Who the Machine Names for the Local Bike Shop: Share of Answer, the Map Pack, and the Service Moat

Introduction

A bicycle is not a bar of soap. It is a specialty durable good, chosen after consideration, fitted to a body, and then kept running for years through a relationship with whoever holds the wrenches. That second half, the service attach, is what sets the bike category apart from most of retail. A shopper does not buy a bike once; they come back for a flat, a tune up, a new chain, a fitting, a warranty question. For the independent local bike shop, abbreviated in the trade to LBS, that recurring service counter is the economic heart of the business and, as this study argues, the engine of its digital visibility.

The competitive terrain is specific to this buying mode. On one side sit the brand experience stores (Trek and Specialized concept stores), the outdoor chains (REI, Scheels), the mass merchants (Dick's Sporting Goods, Walmart, Target), and a wall of online sellers from Amazon to specialist etailers such as Jenson USA and Competitive Cyclist. On the other sits the neighborhood shop, whose advantage is not price or catalog depth but proximity, expertise, and the service bay. The question this study holds constant across the RavenEye Retail Visibility Study series is plain: when a shopper looks for this product, who does the machine name, and where do independent local shops actually stand?

This is original measurement, not a client case study. Raveneye Global has no bicycle-retail clients and names none here; every shop mentioned is treated as an anonymized, publicly observable fact. On 2026-07-22 we captured 40 Google result sets and five Perplexity answers across five metros, read real US search volumes, and audited a small sample of independent shop websites for their machine readiness. Every figure below is dated to its capture and tagged with an evidence tier. Where an engine did not return an answer, we say so rather than estimate one.

Background and literature

Economic theory has long sorted goods by how a buyer can verify quality. Nelson (1970) separated search goods, whose quality can be judged before purchase, from experience goods, judged only after use. Darby and Karni (1973) added credence goods, whose quality a buyer struggles to judge even after the fact, which is exactly the position of a novice cyclist weighing a fit, a repair, or a component recommendation. The bike sale itself is a search good, since specifications are public, but the service and expertise wrapped around it behave as experience and credence goods. That gap is why a trusted local intermediary keeps the value the internet was supposed to erase, and why the signals that certify trust, chiefly reviews, carry so much weight in this category.

The product is only half of what the shop sells. Vandermerwe and Rada (1988), who coined the term servitization, described the modern firm as one that adds value by offering bundles of customer focused goods, services, support, and knowledge rather than a bare product. The independent bike shop is a servitized retailer by nature: the sale, the fit, the assembly, the tune up, and the standing advice arrive as one bundle. Betancourt and Gautschi (1988) formalize the other half of this, modeling a retailer's true output not as the object on the shelf but as a set of distribution services (assortment, information, assurance, ambience, and, here, hands on repair) that shift cost and effort off the buyer. An online marketplace competes by stripping those services out to lower the price; the specialty shop competes by selling them as the point. This is the structural reason the two channels rarely answer the same question.

The aftermarket is where both the margin and the moat sit. Cohen, Agrawal, and Agrawal (2006), studying durable goods, found that spare parts and after sales service routinely earn higher margins than the original sale and bind the customer into a relationship a one time transaction cannot. For the bike shop the service bay is that aftermarket in miniature, and its economics run the same way: the first sale may be thin, but the years of repairs, replacements, and fittings that follow are not. The point that matters for this study is that the moat and the visibility signal are the same asset. Every completed service is both recurring revenue and an occasion to earn the recent, local, human review that the map pack reads, so a shop that runs its service counter well is compounding its machine visibility as a byproduct of its core economics.

Reviews are not decoration. Luca (2016), studying Yelp, found that a one star increase in a business rating drove a measurable revenue increase for independent restaurants, an effect concentrated among independents rather than chains whose reputation is already established. The logic transfers cleanly to bike shops: the independent with no national brand to lean on lives or dies by its local review profile, and that profile is precisely what the machines read.

How buyers reach an answer is described by information foraging theory (Pirolli and Card, 1999), which models searchers as foragers following the strongest scent of value for the least effort. When a search engine or an answer engine collapses a decision into a ranked three pack or a short written list, it is doing the foraging on the buyer's behalf, and the businesses outside that shortlist are effectively invisible. Choice overload (Iyengar and Lepper, 2000) sharpens the stakes: faced with too many options, buyers defer to whatever curated set is presented, so being in the machine's shortlist matters far more than being listed somewhere on the page.

