Smaller brands get named more than you think. On the right questions.
Field notes #1: 874 answers from ChatGPT, Perplexity and Google AI Overviews about eight brands, asked the way their buyers ask. Every figure below comes with its range, and the questions are published word for word so anyone can ask them again.
Field notes #1: what we found
Eight brands, mostly smaller than the leaders of their category. Questions drafted by the product from each brand’s homepage, each asked three times per assistant, on 6 to 8 October 2026.
Named in half the answers or more
On questions built around what each brand does differently, seven of the eight were named in 51% to 87% of answers. Broad “best in category” questions are where brands disappear, so each set kept only a few.
Steadier than its reputation
Asked the same question three times on the same assistant, whether the brand was named came out the same all three times in 213 of 241 cases (88%). Which brands make a list changes from run to run; whether yours is on it changes much less.
ChatGPT reads the brand’s own site, Perplexity reads everyone else
ChatGPT cited the brand’s own site in 25 Brooklinen answers and 41 Plausible answers; Perplexity did so in 3 and 7, leaning on review sites and roundups instead. Owned pages and third-party coverage are two different jobs.
Buying questions are the weak spot
For four of the eight brands, questions asked at the point of purchase named them least: Kubera in 20% of them against 90% for comparison questions, Graza in 40%, Bombas in 2 of 27. That gap is where an agency has the most to work on. Cal.com and Tuta went the other way, named in 20 of 24 and 20 of 25 buying answers.
The control questions stayed quiet
Questions about a neighboring category each brand does not sell named it in 2 of 128 answers. The counting finds the brand where it belongs, not everywhere.
The assistants mostly agree, not always
Where both were asked, ChatGPT and Perplexity agreed on whether a brand was named for 62 of 77 questions. The other 15 are a reason to report each assistant separately rather than blend them.
Google almost always answers first
Across three brands, Google showed an AI Overview above its results in 112 of 128 searches: 44 of 44 for Tuta, 35 of 37 for Cal.com, 33 of 47 for Bombas. For buyers who start on Google, that box is the answer they read before any link.
The best-known brand was named least
Bombas, the most familiar name of the eight, was named in 10% to 26% of answers depending on the assistant. Its homepage blocked our reader, so its questions were drafted from the category alone and came out broad, the kind where brands disappear in every set. Questions that fit what a brand does differently are the first thing to get right.
| Brand | Category | Assistant | Named in | Range (95%) | Buying questions, all assistants |
|---|---|---|---|---|---|
| Brooklinen | bed sheets and bedding | ChatGPT | 51% (32 of 63) | 39–63% | 60% (18 of 30) |
| Perplexity | 54% (34 of 63) | 42–66% | |||
| Plausible | website analytics software | ChatGPT | 90% (44 of 49) | 78–96% | 64% (9 of 14) |
| Perplexity | 84% (41 of 49) | 71–92% | |||
| Graza | olive oil | Perplexity | 62% (37 of 60) | 49–73% | 40% (6 of 15) |
| Kubera | net worth tracking software | Perplexity | 67% (42 of 63) | 54–77% | 20% (3 of 15) |
| Tally | online form builder | Perplexity | 51% (32 of 63) | 39–63% | 80% (12 of 15) |
| Cal.com | scheduling software | ChatGPT | 77% (30 of 39) | 62–87% | 83% (20 of 24) |
| Perplexity | 67% (26 of 39) | 51–79% | |||
| Google AI Overviews | 82% (23 of 28) | 64–92% | |||
| Tuta | encrypted email service | ChatGPT | 87% (34 of 39) | 73–94% | 80% (20 of 25) |
| Perplexity | 77% (30 of 39) | 62–87% | |||
| Google AI Overviews | 86% (30 of 35) | 71–94% | |||
| Bombas | socks and basics | ChatGPT | 26% (10 of 39) | 15–41% | 7% (2 of 27) |
| Perplexity | 13% (5 of 39) | 6–27% | |||
| Google AI Overviews | 10% (4 of 39) | 4–24% |
Shares count answers to the brand’s own questions; control questions are left out here and reported in finding 5. Plausible ran 17 of its 24 questions before the run reached its budget, three brands were measured on Perplexity only, and Cal.com, Tuta and Bombas were also asked on Google AI Overviews, where 16 of 144 searches failed at the data provider and are left out. Bombas’s questions were drafted without its homepage, which our reader could not open. Eight brands in one week are field notes, not a market study: read the ranges before the shares.
