SwarmSystem Playbook

The About page playbook

Every other page on a home service website answers the question "can you fix this?" This one answers "are you real?" It is the page somebody reads right before they decide to let a stranger into their house.

It is also the only page on the site whose facts go false on their own.

Written for the person who builds a plumbing company's website. Version one, September 22, 2026.

One company. One page. One question: are you real.

A homeowner has a leak. They searched, they read a service page, they are close. Then they click About, because a person they have never met is about to be in their kitchen with the water off.

That is the entire job of this page. Not traffic. Not a story. Somebody is hovering over the call button and needs one more reason to press it.

Here is the sentence that governs everything below.

Which gives you the test. Take your About page and delete the business name and every number on it. Hand what is left to your closest competitor. Could they publish it, word for word, as their own?

If yes, the page failed. It was never about you.

The principle behind this whole document: nothing here is a rule we have not applied to this page. The counts below are real counts. Where a number is ours rather than somebody's published research, it says so.

Do you even need two pages? Count the licenses.

Most advice on this subject assumes you are building a team page. Start somewhere else, because the honest answer for most shops is that you should not.

Most home service companies hold one master license, in the owner's name, and everybody else works under it. That is how the trade is structured in most states. So for most shops the count is 1, and the right answer is one page with the crew on it.

Why the licensed head count and not the total head count? Because a credential under somebody's own name is the only thing on a team page that a stranger can go and verify. A dispatcher with a three sentence bio does not survive that test, and a page of people who do not survive it is a page of stock-photo energy with real faces.

Three ways people get this wrong:

• Splitting on headcount. Twelve employees is not a reason. Twelve employees and one license is still one page.
• Splitting because a competitor did. Their license count is not yours.
• Splitting on ambition. A team page built for the company you plan to be reads as thin for the whole time you are not that company yet.

What happens when you split later

You will hire a second licensed person eventually, and then the rule fires and you split. When that happens you are moving content out from under a live URL, and that is a redirect job, not a copy and paste job.

The bios that were sections on your About page become their own pages. The About page keeps the company, the proof, and the licensed people. Anything you move gets a redirect from where it used to live. If you had an anchor link that people bookmarked, it still needs to land somewhere sensible.

Splitting is cheap if you plan for it and annoying if you do not. Build the single page so the seams are already in the right places.

The buying question is "who am I letting in my house?"

Service pages answer a job question. Can you fix a water heater, what does it cost, how fast can you get here. This page answers a different one, and it is closer to the bone.

Somebody is going to be alone in a house with a spouse, or a kid, or a dog, and a set of tools. That is the actual transaction. The plumbing is almost incidental.

So the content that works here is not the content that works on a service page. Three things carry it:

• Faces. The people who will physically arrive. Not a stock photo of a smiling man in a clean uniform.
• Accountability. Whose name is on the license, and what happens if the work is bad.
• Things a stranger can verify. Not adjectives. Records.

And the thing that does not work, which is most of what gets published: the founding story told as a story. More on that below, because it is the single most common way this page gets wasted.

The order of the page, and what each block is for.

Nielsen Norman Group studied how people look for information about a company across three rounds of testing. They observed 70 or more users, tested 100 websites and reviewed another 65. Their recommendation, in their words:

That settles the About versus team question without anybody needing an opinion. A team page is a subcategory page at their third level, sitting under the About summary at the second. It is not a replacement for the About page and it never was.

The combined page, which is the one most of you are building

1. A real photo, and the headline. Not the logo. Not a stock kitchen.
2. What you do and where, in two or three sentences. The reader may have landed here from a search, not from your home page.
3. The proof box. The facts a stranger can check, with the links to check them. This goes high, not at the bottom.
4. The people. Faces, names, roles.
5. How the company got here. Short. This is where the story lives, and it is fourth for a reason.
6. What you will not do. The most persuasive block on the page and almost nobody writes it.
7. What the first visit looks like. Removes the fear of the unknown appointment.
8. How to reach you. Phone first.

The split set, when the rule says you have earned it

Three shapes, and they do different jobs. Do not let them repeat each other.

Page	Carries	Does not carry
About	The company, the proof box, the licensed people	The full roster
Team hub	Everybody, with faces and roles, and a review date	The proof box, which lives on About
Bio	One person, their credential, what they actually work on	The company story again

The first screen has to earn the scroll.

