.6% among those 65 and older. Across all non-English speakers, 48.2% of Chinese speakers and 69.8% of Tagalog speakers spoke English "very well." Nationally, 78.3% of the population age five and older spoke only English at home, leaving 21.7% who did not.

There is no published evidence at all for a search-behaviour split by generation, and you should be suspicious of anyone who presents one. No dataset I would cite segments local search, call rates or review-writing by immigrant generation. Google publishes search volume by no such dimension. The Census measures language spoken at home and English proficiency, which is a proxy for language, not for which app someone opens.

So the honest position is this. The age and proficiency gradient is documented. The channel behaviour is a pattern I have watched at a host stand for sixteen years and it is worth testing, but it is not a fact about your customers until you count them. An agency persona deck that assigns discovery channels to ethnic generations has invented the mapping.

02

What actually differs, if not the people?

Three practical things: how they find you, how they contact you, and what they are measuring you against.

Discovery. A customer with an established network here often hears about you from a person. A customer without that network, or one who simply grew up searching for everything, finds you on a map. Neither is more loyal. They just enter through different doors.

Contact. Some customers call. Some will never call and will leave if the only option is a phone number. This one is worth testing rather than assuming, and there is a documented trend making it urgent. Sterling Sky's "State of Local SEO in 2026," using third-party data from Jepto compiled across 179 Google Business Profiles at 34 US law firms over two years, reports that clicks-to-call from Business Profiles are declining even for profiles whose rankings held steady, while website clicks did not fall the same way. Joy Hawkins attributes the difference to the surface: website icons appear on desktop while call icons dominate mobile, so the loss is mobile-specific and it is the call button. The same report notes that the AI-powered local packs now appearing on roughly 7% of tracked keywords show one or two businesses instead of three and carry no call buttons at all.

The comparison. This is the interesting one. A customer who grew up with your product may be comparing you to a remembered version of it. A customer discovering it may be comparing you to the last three places they tried. Those are different standards and you can satisfy both, but not by accident.

Both of those customers may be at the same table. That is the part most marketing misses entirely.

03

Who books and who decides?

Frequently different people, and this changes everything about where you spend.

Standing at a host stand for years, the pattern I saw constantly was a party where the adult child made the reservation, drove, read the reviews beforehand, and handled the check, while a parent decided whether the food was right. Both were customers. Only one of them was ever going to see a Google review.

If that describes any of your business, your search visibility and your product quality are aimed at two different people. Your reviews and your photos have to satisfy the person who books. What is actually on the plate has to satisfy the person who decides.

Get one right and lose the other and you get a specific outcome: good reviews and no repeat visits, or full tables of people who never write anything.

This is also why review count alone tells you so little. A steady flow of recent reviews matters more than a large historical total, especially when only part of your customer base writes them at all. In Whitespark's 2026 Local Search Ranking Factors survey, in which 47 local search practitioners scored 187 factors, Recency of Reviews ranks eleventh and "Sustained Influx of Reviews Over Time (rather than bursts)" ranks fourteenth. Quantity of reviews with text ranks ninth. Recency and steadiness are scored as separate things from volume, and a business whose review writers are all in one half of its customer base will show a thin, lumpy flow no matter how many total reviews it has accumulated.

04

How do I run the two-door audit?

Worked example

Count sixty days of new customers by how they arrived, then test each door for a blocker. Here is the arithmetic.

Step one, log the entry. Over sixty days you log 148 new customers. Ninety-one found you in search, 44 came on a personal recommendation, and 13 walked past. That is 61%derived through the search door and 30%derived through the referral door.

Step two, test the search door. Can a customer complete a booking or an order without picking up the phone. If the answer is no, you are relying entirely on a call button that Sterling Sky's data says is being clicked less on mobile every year, and that does not exist at all in an AI local pack.

Step three, test the referral door, which almost nobody tests. A referral arrives as a name said out loud. The customer then types what they heard. If your profile and website carry one spelling and your customers say another, some of those people never arrive and you never learn about it. Ask the 44 who came by recommendation whether they had trouble finding you. Say seven of them did. That is 16%derived of your referral door leaking, on a door carrying nearly a third of your new business.