Brooklinen: the 24 questions, verbatim
- Which soft sheets feel comfortable for sensitive sleepers?
- What are good options for sheets with a soft feel that lasts?
- Which sheets feel soft without feeling too flimsy?
- What are good sheets for hot sleepers?
- Which sheet brands are good alternatives to Parachute sheets?
- What are some alternatives to Boll & Branch sheets?
- Which premium sheets make a thoughtful wedding gift?
- Where can I buy high-quality sheets online?
- Are premium sheets worth the price for everyday use?
- Which sheets are best for a California king bed?
- What are good options for a cozy comforter for cold winters?
- Which comforters feel warm and cozy without being too heavy?
- What are the best comforters for year-round use?
- Which comforter brands are good alternatives to Snowe?
- Which premium comforters are worth buying as a housewarming gift?
- Where can I find a king-size comforter from a premium brand?
- Which pillows come in both soft and firm support options?
- What are good firm supportive pillows for side sleepers?
- Which soft pillows work well for stomach sleepers?
- What are the best supportive pillows for neck pain?
- Which pillows are a good choice for a guest room?
- Which mattress toppers add a plush feel to a firm bed?
- What are good sleep masks for blocking out light while traveling?
- Which pajamas are comfortable for warm summer nights?
Plausible: the 17 questions, verbatim
- Which privacy-focused tools are good alternatives to Google Analytics for a small website?
- What are the best lightweight tools for tracking website visits without slowing pages down?
- Which website analytics tools work without cookie consent banners?
- What are good alternatives to Fathom Analytics for privacy-friendly website stats?
- What are good alternatives to Simple Analytics for a clear, easy-to-read dashboard?
- Which tools are good alternatives to Matomo for tracking website traffic?
- Which open-source tools can I use to understand traffic on my website?
- What are good website analytics options if I want to own and control my data?
- Which website analytics tools charge based on traffic instead of users or events?
- What are the simplest tools for seeing where my website visitors come from?
- Which privacy-friendly tools can replace Google Analytics on a blog?
- What website analytics tools are a good fit for an online store?
- Which tools show website traffic in a straightforward dashboard without lots of reports?
- Which tools can track website visits without collecting personal visitor data?
- Which website analytics tools offer a free trial?
- What website analytics software has a free plan for a low-traffic site?
- How much do privacy-friendly website analytics tools cost in the US?
Graza: the 24 questions, verbatim
- Which extra-virgin cooking oil comes in a squeeze bottle that's easy to control?
- What are good extra-virgin oils with refill cans for everyday cooking?
- Which extra-virgin oils come in glass bottles instead of plastic?
- What are good olive oil gift sets with more than one style of oil?
- Which cooking oils are made for high-heat cooking?
- What are some olive-oil-based mayonnaises to try?
- What are good extra-virgin olive oils for finishing dishes straight from a squeeze bottle?
- What are good alternatives to Brightland olive oil for everyday cooking?
- Which extra-virgin olive oils come in refillable or recyclable packaging?
- What are good alternatives to California Olive Ranch extra-virgin olive oil in a squeeze bottle?
- What are good alternatives to Colavita extra-virgin olive oil for drizzling?
- Which olive oil gift set has separate bottles for cooking and finishing?
- What are good olive oil brands that sell both glass bottles and refill cans?
- Which high-heat cooking oil is a good alternative to extra-virgin olive oil?
- Which olive oil trio is a good way to try different cooking styles?
- Where can I buy an olive oil starter set in the US?
- What's a good olive oil gift set under $50?
