The Replenishment Blind Spot: Who the Machine Names When Shoppers Look for a Local Pet Store

Introduction

Most retail categories are bought when a need appears. Pet supplies are not. A household with a dog or a cat runs a standing set of consumables, food, litter, flea and tick control, treats, that empty and refill on a predictable cadence. The purchase is less a decision than a rhythm, and that rhythm is the most valuable asset in the category. Whoever owns the reorder owns the customer for years, because the next bag of food is not a fresh choice. It is a habit firing on schedule.

National marketplaces understood this early. Chewy and Amazon built the pet business around autoship, the standing subscription that ships the same food and litter at a set interval for a small discount. The model turns a recurring need into a recurring revenue line and, more consequentially, deletes the moment a shopper might reconsider. For the independent pet store, that deleted moment is the whole contest. The real question is whether the recurring buyer searches at all, beyond whether the shop can rank when someone does.

This study reads the terrain directly. We treat the search engine as the place a local pet shop is either named or invisible, and we measure what the machine actually returns. The research question that runs through this series arrives here in sharpened form: when a shopper looks for pet supplies, who does the machine name, and where do independent local shops stand once the replenishment purchase has been trained to skip the search entirely.

Data and method, in brief. We captured live Google result pages for eight real shopper prompts across five US metros on 22 July 2026, and separately measured US search demand for the category’s buyer language. Every figure below is dated to its capture and tagged with an evidence tier. Nothing here is a client case. Raveneye Global has no retail clients, and none of the shops named in our data are ours.

Background and literature

The category is large and still growing. The American Pet Products Association put total US pet industry spending at 158.0 billion dollars in 2025, up 3.7 percent year over year, with pet food and treats alone at 68.3 billion. The line closest to a pure supplies retail figure, supplies plus live animals plus over the counter medicine, was 33.3 billion in 2024. IBISWorld counted 13,483 US pet stores as businesses in 2025. This is not a small or shrinking shelf, which makes the visibility question consequential rather than academic.

The buying mode has deep support in the literature, and that support is more specific than a general appeal to habit. Wood and Neal (2009) describe habitual consumption as a context-cued automatic response that runs largely independent of current goals. Verplanken and Aarts (1999) sharpen the definition to goal-directed automaticity that depends on a stable situational cue, so change the cue and you change the behavior. Neal, Wood, Wu and Kurlander (2011) supply the decisive mechanism: in their cinema experiments habitual eaters consumed stale popcorn as readily as fresh, but only while seated in the theater that cued the habit, and the pull released once that context was removed. Autoship is the commercial application of exactly this finding. It strips out the old cue, the trip to the store, and installs a new one, the scheduled delivery, so the replenishment habit keeps firing while the local shop’s context never appears. Samuelson and Zeckhauser (1988) add the last turn of the screw: a subscription is a default, and people disproportionately keep the default they are on, so the reorder persists not because the shopper re-chose the marketplace but because they never revisited the choice at all.

Switching-cost theory explains why that default holds. Burnham, Frels and Mahajan (2003) separate procedural, financial, and relational switching costs, and a subscription builds the first two by design through saved payment details, a set cadence, and a subscribe discount. Klemperer (1987) frames the whole pattern as a lock-in land grab, where firms discount hard early to harvest a captive base later, which reads as a fair description of marketplace pet pricing.

What can an independent hold against that. The theory points to relational and informational advantages rather than convenience. Berry, Seiders and Grewal (2002) decompose service convenience into stages and show that a local retailer cannot beat a marketplace on access or transaction convenience but can win on decision and post-benefit convenience, the expert help before and after the sale. Nelson (1970) and Darby and Karni (1973) classify goods by how buyers verify quality: pet food is heavily an experience good, and premium or therapeutic nutrition carries credence attributes a buyer can never fully verify, which is precisely where a trusted human intermediary keeps pricing power. Kirk (2019) and Epley, Waytz and Cacioppo (2007) explain the emotional engine underneath, showing that psychological ownership and anthropomorphism drive higher willingness to pay and stronger word of mouth for pets treated as family.

Two more bodies of work frame the visibility fight itself. Luca (2016) found that a one-star rise in Yelp rating lifts revenue by 5 to 9 percent, and that the effect concentrates in independent businesses because chains already carry brand-level reputation, which makes reviews the independent’s reputational equalizer. Aggarwal and colleagues (2023, KDD 2024), in the founding paper on generative engine optimization, show that source content can be tuned to lift its inclusion in generated answers by up to roughly 40 percent, a lever that matters as buyers begin asking assistants rather than search boxes. Gensler, Neslin and Verhoef (2017) supply the cautionary backdrop: the local store often serves as a free showroom whose value the marketplace captures at checkout.

Findings

We captured 40 live Google result pages, eight prompt families across five metros, on 22 July 2026. The first result is a structural one about the surface itself. Across all 40 queries in the primary grid, not one returned a Google AI Overview. For these local, high-intent retail prompts, both the navigational ones like pet store near me and the evaluative ones like best pet store in the city, Google answered with a map pack and a list of links, not a synthetic summary. A separate k equals 5 intra-day recapture the next day did surface sporadic AI Overviews in a minority of cells, so the reading is that for this category, at capture, the local pack is the answer, and the generative overview that dominates informational search is still rare and emerging here rather than permanently absent.