Three things, and then get out of the way.

A real photograph. Your trucks, your building, your crew on an actual morning. Not the logo, not a stock photo of pipes. The reader came here to see whether you are a real company, so show them the real company. It does not have to be a good photo. It has to be a true one.

A headline that is not the title tag. These are different strings doing different jobs, and the next section is about why.

What you do and where, immediately. Not a mission. A plumber in Austin who works on houses. People who arrived from a search may have no idea what they clicked.

What does not belong on the first screen: a founding year in large type with nothing around it, an award badge from an organization nobody has heard of, and a carousel.

Six facts a stranger can check without asking you.

This is the block that makes the page work, and it is the block almost nobody builds.

Everybody writes "licensed and insured." It is on every plumbing website in the country, which is exactly why it carries no weight at all. It is a claim with nothing behind it.

Replace it with a box of facts, each one carrying the link that proves it. Aim for 6 facts and at least 4 outbound links to records you do not control.

For a home service business, the candidates are:

• The license number, and whose name it is under. Linked to the state board lookup. This is the strongest one and we will come back to it.
• Insurance. Not "we are insured." What you carry, and that you will send the certificate the same day somebody asks.
• The business registration. Your state's business records are public and searchable. Link them.
• Bonding, where the trade uses it.
• A trade specific registration. A backflow tester registration, a gas license, a city permit account.
• Whether the owner is named on the Google Business Profile. Anybody can look. Most cannot find a name.

Say it in plain words, and pass the swap test

Every fact in this box has to survive one check.

Half your facts have an expiration date.

Here is the thing that makes this page different from every other page on your site.

A service page about water heater repair is as true in March as it was in January. Nothing on it rots.

An About page is not like that. A license renews. An insurance certificate lapses. A registration expires. A person quits. The page does not change, and it slowly stops being true anyway, while you are not looking at it.

And it is worse than a normal stale page, because the whole point of this page is that it can be verified. A page whose job is proof, carrying a fact that no longer checks out, has failed at the one thing it was for. Somebody clicks your license link, finds it expired, and now they are not wondering about the license. They are wondering about everything else you said.

Sort your facts into two piles

Permanent	Perishable
The year you were founded	The license, and its renewal date
The state filing	Insurance coverage
Who founded it	Bonding
Where you are based	Trade registrations
	Accreditations
	Every single person on the page

Two things this buys you. The page tells the truth about its own shelf life, which is the most credible thing a page can do. And somebody now has a specific date in a calendar instead of a vague intention to keep the website current.

The owner facts interview takes fifteen minutes.

The hard part of this page is not the writing. It is that the person writing it does not know anything only this company knows, so they write the things every company says.

Fix it by asking. 15 minutes with the owner, 8 questions, and you will have more usable material than a week of staring at competitors.

Question three is the one that pays. Local mechanism is not decoration. A plumber who can explain what Austin's clay soil does to a slab in August is saying something a competitor cannot copy without also knowing it. Generic prose gets paraphrased. Specific prose has to be quoted, and a quote carries your name with it.

How it should sound, and the words that get cut.

Write it the way the owner talks to a customer in a driveway. Short sentences. Common words. One idea at a time.

Then cut these:

Cut	Because
Family owned and operated	Every competitor says it. Say who the family is.
Committed to excellence	Survives the swap test. Says nothing.
Your trusted local plumber	Trust is something they decide, not something you assign yourself.
We treat every home like our own	Nobody believes this sentence.
With over 20 years of combined experience	Combined experience is arithmetic, not expertise. Say who has how long.
Passion for plumbing	No.

The founding story, and why it goes fourth

The most common piece of About page advice is "tell your brand story." It is bad advice here, and the reason is mechanical rather than aesthetic.

A story is unquotable, and unquotable means uncited. When an answer engine assembles a reply about plumbers in your city, it needs a sentence it can lift. "Ray started as an apprentice in Round Rock in 2003 and holds Texas master plumber license M-41982" is liftable. "We have always believed that plumbing is about people" is not liftable by anything, because it means nothing specific enough to repeat.