- Is a premium olive oil trio worth the price as a gift?
- Where can I buy extra-virgin olive oil potato chips?
- Which olive oil subscriptions let me reorder refill cans?
- What are the best cast-iron skillets for home cooking?
- Which avocado oils are best for high-heat cooking?
- What are good countertop air fryers for a small kitchen?
- Which ceramic nonstick pans are good for everyday cooking?
Kubera: the 24 questions, verbatim
- Which apps can track bank and brokerage accounts alongside crypto and real estate?
- What tools combine US and international financial accounts in one wealth dashboard?
- Which wealth apps handle multiple currencies across accounts and assets?
- What are good options for tracking private investments, LP positions, and public stocks together?
- Which platforms can include homes, cars, jewelry, and watches in a personal balance sheet?
- What software lets me separate assets across trusts, LLCs, and personal ownership?
- Which wealth trackers let a spouse or advisor view the same financial picture?
- What apps update investment and account balances close to real time?
- Which platforms track crypto, commodities, and traditional investments in one place?
- What tools can show cash balances alongside investments and debts?
- Which services create a shareable proof-of-wealth report for a lender?
- What are alternatives to Empower for tracking investments, property, and other assets?
- Which apps offer private benchmarks for comparing wealth with similar households?
- What software can send account access instructions to someone I choose if I become incapacitated?
- Which wealth tracking apps connect to international banks and brokerages?
- Is there a wealth app that can track ownership stakes across multiple entities?
- Which software offers a trial before I connect all my financial accounts?
- What does premium wealth tracking software typically cost per year?
- Which personal finance apps have a free plan for tracking investments and assets?
- Is a paid wealth dashboard worth it for someone with private investments and real estate?
- What are alternatives to Monarch Money for tracking investments and total assets?
- What estate planning services help organize account and asset information for heirs?
- Which password managers are best for sharing emergency access with family?
- What tax software works well for people with rental property and investment income?
Tally: the 24 questions, verbatim
- What are good alternatives to Typeform for collecting customer feedback?
- Which online form tools let me collect unlimited responses for free?
- What are good alternatives to Jotform for creating registration forms?
- Which form builders support conditional questions based on earlier answers?
- What are good alternatives to Google Forms for polished client intake forms?
- Which no-code form tools can calculate a quote from someone's answers?
- What form builders let me use a custom domain without coding?
- Which tools make it easy to embed a form on my website?
- What are good form builders for collecting sensitive information with better privacy controls?
- Which form tools can send submissions to my other business apps?
- Which tools are good for building a polished job application form?
- What form builders work well for a simple event RSVP?
- Which form tools let me add a pop-up signup to a website?
- What are some easy form builders for someone with no technical experience?
- Which online forms can I customize with my own colors and fonts?
- What form builders are good for creating a quick product survey?
- Which form builders have a genuinely free plan with no response limit?
- What's the best free online form builder for a personal project?
- Which form tools offer a free plan with conditional logic?
- How much do popular form builders cost if I need a custom domain?
- Is a paid form builder worth it for a small business collecting leads?
- What are good appointment scheduling tools for a solo consultant?
- Which online payment providers are good for collecting deposits?
- how much does online form builder usually cost
Cal.com: the 16 questions, verbatim
- Which booking tools let me customize the experience to match my website?
- What are good options for adding a booking flow to my SaaS product?
- Which scheduling APIs are built for platforms where users meet with other users?
- Best booking software for an independent consultant taking client calls?
- Which tools let customers choose a time to talk with me directly from my website?
- What are good alternatives to Calendly for customizable booking pages?
- Which platform is best for building a branded appointment-booking experience into an app?
- What booking tools give each provider a personal page where clients can request a call?
- Which developer-friendly platforms can handle bookings between users in my marketplace?
- Best scheduling software for a business that takes calls from prospective customers?
- Which booking platforms offer a free plan for an individual taking client calls?
- What scheduling software has a free trial for testing an embedded booking flow?