Inside that local pack, independents are strong. Of 120 local pack slots, 73 percent were independent local shops, against 16 percent national chains and about 6 percent big-box, pharmacy, or grocery. An independent appeared in 39 of the 40 queries. Real names recurred across metros: Cozy Paws and Noble Beast in Phoenix, My Natural Pet and Pamper Ur Pets in New York, Liz’s Pet Shop and Ruff Haus Pets in Chicago, The Upper Paw and Odyssey Pets in Dallas, Highland Pet Supply and Park Pet Supply in Atlanta. Eligibility to be named is not the gap. These shops hold a Google Business Profile, which is the price of entry to the pack, and many rank a real website too.

The organic column below the map tells a harder story. Across 390 organic results, the single most common domain was Yelp, appearing 49 times, ahead of Petco at 37, PetSmart at 20, and Walmart at 17. At the very top organic position, a directory or a national chain held the first clickable link 75 percent of the time, and an independent retailer held it only 20 percent. So the same independent that wins a map pack slot is usually buried beneath an aggregator page and a chain store locator in the organic list, the place a shopper scrolls when the three named shops on the map are not enough.

The marketplace pattern is the most revealing. Chewy and Amazon were almost invisible in local search, roughly 6 percent of organic results and a single local pack slot in the entire grid. This is not weakness. It is that the marketplaces do not fight on the near me query at all. They win before the search happens. Our demand data shows the mechanism plainly. The single largest term we measured, by a wide margin, was the marketplace brand name itself, at 2.74 million US searches a month, while the chain-branded near me terms drew a million each. The generic local-discovery language a shop might rank for is a far thinner slice, pet supplies near me at 49,500, and the words we use for these businesses, local pet store at 2,900 and independent pet store at 210, are barely searched at all. Those are our words, not the buyer’s.

Finally, the replenishment gradient. The more a query looks like a routine reorder, the more the chains and marketplaces intrude. Cat litter near me, the most consumable prompt in the set, showed the lowest independent share of the local pack and the highest marketplace share of the organic column, and the reorder and delivery terms carried the steepest commercial bids we saw, dog food delivery at a 29 dollar cost per click and cat litter delivery at 11 dollars. That is the sound of the marketplaces bidding to own the recurring purchase, the exact purchase the independent most needs to keep.

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

The Google grid above says the local pack still answers pet-retail searches and that the synthetic overview is rare here. That leaves an open question the study can now begin to answer directly: when a shopper skips search and asks a chatbot instead, does the machine still name the independent, or does it route to a chain or a marketplace. On 23 July 2026 we put the same buyer question to three synthetic engines, ChatGPT running, Perplexity running, and Gemini, all with web search live, across three of these metros (New York, Chicago, Dallas), with two prompt framings each, a reputational one asking for the best independent stores and an intent one asking where to buy pet food and supplies, and repeated every cell twice. All 36 captures returned. The headline is that all three engines named real independent local shops in all three metros, and on the reputational prompt they named almost nothing but independents. The chatbot, like the map pack, is willing to name the small shop.

The engines differed sharply in how they backed those names up. ChatGPT cited the shops’ own websites and Google Maps entries directly, 1800whiskers.com, odysseypets.com, lizspetshop.com, ruffhauspets.com and their peers, with no directory pages in its citation set, so its answer reads as a set of primary business sites. Perplexity named the widest roster of distinct independents but sourced them almost entirely from directories, local media, and listicles, Curbed, Yelp, Reddit, the Dallas Observer, CBS News, the Chicago Tribune, the Better Business Bureau, and a cluster of pet-blog roundups, so its independents arrive filtered through third-party editorial rather than the shops themselves. Gemini wrote the richest descriptions and explicitly separated big-box chains from neighborhood shops, but every one of its citations was masked behind a single Google redirect host, so its sources cannot be classified at all, which is a real limit on what we can verify about how it chose. On the intent prompt each engine also folded in the national chains, Petco, PetSmart, Pet Supplies Plus, Hollywood Feed, and Gemini and ChatGPT also surfaced the marketplace as a delivery option, so the chains and marketplaces re-enter exactly where the question turns from which shop is best to where do I buy.

Across the two runs of each cell the engines were directionally stable but not fixed. A core of shops recurred, Odyssey Pets, Uptown Pup, The Upper Paw and Green Pet in Dallas, Ruff Haus Pets, Liz’s Pet Shop and Wigglyville in Chicago, Whiskers in New York, while a tail of names rotated in and out between the first and second capture. ChatGPT was the most repeatable, its Dallas and Chicago lists nearly identical run to run and only New York reshuffling substantially; Perplexity and Gemini held a stable core and rotated the longer tail. Across engines the picture was convergent in Dallas, where Odyssey Pets, The Upper Paw, Uptown Pup and Green Pet appeared in all three, looser in Chicago, and most divergent in New York, where ChatGPT led with Whiskers and Pet Market, Perplexity led with Petqua and Gone to the Dogs, and Gemini led with Whiskers and Bushwick Bark, sharing little beyond Whiskers and Little Creatures. The three engines agree that independents deserve to be named; they do not agree on which independents.