So keep the story. Put it fourth, keep it to a few paragraphs, and make sure it contains dates, places, and names rather than feelings. The story is what makes the page human. It is not what makes the page findable, and pages that lead with it spend their best position on their least useful content.

What this page is actually trying to rank for.

A service page earns its place with a keyword. Somebody types "water heater repair austin" and the page that matches wins. This page does not work that way, and treating it like a service page is how it gets ruined.

What replaces the keyword is the entity. This is the page your business name resolves to. It is the page that answers "who owns it", "how long have they been around", "are they licensed", "is it a real company or a call center". Those are the queries, and none of them are service keywords.

So do not force a service keyword into the opening. The first hundred words of this page are the company. A plumber whose About page opens with "Austin water heater repair and drain cleaning services" has taken the one page that could establish who they are and turned it into a worse version of a page they already have.

What to actually do

• Business name and city in the first screen. Not stuffed into the first sentence. Present above the fold, where a reader and a crawler both find it without hunting.
• The business name spelled exactly the same way everywhere. One page that calls you two things splits your own entity. If the sign says Copperline Plumbing then the page says Copperline Plumbing, not Copperline Plumbing LLC in one paragraph and Copperline in the next.
• Indexable. An astonishing number of About pages carry a noindex tag somebody set in a theme years ago. Check yours.
• One top level headline, and the heading levels in order. No jumping straight from the headline to a fourth level subheading because the theme styled it that way.
• A unique meta description written as a pitch, not the first sentence of the page repeated.

The title tag and the headline do different jobs.

The title tag competes in a list of ten results. The headline confirms the reader landed in the right place. One string cannot do both, and using the same string for both wastes one of them.

Google does publish guidance, and it is worth following:

• Descriptive and concise. Their example of what to avoid is exactly ours: "Avoid vague descriptors like 'Home' for your home page, or 'Profile' for a specific person's profile." A title that says "About Us" is that mistake.
• Distinct per page. Your About and your team page cannot carry the same title.
• Brand at one end, separated by "a delimiter such as a hyphen, colon, or pipe."

Our house rule narrows that last one. Use a colon or a pipe, never a dash. A dash separator in a title looks like a search result instead of a sentence, and we stopped using them across everything we ship.

Page	Title pattern	Headline
Combined About	About [Business], a Licensed [City] [Trade]	The people who will be standing in your house
Split About	About [Business], a [City] Shop Since [Year]	Who is behind the trucks
Team hub	Meet the [N] People at [Business] in [City]	[N] people, and [N] of them hold the license
Bio	[Name], [Role] in [City]: [Business]	[Name]

How long, and why length is not the lever.

Length is the most asked question about this page and the least important thing about it. A page that hits a word count and says nothing checkable is worse than a short page carrying a license number.

Bands exist anyway, because a page that is too short cannot do the job and a page that is too long has padding in it. Ours:

Page	Words	Sections
Combined About	900 to 1,600	4 to 10
Split About	700 to 1,200	3 to 7
Team hub	500 to 900	3 to 6
One bio	150 to 400	2 to 4

Every one of those is ours, not research. They come from building pages to this standard and finding where they stop working. A combined page under 900 words has usually skipped the proof box. A bio over 400 is a resume nobody asked for.

If you are short, the fix is never padding. It is the owner facts interview. You are short because you do not have anything specific to say yet.

Faces of the people who will be in the house.

How many photos: one per person on the roster, plus one real photo of the company. That is the whole rule. A combined page with five people carries 6 images.

Not a gallery. Not twelve shots of pipes. The reader is here to see faces, and every image that is not a face or the building takes up the space where a face should be.

The specs

Thing	Standard
Aspect ratio	One ratio across the whole site. Square or portrait. Pick one and never mix.
Source size	800 by 800 pixels or larger, served at about twice the rendered size
Format	WebP or AVIF, with a JPEG fallback
File size	100 KB or less after compression
Filename	ray-delgado-master-plumber.jpg, never IMG_0023.jpg
Markup	A real <img> with width and height set

Google is explicit about two of these. On filenames: "When possible, use filenames that are short, but descriptive." And on how images get found at all: "Google can find images in src attribute of <img> element... Google doesn't index CSS images."

That second one matters more than it sounds. A headshot set as a CSS background is invisible. Plenty of themes do exactly that with team grids, and the photos you paid for are not in the index at all.