Two well documented retail dynamics bracket the category. Showrooming and its inverse, webrooming (Gensler, Neslin, and Verhoef, 2017), describe buyers who inspect in one channel and buy in another, a constant tension for a shop whose fitting expertise can be captured by an online seller. And Bakos (1997) showed that electronic marketplaces win commodity purchases by driving down buyer search costs, while leaving high service, high trust transactions with local intermediaries. The service attach is the LBS answer to both forces, and, as the demand data below shows, it is where a large and steady stream of buyer intent actually points.

Finally, the surface itself is moving. Aggarwal and colleagues (2024), in the peer reviewed GEO study, show that inclusion in a generated answer turns on content and citation signals that differ from classic ranking, which means visibility in an AI answer has to be engineered as its own objective. This study measures where that frontier sits for local bike retail today, and finds it, on Google at least, largely unbuilt for this category as of the capture date.

For market context, US bicycle sales reached a pandemic peak of 5.4 billion dollars in 2020, a roughly 64 percent surge, then settled to 4.1 billion dollars in 2023, up 23 percent on 2019 but down 24 percent on the 2020 peak (Circana, reported by The Christian Science Monitor, 2024). The specialty channel has thinned for two decades, from about 6,195 dedicated bike shops in 2000 to roughly 3,600 by 2022 (industry channel data, as compiled), and consolidation continued through the post boom shakeout, with 115 bicycle brands exiting the US market in 2024, more than four times the 2023 figure (research.bike, 2024). Against that backdrop, the digital shelf is not a nice to have for the surviving independent. It is the survival question.

What the machine names on Google

We captured 40 Google result sets on 2026-07-22 across the five metros, using ten shopper prompts that mixed proximity language (bike shop near me, bike repair near me) with metro templated language (best bike shop in Dallas, local bike shop Chicago). Each result was classified on one scheme: independent local shop, national chain or brand owned store, online marketplace, or directory and aggregator. The local map pack was classified by business name, the organic top ten by domain.

In the map pack the pattern is emphatic. Across the captured cells, independent shops held 104 of 113 local pack slots, about 92 percent, with the remainder going to chains and brand stores such as Dick's Sporting Goods and the occasional Trek concept store. In two metros, New York and Phoenix, every local pack result we captured was an independent. This is the surface where a nearby buyer's attention lands first, and it belongs, for now, to the neighborhood shop.

The organic top ten tells a harder story. There independents accounted for about 66 percent of results, but that figure is an upper bound: the classifier labeled any domain not on a known chain, marketplace, or directory list as independent, which sweeps in local cycling clubs, community sites, and unrecognized aggregators. The reliably classified competition took nearly a third of the page, directories and aggregators about 20 percent (Yelp foremost), chains and brand stores about 11 percent, and online marketplaces about 3 percent (Walmart and Amazon). A shop that wins the map pack sits above this contest; a shop that does not is dropped into it, competing for clicks against Yelp and the national names on their terms.

The most striking result is a negative one. On all 40 captures we requested Google's AI Overview, and it rendered on none. As of the capture date, Google is placing no generative answer above these commercial bike queries; the map pack and the organic list, both anchored in the Google Business Profile and reviews, still govern the local bike decision. That absence is itself a strategic fact, and we return to it in the discussion.

What the AI engine says

Google withheld a generative answer, but the dedicated answer engines did not. We captured Perplexity's response to best bike shop in {metro} for all five metros on 2026-07-22. Perplexity answered every query by drawing on Google Maps and Places data, presenting a shortlist of rated businesses and a short written comparison, and the businesses it named were, once again, overwhelmingly independent.

Across the five metros Perplexity named about 30 distinct businesses. Roughly 26 of them, about 87 percent, were independent local shops: Kozy's Cyclery and Comrade Cycles in Chicago, Waterfront Bicycle Shop and NYC Velo in New York, Landis Cyclery and Rusty Spoke in Phoenix, Outback Bikes and Loose Nuts Cycles in Atlanta, The Meteor and In City Wheels in Dallas. The rest were brand owned stores (Trek concept stores in Dallas and New York) and one outdoor chain (REI in Atlanta). Not a single online marketplace was named. When the question is explicitly local, the answer engine does not route the buyer to Amazon; it names the shop with the strongest local review profile.

One classification nuance deserves stating plainly. Because Perplexity keys on Places, it surfaces any physical location, so a Trek concept store appears beside the independents as just another nearby, well reviewed pin. The engine is not distinguishing independent from brand owned; it is ranking on proximity and rating. That is good news for the independent with strong reviews and a warning for the one without, because the brand store arrives with a national reputation already attached.

