Draft: hidden from Google until Dan signs it off.
SEO (search engine optimisation) means making your website the one Google, and now AI tools like ChatGPT and Google's AI Mode, choose to show when people search for what you sell. I'm an SEO consultant on the Central Coast, NSW, working with businesses across Australia. Most SEO I see chases keywords and forgets the people typing them. I start with who your buyers are and what they're worried about, worked out from your own search and site data, then build and angle each page so the right person feels it was written for them.
Written by Dan Picken, founder. Last checked 27 October 2026.Persona A
Worry · Cost
'How much is this going to cost me?'
Page angle → Upfront pricing, what moves the price
Persona B
Worry · Reliability
'Will they actually turn up?'
Page angle → Reviews, arrival windows, a named plumber
One search, two pages. Each one wins its own click.
Audience first
Every search has a person behind it, and the same search can come from people with very different worries. Most SEO writes one page for an average reader who doesn't exist. I start with who your buyers actually are, built from your own data, then angle each page so the right person feels it was written for them.
Find the audiences
Group your buyers by what they are trying to get done and what worries them, not by age and postcode.
Check them against real searches
For each audience, list the searches they'd make, then check them against what people actually typed in Search Console and Google Ads. If nobody searches it, it's a guess, and it gets dropped.
Angle every page
Decide which audience each page is for, then write the opening, the proof and the next step for that person.
Worked example: home batteries
Two people can search for the same product with completely different reasons. Same topic, two pages that open differently:
The bill watcher
'is a home battery worth it'
The blackout worrier
'battery backup during power outage'
One generic page would half-suit both and fully suit neither. Two angled pages each get the click, and each keeps the reader.
From my audits
Most sites I look at have the same few problems, whatever the industry. These are the six I check first.
Traffic is up, enquiries are flat. The site ranks for information searches from people who were never going to buy, and not for the ones who were.
Two or three pages aimed at the same search, so Google can't tell which one to show and none of them ranks well.
Swap the business name and it could be any competitor's page. Nothing on it gives Google, or a customer, a reason to pick it.
Blog posts get the effort while the service and category pages, the ones people actually buy from, stay thin.
A new website goes live, old addresses aren't redirected, and years of rankings disappear in a week.
It's free, it comes straight from Google, and it often shows pages Google found but chose not to keep. Most businesses have never opened it.
How Google picks what to show
It ranks pages, one search at a time. Google describes three stages: find the page, understand it, then decide if it's the best answer. Its AI answers add a fourth step on top. Most SEO problems sit in one of these, and the fix is different for each.
Google's bots find your pages by following links and reading your sitemap.
What I do: Make sure every page that matters is linked from somewhere, and nothing important is blocked.
Google reads each page, works out what it is about, and decides whether to store it.
What I do: Check Search Console for pages Google found but didn't keep. That list is usually telling you something.
For each search, Google picks the indexed pages it thinks answer it best.
What I do: Build one clear page per search intent, so your own pages stop competing with each other.
Google's guide asks whether a page offers real value compared with other pages in the results.
What I do: Add what the top results don't have: your data, your experience, a worked example, a tool.
Google's AI runs a set of related searches behind each question and pulls from the pages that answer them.
What I do: Answer the follow-up questions too, each in its own short section.
Worth knowing: Google says it doesn't accept payment to rank a site higher, and doesn't guarantee any page will be shown. It also says no one can guarantee a #1 ranking. If someone promises you one, Google's own advice is to find someone else.
Method: Information Gain Scoring
Google's own guide asks whether a page offers original information or analysis, and real value compared with other pages in the results. Most pages don't. They say what the top ten already say, a bit longer. Open your page next to the top three results for its main search, then tick only what yours has and theirs don't.
Your page adds
0/10
Nothing the top results don't already have. To Google, it's one more copy.
Fix this first
Add one number only you could know. Even a count of jobs or orders beats a statistic everyone else quotes.
A self-check, not Google's score. Google doesn't publish one.
Consumer psychology
A search is a string of tiny decisions: which result to click, whether to stay, whether to believe it, whether to act. Each one is shaped by how people actually think, and each has been studied. Here's what the research says at each moment, and what I do about it.
01Choosing a result
People judge a link by how strongly its words and context smell like what they're after, before they ever click.
The title and description in Google are your shop window. Clever wording loses to the searcher's own words.
What I do: Write titles and descriptions in the exact words people use in Search Console, not the words used inside your business.
02Choosing a result
Kahneman and Tversky found people weigh losses more heavily than equal gains.
Someone with a problem is already worried about losing something: money, time, a weekend.
What I do: For problem searches, test a title about what they stand to lose against one about what they'll gain.
