What does a stranger learn about you in the eight seconds after they type your name into Google? That question is the whole job. Investors check it before a first call. Journalists check it before they reply to a pitch. Candidates check it before they accept an offer. Most founders have never looked at the page with fresh eyes, and the ones who have usually find a mix of their own homepage, a stale directory listing, an old podcast they forgot about, and at least one result they would rather not explain.
That page has a name: the brand SERP. Fixing it is not reputation management in the crisis sense, and it is not classic SEO either. It is the slow, deliberate work of deciding which ten pages should describe you, then giving Google good reasons to agree.
What Is a Brand SERP, Exactly?
Jason Barnard, founder of Kalicube, coined the term “Brand SERP” in 2012. Kalicube defines it as “the search engine results page that appears when someone searches for an exact brand name,” and Barnard calls the traditional version a brand’s Google business card. His sharper line, from a November 2020 WordLift article, is that a brand SERP is “Google’s opinion of the world’s opinion of you.” That phrasing matters. You do not write the page. Google assembles it from what your site says, what other sites say, and what searchers click.
The stakes come from click behavior. Backlinko’s study of roughly 4 million Google results, published by Brian Dean on August 27, 2019 and updated April 16, 2025, found the first organic result earned a 27.6% click-through rate and that only 0.63% of searchers clicked anything on page two. For a name search, page one is the entire record. Whatever sits in those ten positions is your reputation for most of the people who look.
Kalicube has also extended the idea to AI. Barnard coined “AI Résumé” in 2024 for the conversational version, the answer an assistant gives when asked about you. Same input sources, different presentation. Keep that in mind, because the work in this guide feeds both.
The Features That Crowd Your Ten Blue Links
A modern results page for a name rarely shows ten plain links. Several features compete for the same screen, and each one pulls from a different source. Knowing which source feeds which feature tells you where to spend effort.
Sitelinks are the indented shortcuts under your homepage result. Google’s sitelinks documentation, last updated December 10, 2025, is blunt: sitelinks are automated, Google shows them only when it thinks they help, and there is no tool to add one. You can remove a bad sitelink only by deleting the page or setting it to noindex. Your influence comes from page titles, site structure, and internal anchor text.
The site name is the label Google prints above your homepage result. Google’s site names documentation says it reads WebSite structured data on your root URL, with a required name and url plus an optional alternateName for acronyms or short forms. It works at the domain and subdomain level, not for subdirectories.
The knowledge panel sits on the right on desktop and near the top on mobile. It pulls from the Knowledge Graph, which reconciles facts from your site, Wikipedia, Wikidata, and other structured sources. We cover it in depth in our Google knowledge panel guide.
People Also Ask boxes surface questions searchers ask about you, with answers lifted from whatever page Google trusts most for each question. Top Stories carousels appear when recent news coverage mentions your name. Video carousels pull mostly from YouTube. Social profile links, image packs, and review stars round out the list.
Then there is the AI Overview, which now sits above everything for many brands. Ahrefs published a study by Ryan Law on October 2, 2026 covering 232.8 million US desktop search crawls from July to September 2026. As of September 29, AI Overviews appeared on about 82.9% of results for branded keywords. In a separate panel of 100 brands drawn from Interbrand’s Best Global Brands 2025, bare brand-name queries with an AI Overview climbed from 5 of 98 usable results on July 1 to 63 by September 30, and 48 of those 63 sat in position one. Among the overviews with visible sources, Wikipedia was cited 71.9% of the time, YouTube 38.6%, and LinkedIn 22.8%. Your own site is no longer guaranteed to be the first thing a searcher reads about you.
Run an Audit Before You Touch Anything
Start with evidence. Open a private browser window, sign out of Google, and search your exact brand name. Then search it again on a phone, because mobile layouts move the knowledge panel and AI Overview into different positions. If you serve more than one country, repeat the check with a VPN or a rank-tracking tool set to each market. Screenshot every version.
Next, write down all ten organic results in order, plus every feature you see. For each organic result, note the domain, who controls it, the date of the content, and whether the page describes you accurately. For each feature, note the source. Which pages does the AI Overview cite? Which site answers each People Also Ask question? Does the knowledge panel description come from Wikipedia, your About page, or somewhere you have never heard of?