The AI answer engines: who ChatGPT, Perplexity, and Gemini name

The Perplexity read above was one engine, one prompt, on 2026-07-22. Google withheld a generative answer for this category, so to see the generative layer plainly we widened the capture. On 2026-07-23 we put the same buyer question to three answer engines with web search enabled, ChatGPT, Perplexity, and Gemini, across two prompt types (a reputational prompt, best independent bike shops in {metro}, and an intent prompt, where should I buy a bike or bike repair in {metro}), three metros (New York, Dallas, Chicago), captured twice each. Thirty-six of thirty-six attempts returned an answer. We read the full text of every answer, classified each named business as independent shop, national chain or brand store, or online seller, and classified the citations wherever the engine exposed them. One caveat governs the reading: Gemini masks its citations behind a single Google redirect host (vertexaisearch.cloud.google.com), so for Gemini the classification rests on the businesses named in its text, not on its sources.

The convergent result is strong and, for the independent, encouraging. All three engines named independent local shops in all three metros, and across the full thirty-six answers not one engine named an online marketplace. Amazon, and specialist etailers such as Jenson USA and Competitive Cyclist, simply did not appear when the question was local. The showrooming fear that frames so much of this category, that the machine will route a local shopper to a national seller, did not materialize in a single generated answer. ChatGPT listed five to eight shops per metro that were almost entirely independent, and the only chain it ever named was REI, and only on the buy-or-repair prompt. Gemini wrote editorial profiles of independents such as 718 Outdoors and King Kog in New York, Red Star Bicycles and Dallas Bike Works in Dallas, and Comrade Cycles and Blue City Cycles in Chicago, with only a Trek concept store and the regional chain Sun and Ski slipping into the Dallas intent answer. Perplexity named the widest field, dozens of shops still overwhelmingly independent, and was the one engine that routinely placed brand and chain stores (Trek Bicycle Dallas Park Cities, REI, Specialized Chicago Lincoln Park, Giant, Pedego) into the list beside the neighborhood shops.

Two qualifications follow from reading the runs side by side. First, these answers are not deterministic, and the three engines are not equally stable. ChatGPT was the steadiest: its named set overlapped substantially across the two runs, and in Dallas and Chicago the reputational lists were nearly identical. Perplexity and Gemini reshuffled more, keeping a recognizable core while the longer tail changed from run to run, and Gemini reshuffled the most, especially in New York and Chicago where only an anchor shop or two survived between the two captures. Second, the engines converge on a small core of the same independents per metro (Comrade Cycles in Chicago; Dallas Bike Works, Oak Cliff, and The Meteor in Dallas; NYC Velo and Toga in New York were each named by all three), then diverge into their own long tails, so agreement is real at the top of the list and thin below it.

The citation behavior differs by engine, and the difference is strategic, not cosmetic. ChatGPT cited the shops' own websites, so for that engine a shop's own web presence is read directly. Perplexity cited directories, city guides, and community media (Curbed, Time Out, CBS, Reddit, Yelp, Yellow Pages), so being featured and discussed in those places is the lever there. Gemini's sources are masked, so its selection mechanism is unreadable from the outside. The reading holds to the study's thesis and extends it: the surfaces machines use for local bike shopping name the independent, and the generative engines do so too, without routing to a marketplace. The contest is not independent versus Amazon. It is which independents make the shortlist, and whether a brand or chain store arriving with a national reputation slips in beside them. This is a single-day, k=2, three-metro reading and is reported as emerging.

The demand terrain and the readiness gap

Who buyers ask for reveals where the visibility game is actually played. We pulled real US monthly search volumes on 2026-07-22, and the shopper's language is proximity language, enormous relative to everything else. Bike shop near me and bicycle shop near me each draw about 673,000 US searches a month, as much as 1.35 million combined if the two near-synonyms do not overlap, with bike store near me adding another 135,000. These are Google's bucketed monthly estimates, so the combined figure is an upper bound rather than a clean sum. The industry's own vocabulary barely registers: local bike shop draws 3,600 and independent bike shop just 260. And buy a bike online, the phrase the whole showrooming panic is built on, draws only 590. Buyers are not, in the main, searching to buy a bike online. They are searching for a shop near them.

The service attach shows up directly in the demand data. Bike repair near me draws about 74,000 US searches a month and bike repair shop near me about 49,500, with bicycle repair and bike tune up near me adding tens of thousands more. This is the recurring revenue that Circana and shop owners describe as the survival model, and it is also a visibility asset, because every completed service is an occasion to earn the review that feeds the map pack. Service is both the moat and the signal generator.