Set width and height on every image. It stops the grid jumping as photos load, which is a layout shift problem and a Core Web Vitals problem at the same time.

Alt text, and the line you should not cross

Describe the person the way you would to somebody on the phone. Name, what they do, where.

"Ray Delgado, Owner and Master Plumber at Copperline Plumbing in Austin" is right.

What makes it wrong is turning it into a template and running it across a dozen photos with keywords wedged in. Google, in the same document: "Avoid filling alt attributes with keywords (also known as keyword stuffing) as it results in a negative user experience." An alt reading "austin plumber best plumber near me Ray Delgado emergency plumbing" is that.

Ask before you publish somebody's face

This gets skipped and it should not. You are putting an employee's face and full name on a public page, in a trade where people go into strangers' homes. Some of your crew will have reasons not to want that, and those reasons are theirs.

Ask. Get a yes. Offer the option of a first name and a last initial. And when somebody leaves, take them down, which you should be doing anyway.

How it should look, and what not to let a designer do.

• The grid collapses cleanly. Three or four across on a desktop, two on a tablet, one or two on a phone. No horizontal scroll at any width.
• The whole card is the link, where a card links to a bio. Not just the name.
• Uniform card heights. A staggered grid reads as a broken page before anybody reads a word.
• No carousel. Everybody after the first slide is effectively not on the page.
• Names and titles as real text, never baked into an image.
• Bios render without JavaScript. If a bio only appears when a modal opens or a filter runs, it is not reliably on the page for a crawler and it is gone for anybody whose script failed. A filter is fine as an enhancement. The unfiltered list has to work without it.
• Keyboard reachable, with a visible focus state. If a bio expands in place, the toggle is a real button with aria-expanded, not a styled div.
• Contrast. Do not set names in white over a light headshot.
• Abbreviate credentials one way. Pick it once and hold it across the site.

Who gets their own page, and who stays a card.

Different question from the split rule. That one decided how many pages the company has. This one decides which humans get a URL.

Give them a page when	Keep them a card when
People search their name	Nobody searches them individually
They hold a credential a stranger can verify	The role is generic
They write content on the site	They do not publish
Their bio runs past 150 words	A short blurb covers it

Being on a card is not a demotion. A page with three sentences on it helps nobody, and it is one more page that goes stale when that person leaves.

URLs and navigation

• Team hub at /our-team/ or /about/team/. Short and stable.
• Bios at /team/first-last/. One person, one crawlable URL.
• The team page sits under About in the main navigation, and appears again in the footer.
• Never put a bio only in a modal, a tab, or a PDF.

Opening a bio

Open on what makes this person different. Not on where they went to school.

"Ray takes the calls where somebody else has already been out once and the problem came back" is an opening. "Ray received his certification from" is a resume, and the reader has already stopped.

A byline is a promise to link somewhere.

Google asks a set of questions about content quality. Three of them are about this page, in their words:

A byline that links nowhere answers the first and fails the second. The bio page is the further information that question is asking for. That is what an author link is, and it is the only reason to build one.

Google's own framing of why any of this matters: "Something that helps people intuitively understand the E-E-A-T of content is when it's clear who created it." And on the four letters, "Of these aspects, trust is most important."

Which is the argument against treating E-E-A-T as a score to optimize. It is not a dial. It is corroboration you either supply or do not.

Links go out to proof and sideways to the work.

Most internal linking advice is about passing authority around. On this page it is about something simpler: letting somebody finish the check they came here to do.

Out, to things you do not control

This is the unusual one. Most pages avoid sending people away. This page should, because a link to a state licensing board is worth more than any sentence you could write instead. Target at least 4 outbound proof links.

Sideways, to the rest of your site

Link to	Why
Your main service pages	They came to check you out. Some are ready to book.
The team page or the bios	Both directions, always.
Contact	Obvious and frequently missing.
Service area pages	"Do you even come out here" is a real question.
Reviews or testimonials	Proof of a different kind than a license.
Financing or warranty, where you have them	These are trust pages too.
Careers, if you are hiring	A real share of About traffic is job seekers.

And from the other direction, which is the part people forget: your home page links here, your service pages link to the people who do that work, your blog bylines link to bios, and your footer carries an About link on every page. That last one is the fourth of Nielsen Norman's levels.