Then search the variations people use: your brand plus “reviews,” “pricing,” “CEO,” “lawsuit,” “scam,” and “login.” Google’s autocomplete suggestions for your name are a free list of what the market wonders about you. A results page that looks clean for the bare name can look very different one modifier away.
The output of this audit is a single spreadsheet with one row per slot. It takes about 30 minutes for a company and less for a person. It becomes the baseline you measure against every month, and it feeds directly into the model below.
The Ten-Slot Ledger: Owned, Earned, Borrowed, Open
At Instant Press we run every name-search engagement through a model we call the Ten-Slot Ledger. Kalicube’s work focuses on how Google understands an entity and builds its panels. The Ledger is narrower and more operational: it treats the ten organic positions as a balance sheet and asks who holds each one.
Every slot gets one of four labels. Owned slots are pages you publish and fully control: your website, your blog, your careers page. Earned slots are editorial pages someone else published about you because you gave them a reason to: press articles, podcast episode pages, award listings, interviews. Borrowed slots are profiles on platforms you do not own but can edit: LinkedIn, Crunchbase, YouTube, X, Instagram, G2, Trustpilot, an Apple Podcasts show page. Open slots are everything else: pages about someone with the same name, unrelated businesses, outdated directories, and hostile content.
The target balance we aim for on a company’s page one is roughly three owned, three to four earned, three borrowed, and zero open. Owned pages give you control of the message. Earned pages give you credibility that no amount of self-description provides. Borrowed profiles give you breadth and feed the structured sources AI systems read. A page one stacked with nine owned results looks like a company talking to itself. A page one with no owned results means someone else is telling the story.
The Ledger also forces a useful question about features. For each feature on the page, which column is it drawing from? If your AI Overview cites mostly Wikipedia and an old press release, you know which slots to work on. If People Also Ask answers come from a competitor’s comparison page, that is an open slot hiding inside a feature. Count those too.
The audit gives you a current ledger. The rest of this guide is about moving slots from open to one of the other three columns, starting with the column you control completely.
Lock Down Your Owned Slots
Your website should hold position one for your exact name. If it does not, fix that before anything else, because every other tactic depends on Google being sure which site is yours. That starts with an entity home page, usually your About page or homepage, that states who you are, what you do, and links out to every official profile.
Add Organization or Person schema with a sameAs array listing your official profiles, and add WebSite schema with the exact site name you want displayed. If your brand name is ambiguous, an alternateName helps Google print the right label.
Sitelinks come next. Because Google picks them automatically, you shape them by making your most useful pages obvious: clear titles like “Pricing,” “Careers,” and “About,” prominent links from the homepage, and short descriptive anchor text. If Google shows a sitelink to a dead campaign page or a login screen you would rather hide, the documentation leaves you two options, delete it or noindex it.
Beyond the homepage, one or two additional owned pages can rank for your name on their own. A founder’s bio page, a press page that lists coverage, or a blog with real readership can each take a slot. The test is simple: does the page deserve to be the third thing a stranger reads about you? If the honest answer is no, do not force it.
Where Does Press Coverage Fit on Page One?
Earned slots are where most name searches are thinnest and where PR does its best work. An article on a recognized publication ranks for your name because that publication’s domain already carries authority, and Google treats editorial coverage as independent evidence about you. A single feature in a trade outlet with a clean headline can hold a page-one slot for years.
Coverage also reaches the features. Recent articles are what trigger a Top Stories carousel for a name search. Interviews and op-eds give People Also Ask something credible to quote. Press mentions that name you clearly, describe what you do, and link to your entity home help the Knowledge Graph confirm your facts, which is how coverage ends up influencing a knowledge panel. And because AI Overviews lean on a short list of trusted sources, a well-placed article is one of the few ways to get your own version of events into the answer above slot one.
Not every placement earns a slot. Syndicated press release pages often get filtered as duplicates. Paid listicles on low-quality sites rarely rank for your name and can look worse than nothing. What works is coverage with your name in the headline or subhead, on a domain Google trusts, with substance a reader would want. Our digital PR guide covers how to earn that kind of placement.