- Which customizable scheduling platforms publish clear pricing for developers?
- Which video meeting apps are a good fit for client conversations?
- What are good customer relationship management tools for a solo consultant?
- Which online form builders work well for collecting client details before a call?
Tuta: the 16 questions, verbatim
- Which email providers offer end-to-end encryption by default?
- What are good alternatives to Proton Mail for private email?
- Which email services are preparing for future quantum-computer threats?
- What are the best ad-free email services with encrypted calendars?
- Which private email providers include calendar and contact syncing?
- What are good alternatives to Fastmail for privacy-focused email?
- Which email services are open source and protect message privacy?
- What are good alternatives to Mailfence for encrypted email?
- What are the best private email services with a free plan?
- Which encrypted email providers offer a free plan without ads?
- What does a paid private email plan usually cost per month?
- Which private email provider is easiest to switch to from a regular inbox?
- What are good private cloud storage services for personal files?
- Which password managers make it easy to share logins securely?
- What are good private web browsers that block online tracking?
- Which VPN services are simple to use on a phone?
Bombas: the 16 questions, verbatim
- Which cushioned socks stay comfortable on long walks without bunching?
- What are good no-show socks that stay put in low-cut sneakers?
- Which everyday socks have a smooth toe seam for sensitive feet?
- Best options for socks with arch support that don't feel too tight?
- Which socks have a heel tab to help prevent blisters during runs?
- What are good lightweight merino socks for warm-weather hikes?
- Which socks stay up on your calves without digging in?
- What are the best options for soft, breathable underwear for everyday wear?
- What are good lightweight everyday T-shirts that hold their shape?
- What are good compression socks for long flights?
- Where can I buy well-made everyday socks online?
- Are premium cushioned socks worth the price for daily wear?
- Which soft cotton socks make a useful gift under $50?
- Which walking shoes have a roomy toe box for wide feet?
- What are good lightweight rain jackets for day hikes?
- how much does socks and basics usually cost
Where the assistants look, by category
The sites cited most often in answers to each brand’s questions, counted once per answer. This is where coverage gets read, so it is where an agency starts. A brand’s own site is marked.
bed sheets and bedding · Brooklinen
- sleepfoundation.org · 38 of 126 answers
- goodhousekeeping.com · 38 of 126 answers
- brooklinen.com (own site) · 28 of 126 answers
- nytimes.com · 21 of 126 answers
- ncoa.org · 13 of 126 answers
- llbean.com · 13 of 126 answers
website analytics software · Plausible
- plausible.io (own site) · 45 of 98 answers
- matomo.org · 32 of 98 answers
- usefathom.com · 31 of 98 answers
- docs.umami.is · 22 of 98 answers
- developers.cloudflare.com · 16 of 98 answers
- umami.is · 16 of 98 answers
olive oil · Graza
- target.com · 13 of 60 answers
- foodnetwork.com · 10 of 60 answers
- amazon.com · 8 of 60 answers
- nytimes.com · 7 of 60 answers
- nbcnews.com · 7 of 60 answers
- tastecooking.com · 7 of 60 answers
net worth tracking software · Kubera
- kubera.com (own site) · 26 of 63 answers
- quicken.com · 15 of 63 answers
- robberger.com · 14 of 63 answers
- mycapitally.com · 12 of 63 answers
- vyzer.co · 9 of 63 answers
- mezzi.com · 8 of 63 answers
online form builder · Tally
- paperform.co · 21 of 63 answers
- jotform.com · 21 of 63 answers
- webflow.com · 13 of 63 answers
- softr.io · 13 of 63 answers
- orbitforms.ai · 13 of 63 answers
- zapier.com · 12 of 63 answers
scheduling software · Cal.com
- cal.com (own site) · 49 of 106 answers
- calendly.com · 31 of 106 answers
- simplybook.me · 14 of 106 answers
- cronofy.com · 14 of 106 answers
- developer.nylas.com · 13 of 106 answers
- help.acuityscheduling.com · 12 of 106 answers
encrypted email service · Tuta
- proton.me · 54 of 113 answers
- tuta.com (own site) · 54 of 113 answers
- allaboutcookies.org · 19 of 113 answers
- reddit.com · 19 of 113 answers
- cyberinsider.com · 13 of 113 answers
- blog.mailfence.com · 11 of 113 answers
socks and basics · Bombas
- google.com · 17 of 117 answers
- darntough.com · 12 of 117 answers
- bombas.com (own site) · 12 of 117 answers
- thewearify.com · 11 of 117 answers
- goodhousekeeping.com · 10 of 117 answers
- menshealth.com · 10 of 117 answers
Up to twelve sources were kept per answer. Review sites and roundups dominate consumer categories; in software, competitors’ own pages and documentation do.