The code the page carries.

Structured data is a block of code in the page that states the same facts in a form a machine reads without guessing. It does not make you rank. It removes ambiguity.

Organization, and where Google says to put it

Google's structured data documentation says this, which is the clearest argument for this page existing at all:

Google names the About page. That is where your company's identity block belongs.

There are no required properties. The ones worth carrying for a local service business: name, legalName, url, telephone, address, foundingDate, numberOfEmployees, founder, and sameAs pointing at your profiles elsewhere.

Give it an @id. That is the anchor everything else on your site points back at, and it is what stops you having several versions of your own company in the same graph.

AboutPage

AboutPage is a real schema.org type, and its whole definition is "Web page type: About page." It produces no rich result and Google publishes no feature for it. What it does is say, unambiguously, that this page is about that organization. Point its mainEntity at the Organization's @id.

Where the license number lives

schema.org has a property for exactly this: hasCredential, defined as "A credential awarded to the Person or Organization." Put the license there, with the issuing body in recognizedBy.

So the proof box on the page and the code in the head end up being the same claim said twice. A fact a person can check and a machine can read. That is the whole design.

ProfilePage is earned, not applied

Most advice says put ProfilePage on every staff bio. Read what Google actually says it is for: "ProfilePage markup is designed for any site where creators (either people or organizations) share first-hand perspectives." The valid examples they name are a forum profile and an author page on a news site.

There is an upside in reading it this way. ProfilePage becomes something a person earns by writing, which gives you an actual reason to get the owner writing.

Two more to skip

• knowsAbout as an expertise signal. schema.org defines it as "suggesting possible expertise but not implying it." The property does not assert what people use it to assert.
• A hub page full of ProfilePage blocks. A team hub covers many people, so it is not about one entity. Leave it with the Organization markup, or add an ItemList pointing at the bios.

And keep the name, address and phone in your schema identical to your Google Business Profile. Conflicting versions of the same business across your own pages is how you end up as two half entities instead of one.

What answer engines can see, and what they cannot.

An answer engine is the thing that writes a paragraph at the top of a search result instead of just listing links. AEO is the work of being the source that paragraph is built from.

Before any tactics, the boundary, and this is quotable to a client:

What this page controls: factual accuracy, whether a claim can be extracted, local specificity, valid schema, and one correct business entity. What it does not control: review volume, directory presence, and the third party citations an assistant actually aggregates.

Here is what is genuinely different about this page. Every other page type answers a job question. This is the only page that answers the identity questions, and those get asked constantly: who owns it, how long have they been in business, are they licensed, is it family run, is it a franchise. No service page carries any of that.

Advice we refuse, with reasons

Refused	Why
Tell your brand story	A story is unquotable, and unquotable means uncited.
FAQ schema as the lever	Google cut those rich results for most sites. Answer the questions because people ask them.
About pages do not rank, keep them short	Ranking was never this page's job. Being citable is.
E-E-A-T as a score to optimize	It is corroboration you supply, not a dial you turn.
Question shaped headings everywhere	Blows the section count and reads as a generic FAQ instead of a company page.
Sprinkling the city name through the copy	Theater. See the local section below.

These are dated September 22, 2026, so a later reader knows what we were looking at when we refused them.

Being named is won mostly off this page.

A generative engine is a tool like ChatGPT or Perplexity that writes an answer rather than returning links. GEO is the work of being in that answer.

Almost everything written about GEO is guessing, so here is the honest version. The mechanism we can defend is this: a generative engine restating a fact about a business carries some risk of being wrong, and corroborated facts are the ones it can restate safely. A license number that matches a state registry is corroborated. "The most trusted plumber in Austin" is not, and no system should repeat it.

That is the same doctrine as the corroboration block. It is not a separate discipline with separate tactics, which is what most GEO advice sells.

What we are not going to tell you: that doing this gets you named by an assistant. We do not have that evidence and neither does anybody selling it to you. The boundary above holds. This page makes you accurately describable. What an assistant does with that depends mostly on things off your website.

The map pin is not this page's job. Agreeing with it is.

Local SEO on an About page is mostly cargo cult. Putting "Austin plumber" four more times into your company history does nothing, and anybody telling you it lifts your map ranking is selling something.