03The first few seconds
Eyetracking shows people scan in an F-shape: the first lines and the first words of each line get most of the attention.
And with Google one tap away, it's cheap to go back and try the next result. Pages get seconds, not minutes.
What I do: Answer the search in the first two lines, and start every heading with the words that matter.
04Deciding whether to believe it
In one study, statements that were easier to read were judged true more often than the same statements made harder to read.
How easy a page is to read changes how much people trust it. Dense, clever, jargon-heavy pages feel less believable.
What I do: Short sentences, plain words, strong contrast. The technical term once, then the plain version.
05Looking at prices
Tversky and Kahneman showed estimates stay close to the first number people see, even when it's irrelevant.
On 'how much does it cost' pages, the first figure becomes the yardstick for everything after it.
What I do: Lead cost pages with a realistic range and what moves it up or down, before anyone else sets the anchor for you.
06Comparing options
The famous jam study found fewer choices led to more buying. But a later review of 50 experiments found the average effect was close to zero. It depends on the situation.
'Fewer options' isn't a rule. What helps is making the choice easy, however many options there are.
What I do: On comparison pages, give a clear pick for each type of buyer instead of a list of twenty with no opinion.
I'm a marketer, not a psychologist. This is published research, linked so you can read it yourself, applied to search the way I'd apply it to your pages.
Method: Buyer State Mapping
People search differently depending on how close they are to buying, and each search needs a different page. Most sites send all three to the home page.
1Has a problem
'signs you need a new roof'
Trades
2Comparing options
'xero vs myob for small business'
B2B
3Ready to buy
'physio gosford saturday'
Health
What SEO covers
SEO isn't one job. Most sites need some of these far more than others. Here's what each part is, and when I'd put the effort there.
Making sure Google can find, read and store your pages: speed, structure, redirects, sitemaps.
Best for: Sites with lots of pages, online stores, and anything recently rebuilt or moved.
Writing and structuring each page so it's the best answer to the searches it's for.
Best for: Almost everyone. Usually where the biggest gains are.
Your Google Business Profile, reviews and suburb pages, so you show in the map results.
Best for: Trades, clinics, cafés and anyone people call or visit locally.
Being the source ChatGPT, Gemini and Google's AI Mode pull from and name.
Best for: Anyone whose buyers ask questions before they buy. Google says it needs no special files, just pages worth quoting.
Other sites linking to or talking about you, earned through things worth mentioning.
Best for: Competitive searches where the pages are already good. Earned, never bought.
How it works
I check how you show up in Google and AI search, and send the top 3 fixes. Then we talk them through in a 20-minute call.
Before writing anything new, make sure Google can find, read and keep the pages you've already got.
Work out your audiences from your data, then one strong page for each search that leads to customers, angled to the person searching it.
Every month: what moved, what didn't, and what I'm doing next. In plain English.
What you get
Signals this works on
Out of the 8 signals every page and ad sends.
From your own data first: the words people typed in Search Console, the search terms that turned into sales in Google Ads, how people move through your site, and what buyers say in reviews. Each audience then gets checked against real searches. If nobody is searching the way that audience supposedly would, it's dropped.
It depends on how many pages need work and how competitive your searches are. The free audit tells you where the effort should go first, so you are not paying for work that will not move anything.
Google doesn't give a timeframe. Typically, technical fixes on an established site show up within weeks, and new pages take a few months to settle. New sites with no history take longer. Be wary of anyone promising results in a fixed number of days. [DAN TO CONFIRM: your usual timeframe]
No, and nobody can. Google says no one can guarantee a #1 ranking, and to look elsewhere if someone promises one.
Mostly no. Google says its AI features don't need special files like llms.txt or special markup. What helps is the same thing that helps SEO: clear pages that answer the question and the follow-up questions, with something worth quoting.
Google Ads brings traffic as soon as it is switched on and stops when you stop paying. SEO takes longer but keeps working. In my experience, most businesses do best with both: ads data shows which searches turn into customers, then SEO is built around those.
I earn links and mentions with things other sites want to point to: original data, useful tools, local stories. I don't buy links. Google treats paid links meant to manipulate rankings as spam, and it isn't worth risking your site for.
Yes. Researched and drafted with AI tools, then checked by me, with your knowledge added. The parts only you know, like what you see on real jobs, are what make a page worth ranking.
I check how Google finds and stores your pages, which searches you show up for and which you miss, how your main pages compare with the top results, and how you appear in AI answers. You get the top 3 fixes in writing and a 20-minute call.
No. I'm based on the Central Coast, NSW and work with businesses across Australia, including Sydney.
Takes a minute. You'll hear back within 1 to 2 business days.
What happens next
Keep going