What the full study asks
Narrow on purpose. A study that tries to describe “AI search” as a whole ends up describing nothing that can be checked.
- How much do answers differ between runs? Ask the same buyer question repeatedly, on the same assistant, in the same week, and count how much the set of brands named changes.
- How much do they differ between assistants? The same question on each assistant we measure, compared side by side rather than blended.
- What do they cite? Which kinds of source turn up in answers to commercial questions, and how often the same domains recur.
- How many answers does it take? How large a sample has to be before a share stops moving around. That is the number that decides what the product is allowed to show.
SparkToro asked AI tools the same recommendation questions many times and found the list of brands almost never repeated: fewer than 1 in 100 runs gave the same list.
Somebody else’s work, on which brands make a list. Our field notes measure something narrower, whether a brand is named at all, and found that far steadier.
How the full study is run
The same way the product measures a client, which is the point: if the method is not good enough for a study, it is not good enough to bill an agency for.
Sampling
By default each question is asked three times on each assistant you switch on, every run, and you can ask more. A cell with fewer than three answers in its window shows a dash instead of a number.
Through the API
Each assistant is asked through its developer API with web search switched on, not through the consumer app; Google AI Overviews and AI Mode are read from Google search results through a data provider. What a person sees in the app can differ.
Windows
Answers are grouped by week. Screens add up a rolling window (the last 28 days by default) so a figure rests on more answers; reports compare the start of the period with the end.
What is kept
Recorded on every answer: the raw text, the pages it cited, the model version that produced it, and what it cost to ask.
What gets published with it
A number on its own cannot be checked. These go out with it, or it does not go out.
The questions
The full list of prompts, verbatim, so anyone can ask them again rather than take our word for what was asked.
The counts
How many answers were collected per question per assistant, per week, not only the shares worked out from them.
The ranges
Every share with the interval around it and the number of answers behind it, on the same basis the product uses.
The dates
When each answer was collected. Assistants change their behavior over time, so a figure without a date is not repeatable.
What went wrong
Questions that had to be dropped, runs that failed, anything that would change how the numbers should be read.
What we did not measure
Microsoft Copilot has no API for its answers, so it is not measured, and no report estimates it. Google AI Overviews and AI Mode can be switched on per client; Google has no API for them either, so their answers come from search results through a data provider.
What we will not claim, even with the data
Said now, while there are no results to be tempted by.
- A rank or position in any assistant
- A single 0–100 visibility score
- That a piece of work produced a change: we show what followed, with a comparison where one exists
- A forecast of the share a client will reach
- A revenue figure for visibility
- Anything about assistants we do not measure
Not a market survey
It measures how assistants answer a fixed set of questions. It says nothing about how many people ask them, or what any of it is worth in revenue.
A sample, not a census
A finite number of questions on a finite number of assistants, over a finite period. Every figure will carry the range that follows from that.
Nothing about your client
General findings do not transfer to one brand. A study is a reason to measure your own client, not a substitute for measuring them.
No claim about why
If something moves during the study we will show what moved and over which weeks. Attributing it to a reason is a different study than this one.
Now measure your own client
A general study would tell you how assistants behave. An audit tells you what they say about the client whose retainer is on the line, which is the only figure that settles an argument with that client.