For the study’s thesis the reading is consistent and, at this stage, directional rather than settled. The synthetic engines behave like the local pack, not like the demand curve: they name the independent when asked, which means the shop that is eligible to be named, a claimed profile, a real website, third-party mentions, can surface in a chatbot answer the same way it surfaces on the map. But the divergence across engines and the run-to-run rotation mean no single named position is durable, and the fact that Perplexity reaches its independents only through directories while Gemini hides its sources tells the independent that being present in the directory and review layer, not only on its own site, is what feeds the answer. This is one day, three engines, three metros, and two runs, so the label here is emerging. It says the chatbot door is open to the independent, and that winning it is the same visibility work the map pack rewards, aimed now at the sources these engines actually read.

Discussion

Put the two halves together and the terrain resolves. When a shopper does search locally for a pet store, the independent is not losing. It holds most of the map pack and appears in nearly every query, and the engine, whether Google’s map or a chatbot, is willing to name it. The problem lives one step upstream: for the category’s defining purchase, the recurring replenishment order, the search never happens. Autoship converts the habit Wood and Neal describe into a standing subscription, and with roughly four fifths of the largest marketplace’s sales already running on autoship, a growing share of pet spending has been permanently removed from the moment of discovery. You cannot be named in an answer to a question no one asks.

The economics of that removal compound quietly. Fader and Hardie (2009) show that the value of a customer base lies in the projected duration of the relationship rather than the size of any single order, so the marketplace that captures the reorder captures a discounted stream of future purchases, while the independent that wins a one-time visit captures only the visit. Every cycle the subscriber does not re-search widens the gap, because the default persists on its own momentum, exactly the status-quo pull Samuelson and Zeckhauser documented. The second-order effect is a shrinking, self-selecting pool of searchers: as more routine demand is absorbed into subscriptions, the searches that remain skew toward the non-routine, the new owner, the failed delivery, the specialty need, which are lower in volume but higher in intent and far more winnable on expertise than on price.

This reframes what visibility work has to accomplish for a pet retailer. It is not primarily a ranking problem to be solved once. It is a re-entry problem. The independent has to be present and convincing in the searches that still happen, and it has to convert that thinner, higher-intent traffic into a relationship strong enough to resist the pull of subscribe and forget. The literature names the levers precisely. Relational switching cost, per Burnham and colleagues, is the independent’s durable moat, built through staff who know the specific animal and give advice a marketplace cannot. Reviews, per Luca, are the reputational equalizer that lets a small shop punch at chain weight, and their revenue effect concentrates on exactly the independent businesses this study is about. The cross-surface picture reinforces the same conclusion from a second direction: the chat engines rewarded the shops with a clean profile, a real site, and third-party mentions, the identical signals the map pack reads, so a single body of visibility work now serves two answer surfaces at once.

The RavenEye model reads this as three linked layers. The Visibility Corpus is the terrain map, the per-category picture of where a market’s attention actually sits, which for pet supplies means recognizing that the map pack is winnable, the organic column is crowded with aggregators and chains, the AI Overview is still rare and emerging for local retail queries, and the largest force in the category is a subscription that bypasses search altogether. Search Surface Optimization is the method of engineering a business into the places that map identifies as reachable, the local profile, the entity consistency, the review system, the schema, so the engine can name the shop cleanly when the search does occur. The Machine-Readiness Score is the single honest number that tracks whether that position is improving. None of it promises a rank, and for this category none of it can promise to reverse the subscription tide by itself. What it can do is make sure the shop is the one the engine names in every search the marketplace has not yet absorbed, and give the owner a measured read of where they stand rather than a guess.

Implications for retailers

For an independent pet store, the practical reading of this study is bracing but useful. Do not fight the marketplace where it is strongest, on price and access convenience for a locked-in subscriber, because that battle is already lost at the point of enrollment. Fight where the engine still names you and where a human relationship still decides the sale.

Concretely, that means three moves. First, own the local pack you already largely hold, by making the Google Business Profile complete and accurate, keeping name, address, and phone identical everywhere the engine reads them, and running a compliant, real-customer review system, because reviews are the one signal proven to move revenue for independents specifically. Second, get findable in the organic and chatbot searches that still happen, so that a new owner researching the best food for a specific breed or condition finds the shop’s expertise rather than only an aggregator page or a chain locator, and so the directory and review layer the answer engines read carries the shop’s name. Third, treat the recurring purchase as a relationship to earn in person, since that relational tie is the only switching cost strong enough to counter autoship. The caveat holds throughout: visibility work makes a business easier to find and choose, it does not guarantee a ranking, and it does not by itself convert a subscription buyer who never searches. It makes sure that when the search does come, the answer is you.