What actually matters here is agreement. Three things:

• Name, address and phone identical to your Google Business Profile. Not similar. Identical, down to whether it says Suite or Ste. This is the second most authoritative statement of your business identity on the internet, after the profile itself, and the two disagreeing is a real problem.
• The license number. The one local trust signal almost nobody publishes, and it ties your page to a government record that already has your business name and address in it.
• A service area list that links to real pages. If you have town pages, link them. If you do not, a plain list of the towns you cover is still useful and honest. What you must not do is list forty towns you do not serve.

The About page is what somebody clicks after they found you on the map. It is not what put you there.

When somebody quits, this page starts lying.

The team page is the fastest rotting page on your website, and the damage is worse than a stale page anywhere else.

Think about what it means. A customer meets the tech at the door. They looked at the site last night. The person in front of them is not on it, and somebody who left in March still is. Every other claim on that page just became a maybe.

The page whose entire job is proof turned into the page that proves you do not maintain your own website.

This is the section other About page advice does not have, and it is the one that costs you if you skip it. Everything else here is a build task you do once. This is the only thing on the list that fails by itself while you do nothing.

The stop doing list.

Run this against the page you have now. Each one is a yes or a no, and you need no tools.

Check	Fail looks like
The swap test	A competitor could publish your page with the name changed
The license number	The page says licensed and does not say the number
Outbound proof	Fewer than 4 links to records you do not control
A review date	No date on the page, or one somebody picked
Real photos	Stock imagery, or a face that does not work there
Alt text	Empty, or a keyword stack
CSS backgrounds	Headshots set in CSS, so they are not indexed
Title and headline	The same string, or a title that says About Us
Dash separators	A dash used as a separator in the title
The split rule	A team page with fewer than 2 lookupable credentials on the roster
Author links	Bylines that link nowhere
ProfilePage	On a bio for somebody who has never published
Schema agreement	Name, address or phone differs from the Google Business Profile
JavaScript	Bios only visible after a click or a filter
An owner	Nobody is responsible for updating it

Two ways in, and what we still do not know.

If you build the pages

1. Count the lookupable credentials. Produces: one page or two.
2. Run the owner facts interview. Produces: eight answers that fail the swap test. Trap: accepting the first generic answer. Ask again.
3. Build the proof box. Produces: six facts, four or more outbound links, each perishable one dated.
4. Compute the review date. Produces: the earliest expiry, on the page. Trap: letting somebody choose it.
5. Shoot the photos in one session. Produces: one ratio, one lighting setup, consent from everybody. Trap: filling gaps with stock.
6. Write the page in the block order. Produces: the story fourth, not first.
7. Write the title and the headline as different strings. Trap: a dash separator.
8. Add the schema. Organization with an @id, AboutPage pointing at it, the credential in hasCredential. Trap: ProfilePage on somebody who does not publish.
9. Wire the loop. Bylines to bios, bios to services, services to people.
10. Name an owner and put the review date in their calendar.

If you own the trucks

Three questions for whoever runs your website. You do not need to know how any of it works to ask them.

1. Is my license number on the About page, linked to the state board? If not, why not. It takes ten minutes.
2. What is the date we are supposed to look at this page again, and whose calendar is it in? If the answer is vague, nobody owns it.
3. Is everybody on our team page still working here? Go look right now. This one catches more problems than the other two.

What we do not know yet

This page separates two kinds of claim on purpose. Numbers with a source are somebody's published documentation. Numbers marked as ours are house standards, arrived at by building pages this way and seeing where they break. Hover any number to see which it is.

Honestly: most of the numbers here are ours. The sourced material is Google's documentation on structured data, titles, snippets and images, plus Nielsen Norman Group's research on About pages. The word bands, the section counts, the six facts, the four outbound links and the eight interview questions are all house standards, and they are dated. A future reader should re-examine the house half and trust the sourced half.

What we are not claiming: that building a page this way produces a measurable result. We do not have that evidence yet. The pages built to this standard are new, and the earliest honest read on them is a Search Console comparison we have not run. When we have it, this section is where it goes, and if the numbers do not move, this section is what changes.

We also do not know whether the review date changes behavior. The theory is that a specific date in a calendar beats a vague intention. That is a reasonable belief and it is not yet a measured one.