Step four, price the fix. Adding every spelling your customers use to your Business Profile services and products and to your menu page is an hour of work and no cash. Seven customers over sixty days is 42 a year, which for most operators is a larger number than anything an equivalent hour of paid advertising would buy.

Two honest caveats. Self-reported attribution undercounts, so 44 is a floor rather than a measurement. And the seven who told you they had trouble are the ones who arrived anyway. The ones who gave up are not in your log at all, which is why the fix is worth more than the count suggests.

05

Do I need separate marketing for each?

No. You need one business with more than one entry point.

Running two brand voices at the same business is expensive and it reads as insincere to both groups. What you actually need is coverage:

1. Your Google Business Profile and reviews in good order, because that is the door for anyone searching. 2. Real presence in the networks where recommendations move, because that is the door for anyone asking a person. 3. Both languages available where the decision happens, so neither customer hits a wall. 4. A phone that gets answered and a booking option that works without one.

That is not two marketing programs. It is one program with no missing pieces.

Where you genuinely should differentiate is channel spend. Local community press and in-person events still reach people that search does not, and whether community newspapers move business for you depends heavily on which part of your customer base you are trying to reach. Test it with a tracking number rather than an assumption, in either direction.

06

What about the product itself?

Here is the harder question, and it is not really a marketing one.

Businesses serving both groups sometimes feel pressure to adjust the product toward whichever customer is easier to acquire. That is a real strategic decision with real consequences, and it should be made deliberately rather than by drift.

The version that tends to work is being clear about what you are, then being findable to everyone who wants that. The version that tends to fail is quietly changing the product to widen the audience, which reliably loses the customers who were already loyal without winning the new ones.

If you do change something, say so. Menu and service changes should be reflected everywhere a customer might check, and handling seasonal or menu changes on your website is a small discipline that prevents a lot of disappointment at the door.

07

How do I find out which is which in my business?

Look at what you already have, then ask.

Count your reviews by language. Count your calls versus your online bookings. Look at what your reviews praise, because different groups tend to mention different things, and the pattern will be visible once you read fifty of them in a row.

Then ask ten customers directly how they heard about you. Not a survey. A question at the counter, logged somewhere.

This is the beginning of a measurement habit rather than a one-time exercise, and figuring out whether multilingual marketing is actually working requires the same small set of numbers tracked over time.

Where the two groups usually overlap is in person. An event brings both, often together, which makes it one of the few places you can watch the difference directly. Cultural events can work as a real acquisition channel partly for that reason.

08

What to do this week

Count your reviews by language and count your last fifty bookings by channel. Write both down and date them.

Read fifty reviews in a row and note what each one praises. Look for two different patterns rather than one average.

Make sure a customer can complete a booking without a phone call, and make sure a customer can reach a human by phone. Both. Most businesses are missing one, and the call button is the half that is quietly eroding.

Ask ten people how they heard about you, and ask the ones who were referred whether they had trouble finding you.

Be honest with yourself

When you do not need this

If your customer base is genuinely one group, do not manufacture a second program. Serve the customers you have well.

If you are already covered on both doors, search and referral, leave it alone. Adding complexity here has a real cost in maintenance.

If your Business Profile is missing a primary category or has the wrong hours, none of this is your problem yet. Primary category scored 227 points and proximity 225 in that same practitioner survey, and no amount of audience segmentation compensates for a profile that does not describe what you sell.

And if you find yourself building customer personas around ethnicity rather than around behavior, stop. There is no published data mapping generation to channel, so anything you build that way is invention. The useful distinctions are how people find you and what they need from you. Everything else is guessing about strangers, and customers can feel it in the copy.

Sources

Related reading

12

Questions about who your customers actually are?

Email me at eric@seod.com with your business name and I will read through your reviews and tell you what patterns I see in who is writing them, in which language, and about what. If you send your sixty-day entry log I will run the two-door audit above on it. It is a read of evidence you already own but probably have not sat down with.

I answer these personally. If your customer base turns out to be simpler than you thought, that is a useful answer too.

Or keep reading more on multilingual and community marketing.

Call Eric Email Eric