Benchmarks

Where Diners Actually Find You in 2026: Maps, TikTok, or Word of Mouth?

Word of mouth is still the single largest US discovery channel at 38%, but for Gen Z, TikTok has overtaken it as the top way they find a new restaurant. A channel-priority framework by guest type, built from Toast and CGA survey data, not a "be everywhere" listicle.

Alex Riesenkampff

August 12, 2026 · 10 min read · Markdown

Word of mouth is still the single largest way US diners discover a new restaurant, at 38%, ahead of simply passing by (30%) and Facebook (27%). But that topline number hides the fact that matters most for a small venue deciding where to spend its limited marketing time: for Gen Z specifically, TikTok has already overtaken word of mouth as their top discovery channel, at 38%, while the same platform barely registers for Gen X and Boomers. Toast's October 2025 survey of 1,466 US adults is the source, and the generational split is the whole story: there is no single best channel, only the channel that matches who already walks through your door.

This piece builds a channel-priority framework from that survey and a UK companion study, one built for whoever runs a cafe, a bar, a bakery, or a full-service restaurant with limited weekly time to spend on marketing. The framework changes with your guests, not your menu.

What actually gets people through the door

Across all US diners, word of mouth (38%), passing by in person (30%), and Facebook (27%) are the three channels doing the most work, well ahead of search engines, Instagram, YouTube, and TikTok individually. Toast's survey, fielded 9-15 October 2025 across 1,466 US adults, is the clearest single dataset available on this question, and it is worth seeing the full breakdown before any single platform gets outsized attention.

Two things are worth naming honestly. First, this is US data; a UK companion survey from CGA by NIQ (755 consumers, August 2025) separately reported social media usage at 32% and Google Maps usage at 27% among British diners, a different market with its own shape, not a contradiction of the US figures. Second, no survey we could find breaks discovery down by venue type, so resist the urge to assume a bakery's numbers look like a bar's. Use the shape of the data, not the exact percentages, if your venue and market differ from the one surveyed.

There's a practical reason the order matters beyond curiosity: word of mouth, in-person visibility, and a Google Business Profile cost nothing to maintain, while Facebook, Instagram, and TikTok all compete for the same scarce hours an owner or a single staff member has available for content. DoorDash's own merchant guidance makes the same point from a different angle, telling independent operators they are "best served by working through free and low-cost options before committing budget to paid marketing." The topline chart above is really a ranked case for where those free hours pay off first, before any paid or platform-specific push gets added on top; our Google Maps ranking playbook covers the profile side of that in full once you've decided it's worth the time, and our guide to how many reviews you actually need is the natural next stop once reviews are the channel you're prioritising.

For Gen Z, TikTok has overtaken word of mouth

Toast's generational breakdown is the single most useful fact in this dataset: TikTok drives 38% of Gen Z restaurant discovery, overtaking every other channel including word of mouth for that cohort, while the platform's influence "drops off sharply" for Gen X and Boomers, who still lean on the traditional channels above. This is the direct evidence against treating TikTok as a universal must-do, and equally against writing it off entirely.

Ryan Jacobsen, SVP of Growth Marketing at LT.agency, frames the broader sequencing logic that follows from a generational split like this one: he recommends restaurants "start with Google Ads as the ultimate 'pull channel' to capture high-intent searches," and only move into "push channels like Meta or creator collaborations" once that pull channel is fully used. The principle generalises past paid search: capture the guests already looking for you on the channel that already works for their generation, before spending effort trying to manufacture attention from guests who weren't looking in the first place.

Going viral is not the same as being findable

A viral moment can fill a room with people who came to watch, not to eat, and that is a real cost, not a marketing win, if it drives away the regulars a small venue actually depends on. Folderol, a small independent wine bar and ice cream shop in Paris, went viral on TikTok in 2023 for its aesthetic; the resulting crowds came specifically to film rather than to order, disrupted service, and pushed away the venue's actual regulars. The owners' response was not more content, it was a "no TikTok" sign, restricted outdoor seating, and a bouncer to manage the crowd.

Folderol was not an isolated case. Ai Monti Lattari, a sandwich shop in Naples, went through a similar cycle after a staff member's TikTok brought disruptive tourist crowds rather than sustained local business, and the owners eventually asked that employee to stop posting. Neither venue set out to attract a crowd that came to watch rather than to eat; both discovered that a platform's reach and a venue's actual capacity to serve well are two different things, and that the gap between them shows up as worse service for the regulars who were there first.

Neither story argues for chasing virality as a strategy. Read together, they argue for something narrower and more useful: content works when it reaches people who were already a plausible match for the venue, and backfires when it reaches people who were only ever there for the clip.

A channel-priority framework by guest type

The two variables in Toast's own data that predict which channel is worth an owner's time are guest age and how urban the venue's location is, so build the framework from those two axes rather than guessing. Toast found urban diners lean toward social discovery, while suburban diners rely more on word of mouth and physical visibility; layer that against the generational TikTok split above and a usable starting matrix emerges.

The matrix is built from Toast's own generational and urban/suburban findings, not from a separate venue-type survey, so treat each cell as a reasoned starting point, not a guarantee: confirm it against your own guests before committing real weekly time to it.

One 15-minute task per channel, sustainably

Owned channels reward small, repeatable effort more reliably than a single big campaign, so the goal is a task you can actually keep doing every week, not a burst you'll abandon in a month.

What the data doesn't tell you, and shouldn't guess at

No survey we found breaks discovery channels down by venue type, so a claim like "bakeries do better on Instagram than bars" would be invented, not evidenced, and this piece will not make it. Where the evidence runs out, say so rather than filling the gap with something that merely sounds plausible. If the gap you're actually fighting is a single slow night rather than overall visibility, our playbook for fixing a dead weeknight walks through the narrower, cheaper fix; if the open question is whether a website still earns its keep next to a free Google profile, we've run those numbers too.

The honest default for most independent venues, whatever the matrix above suggests, is this: fix your Google Business Profile and cultivate word of mouth first, because both compound regardless of who your guests are, and both are already ahead of TikTok in Toast's own topline numbers. Add a second channel only once the first two are genuinely working, and only when your own guest data, not a trend, points you there. The exception is a venue whose regulars are already visibly young and urban: for that specific case, treating short-form content as core rather than optional is the evidence-backed call, not a hedge. Either way, watch for the signal that should reverse a decision: if a channel starts pulling in more lookers than guests, or if content creation is visibly eating into service quality or staff time, that is the moment to step back, not to post more.

Filming, editing, and posting content is real labour, whether it's the owner doing it after close or a staff member pulled off the floor during a shift, and it competes directly with the work that actually keeps existing regulars happy. A venue that adds a content routine it can't sustain past the first few weeks has spent real hours for a channel that will quietly go stale, which is worse than never having started it. Pick the one channel the framework above actually points to, commit to the 15-minute weekly version of it, and only expand once that habit is genuinely holding.

Keeping track of which channel is actually converting into covers, not just views, is the harder half of this problem for a solo owner without a marketing team. Super44 connects to a venue's POS and Google reviews so an owner can ask, in plain language, what a given week's revenue pattern actually looks like and where the reviews that shape a guest's first impression need a reply, rather than reconstructing the picture by hand.