Plan earned coverage as a portfolio, not a one-off. Three or four strong pieces spread over a year, each on a different publication, beat one big hit that ages out of Top Stories in a week and slides down the page as fresher results appear.
Borrowed Slots Are Cheaper Than You Think
Borrowed profiles are the fastest slots to fill because the platforms already rank. LinkedIn, Crunchbase, YouTube, X, Instagram, and review sites carry enough domain authority that a complete, active profile will often reach page one for a distinctive name within weeks.
The catch is the word “complete.” An empty profile with a logo and no description rarely ranks, and when it does it looks abandoned. Use the same name, description, and URL on every profile, matched character for character with your entity home. Link each profile back to your site, and list each one in your sameAs markup. That two-way link is what lets Google and AI systems treat the profiles as confirmations of one entity rather than a pile of similar names.
Pick borrowed platforms by audience, not by count. A B2B software company gets more from LinkedIn, Crunchbase, G2, and YouTube than from TikTok. A consumer brand may need Instagram and Trustpilot more than Crunchbase. Remember the Ahrefs citation data, too: YouTube and LinkedIn showed up in roughly four in ten and one in five AI Overview source lists respectively. A YouTube channel with a few useful videos does double duty, filling a video carousel and feeding the AI answer.
Keep the profiles alive. A LinkedIn page with a post from 2022 tells a searcher something too.
What Santorum’s Decade Teaches About Open Slots
The cost of leaving a slot open has a famous case study. In 2003, columnist Dan Savage, angry over Senator Rick Santorum’s views on homosexuality, recruited readers to attach a crude definition to Santorum’s surname and link to it until it topped Google searches for his name. It worked. According to NPR’s February 25, 2016 report, the result dogged Santorum for nearly a decade.
His fix arrived in 2011, when his campaign hired digital strategist Eric Sherred and made search a priority. Sherred’s team published content with Santorum’s name every day and asked friendly bloggers for links. NPR reports the campaign finally took the number one spot a few weeks before the 2012 Iowa caucuses. “We finally did it,” Sherred said. “That was a big day for us.” By 2016, Santorum was confident enough to tell voters to Google him.
Two lessons carry over to any brand. First, an open slot does not stay neutral. If you do not fill page one, someone with a motive or simply more content will. Second, displacing a hostile result is a volume game played over months, and the tools are the same three columns: fresh owned content, earned links and coverage, and active borrowed profiles. For the tactical side of that work, see our guide on how to bury negative search results.
Most companies never face anything as pointed as Santorum’s problem. Their open slots are quieter: a namesake company in another state, a ten-year-old directory page with the wrong address, a forum thread from a former employee. Each one still costs a slot that could be telling your story.
Jason Barnard’s Rebuild Shows the Order of Operations
The best illustration of how these pages behave comes from the man who named it. In a 2020 article for Search Engine Journal, Barnard described what happened to his own results that summer. Wikipedia administrators deleted his article, after he admitted he had over-edited it while experimenting with how to feed the Knowledge Graph. Within a week, his entity dropped out of Google’s Knowledge Graph and the knowledge panel on his personal results page disappeared. The Wikipedia articles for his folk-punk band, The Barking Dogs, and for the children’s TV series Boowa & Kwala were deleted shortly after.
What he did next is the order of operations this guide recommends. He rebuilt from his owned corner first. Using WordLift’s entity markup on his own site, he created a new entity in about six weeks and got his knowledge panel back. In his words, Google came to use his site rather than Wikipedia as the reference for information about him. For The Barking Dogs, his site became Google’s main source after the deletion, which meant he could change the panel description himself, and he reported updates appearing in about ten minutes.
The point is not that Wikipedia is bad. It is that a brand SERP built on a single borrowed source is fragile, and one built on a strong entity home, corroborated by earned coverage and consistent profiles, recovers. Barnard lost the most authoritative third-party source he had and still got his panel back because his owned slot was doing the real work.
If you want a quick read on where your own ledger stands, the Instant Press free AEO rating checks the same signals that feed your search results and your AI answers, and it is a good first step before you start filling slots.