2 Geeks Web Design
Schema markup strategies that lift small business search visibility
Walk into any Australian chamber of commerce and you will hear the same grumble. The website looks decent, the photos are sharp, the prices are sharp too, yet the inquiry form sits empty while bigger operators dominate the map pack. The gap is rarely the design. More often it is the invisible layer underneath the page, the structured data that tells Google what the content really means.
Structured data, also called schema markup, is a snippet of code that labels the parts of your page in a language every major search engine already speaks. When a search bot reads your homepage, schema helps it recognise a street address, a phone number, a menu, an event date, or a product price without guessing. The payoff appears as rich snippets: star ratings, opening hours, FAQ dropdowns, sitelinks, and price tags in green, all the polish that pushes a plain blue link down the screen.
This guide covers how schema works, which formats deliver the most value for Australian operators, and the hands-on steps to add, test, and maintain it on a hand-built website. Whether you run a dental clinic in Parramatta, a kayak hire in Cairns, or a roastery in Fremantle, the same building blocks apply.
How schema markup actually works
Schema is a shared vocabulary created by Schema.org, a joint project between Google, Bing, Yahoo, and Yandex. Think of it as a tagging system. A normal HTML page might read "16 Smith Street, Parramatta NSW 2150." Add schema and that line becomes a typed item with a street address, a suburb, a state, a postcode, and optional geo coordinates, all explicit and machine-readable.
Three syntaxes have appeared over the years: Microdata, RDFa, and JSON-LD. The first two live inside the existing HTML, wrapping each piece of content in extra attributes. JSON-LD sits in a separate script block, usually in the head or before the closing body tag, and describes the page in plain JSON. Google officially recommends JSON-LD because it is cleaner to write, easier to audit, and stays intact when designers restyle the visible markup.
The end goal is not a guaranteed jump in position. Schema does not hand you a slot at number one. What it changes is how your listing looks on the results page, which lifts click-through rate, which over time lifts ranking. A treatment room in Adelaide with star ratings, opening hours, and a phone button can pull twice the clicks of an identical listing with none of that polish.
Schema types worth implementing first
Not every schema type earns its keep. Some are quietly deprecated, others go unnoticed by Google. For a small Australian business, the following ten or so types cover the great majority of useful cases. The comparison below lays out the strongest options by purpose, ideal page, and the rich result they unlock.
| Schema type | Best applied to | Rich result unlocked |
|---|---|---|
| LocalBusiness | Contact or home page | Map pack card with hours and directions |
| Restaurant | Menu or venue page | Menu links, reservation buttons, price range |
| Product | Individual product or service pages | Price, availability, review stars |
| FAQ | Service pages with common questions | Expandable Q&A on the results page |
| HowTo | Step-by-step guides or tutorials | Numbered steps with optional images |
| Article | Blog posts, news, opinion | Headline, author, publish date in a vertical layout |
| Event | Workshops, gigs, markets | Date band, venue, ticket link |
| Recipe | Recipe blog content | Image, cook time, ratings, nutrition |
| BreadcrumbList | Any multi-level page | Sitelink-style path under the headline |
| Service | Trades and professional services | Service name, area served, provider details |
LocalBusiness is the highest-leverage type for almost every Australian operator with a physical presence, since it feeds the Google Business Profile integration that powers the map pack. Restaurant is a specialisation of LocalBusiness and unlocks extra features, including menu display and reservation actions directly on the results page. Anyone running a food business should look at how to create a custom website for a restaurant as a starting point, since the template decisions made early on determine how cleanly the schema slots in later.
FAQ and HowTo schemas are particularly strong for service businesses. A mortgage broker in Brisbane who publishes a "First Home Buyer FAQ" page can capture the question and answer inside the search listing, pushing competitors further down visually. A pest control outfit in Hobart with a "How to Prep Your Home for a Possum Inspection" page can take the steps block and stop users from clicking through to find the actual instructions.
Adding markup to a hand-coded site
A hand-built site, the kind a small studio produces for a tradie in Geelong or a florist in Newcastle, gives total control over where the schema lives. The cleanest pattern is to add a single script block per page, either inline in the template or pulled in via a partial, and populate it with JSON-LD.
A LocalBusiness example for a plumbing business in Geelong might cover the type, the name, an image URL, the phone number in international format, the full address with state and postcode, geo coordinates pulled from Google Maps, opening hours expressed as a structured array, and a sameAs block for the social profiles. Each field is a known property from Schema.org, so any validator will flag typos the moment the page is crawled.
For a service area business that travels to customers rather than receiving them, the areaServed property accepts a list of suburbs or regions. A Geelong plumber who covers the Bellarine Peninsula, the Surf Coast, and parts of Melbourne's west can declare each locality explicitly. This signals coverage to the search engine and helps the listing appear for queries in those locations.
For organisations whose offerings change seasonally, think caravan parks in Queensland or ski lodges in the Snowy Mountains, schedule updates to the Event or Offer markup alongside the homepage refresh. JSON-LD is just text, so a developer can store it in an include file and regenerate it whenever pricing or dates move.
Testing and monitoring your structured data
Adding schema is half the job. The other half is making sure it parses correctly and keeps doing so after template updates. Three tools cover nearly every check.
Google's Rich Results Test accepts a URL or a code snippet and shows whether the page qualifies for any rich result, which schema types were detected, and any errors or warnings. It is the fastest feedback loop during initial development. The Schema Markup Validator, maintained by Schema.org itself, performs a deeper structural review and flags properties that do not match the official definitions. Search Console gives the long view. The Enhancements section reports which pages Google has indexed with valid markup, which ones have issues, and how coverage trends over time.
For Australian businesses, the Privacy Act 1988 and the Spam Act 2003 shape what user-generated content ends up in markup. If a testimonial is collected without clear consent, marking it up as Review or aggregateRating can expose the business to complaints handled by the Office of the Australian Information Commissioner. A conservative rule is to mark up only reviews collected through a documented opt-in, with the option to withdraw.
Re-validate after every meaningful site change. Template restructures, CMS migrations, and even a stray DOM reorder can strip script tags or shift their content. A monthly crawl with a tool like Sitebulb, configured to flag missing or malformed schema, keeps the layer invisible to humans and obvious to bots.
Habits that keep your markup healthy over time
Schema decays slowly, then suddenly. A restaurant adds a new menu section and forgets to update the Menu schema. A retailer swaps product images and forgets to keep the image URL in the product markup valid. Within weeks, Google de-prioritises the rich result and the listing returns to plain blue links. The fix is light, but it has to be routine.
Treat the structured data as part of the editorial workflow, not a one-off engineering task. When a writer edits a price, an hours block, or a service description, the schema block in the template goes through the same review. For shops that update stock weekly, a small automation that regenerates the Product JSON-LD from the product database removes human error entirely.
Be wary of markup that promises too much. Marking up an editorial review as a Customer Review, or adding aggregateRating when no genuine rating system exists, breaches Google's spam policies on structured data and can trigger a manual action that quietly demotes the entire site for months. The Australian Consumer Law, administered by the ACCC, reinforces the same principle at the human level. False or misleading representations, including faked reviews exposed through markup signals, draw penalties that apply regardless of where the schema was published.
For a working example of a custom-built project where every detail is documented and every component is intentional, take a look at the Astromech build journal. It is unrelated to SEO, but the discipline of naming each subsystem, versioning each file, and reviewing each change is the same mindset that keeps structured data accurate once a site goes live.