Yet the independents are underbuilt for the machines that name them. In a small, directional audit of three independent shop websites, none carried LocalBusiness or Store structured data, the on page schema that tells an engine unambiguously what and where a business is. One had no structured data at all; two carried only ecommerce or generic organization markup. All three appeared in Google Business Profile and were named by Perplexity, so their visibility is real but rests entirely on the profile. That is a single point of failure: a profile suspended, mislabeled, or drifting out of sync with the rest of the web takes the whole visibility with it, because nothing on the shop's own site corroborates the entity.

Discussion

The evidence converges on a hopeful but fragile picture for the independent bike shop. Where machines actually adjudicate a local bike decision, on the Google map pack and in a Perplexity answer, independents win, holding about 92 percent of local pack slots and about 87 percent of named businesses. This is not luck. Both surfaces are built on the Google Business Profile and the review graph, and a real neighborhood shop with a service counter is a natural accumulator of exactly the local, recent, human reviews those systems trust. The buying mode explains the rest. Because the service and expertise around a bike behave as experience and credence goods (Nelson, 1970; Darby and Karni, 1973) sold as a servitized bundle (Vandermerwe and Rada, 1988), buyers lean on review signals (Luca, 2016), and the machine, foraging on their behalf (Pirolli and Card, 1999), surfaces the business with the strongest such signals, which is usually the local specialist. The aftermarket economics close the loop: the service counter that Cohen, Agrawal, and Agrawal (2006) identify as the durable good's profit engine is also the shop's review engine, so the same activity that pays the rent feeds the algorithm.

The fragility has two sources. First, the win is conditional on being inside the shortlist. Information foraging and choice overload theory both predict that businesses outside the three pack or the answer list are functionally invisible, and our organic data confirms the penalty for falling out: nearly a third of the organic top ten is chains, directories, and marketplaces, so a shop that loses the map pack competes against Yelp and national brands on a page that no longer favors it. Second, Google rendered no AI Overview on any of the 40 commercial bike queries. The generative answer, which the GEO literature (Aggarwal et al., 2024) treats as a distinct and engineerable surface, is simply absent for this category on Google today. That absence cuts both ways. It means the map pack still decides, so classic local signals remain the priority, but it also means the category could tip toward generative answers with little warning, and the shops that have already engineered their entity and content for that surface will be named first when it arrives.

Reading Google and the chat engines together sharpens the strategic picture, because they now agree on the thing that matters and disagree on the thing that is contestable. They agree on the enemy that is not there: no online marketplace appeared in the map pack, in the Perplexity answers, or in any of the 36 generative captures, so for a service attached durable good the showrooming route to Amazon simply is not the local machine's answer. Where they diverge is in what each surface reads to build its shortlist, and that divergence is a to do list rather than a contradiction. The map pack and Perplexity read the Google Business Profile and the review graph. ChatGPT reads the shop's own website. Perplexity also reads directories and local media. Gemini reads something it will not disclose. A shop that is one clear, well reviewed, well described entity, on its profile, on its own site, and in the places local media and directories talk about it, satisfies all four at once. The second order risk is concentration: because so much of this visibility flows through a single Google Business Profile, a shop that neglects its own site and its off site mentions is not merely underbuilt, it is fragile, one suspended listing away from vanishing from every surface that currently names it.

This is where the RavenEye model applies. The Visibility Corpus is our reading of where a market's attention actually sits, the terrain this study maps for one category: which surfaces answer, who they name, and what language buyers use. Search Surface Optimization is the method of engineering a business to be found and chosen across those surfaces at once, rather than chasing a single ranking. And the Machine-Readiness Score is the metric that tracks whether the work is moving the position, measured with the engine, locale, and date stamped on every reading, exactly as the captures in this study are. For bike retail the read is unusually clear: the local surface is the whole game right now, it runs through the profile and the reviews, and the on site entity signal is the underbuilt half.

Implications for independent bicycle retailers

The implication is that the independent bike shop is already winning the surface that matters, and the work is to consolidate that lead and remove its single point of failure, not to chase a ranking. None of what follows is a guaranteed lift; it is the set of signals a shop can legitimately engineer, offered as a remedy, not a promise.

The order of priority is dictated by the data. Win and hold the Google Business Profile and the review flow first, because that is what both the map pack and the answer engines read. Turn the service counter, which the demand data shows is a large and steady stream of intent, into a review engine, since every completed repair is a legitimate occasion to ask a real customer for a review under FTC rules. Then close the readiness gap by adding LocalBusiness structured data and consistent name, address, and phone details across the web, so the shop's own site corroborates the profile and the visibility no longer rests on one listing. Finally, prepare for the generative surface before it reaches Google, because the shops that read today as one clear, well reviewed entity are the ones the answer engines will name when the category tips.