Ad Copy

Paid Media

Also: Ad Creative Copy · Advertising Copy

What it isThe text inside your paid ads
Main jobStop the scroll, earn the click
Key partsHeadline, description, CTA
Watch forWriting for the brand, not the reader

Quick definition

Ad copy is the written text in a paid advertisement. It includes headlines, descriptions and calls to action across search, social, display and other paid channels. Good ad copy earns attention, communicates a clear reason to click and matches what the audience already wants.

How it varies across Australia

Across the Australian paid media we review, the most common failure in ad copy is not weak writing but weak specificity. Generic benefit statements compete with everything. Ads that name a concrete outcome for a specific audience consistently outperform ones that describe what the product does in the abstract.

See acquisition performance benchmarks across Australian industries

The parts of an ad

Headline

The first thing people read. Its only job is to earn the next second of attention.

Description

The supporting line that adds context or proof to the headline's promise.

Call to action(CTA)

The instruction that tells the reader what to do next. Specific beats generic.

Display URL

In search ads, the URL shown alongside the headline. Often editable and part of the trust signal.

What it actually means

Ad copy is the part of paid advertising that does the actual persuading. The targeting gets your ad in front of the right person. The copy decides whether they stop, read and click.

Every ad has roughly the same structure regardless of platform. A headline (the line that earns attention), a description (the line that earns interest) and a call to action (the instruction that earns the click). On search, the headline is what people scan before they see anything else. On Meta or LinkedIn, the first line often has to compete with everything else in the feed.

The single most common mistake is writing copy that describes the product instead of copy that describes the reader's situation. 'Award-winning project management software' tells me about you. 'Stop running projects in spreadsheets' tells me about a problem I recognise. The second line stops the right person. The first line impresses no one.

Copy also has to match the medium. Search copy is intent-led: the person typed a query and the headline should mirror what they were thinking. Social copy is interruption-led: they were looking at something else and your ad appeared. The job is different. Writing the same way for both wastes budget on both.

The ad that tries to speak to everyone ends up speaking to no one.

How it shows up

Ad copy shows up everywhere a paid placement exists: Google Search headlines and descriptions, Meta feed and story text, LinkedIn sponsored content, display banners, YouTube bumper scripts, programmatic overlays and out-of-home digital screens.

In platform dashboards, copy performance surfaces as click-through rate (CTR), quality score (in Google Ads) and relevance score or ad quality ranking (in Meta). These signals are proxies for how well the copy matches audience intent. A low quality score is often a copy problem before it's a targeting or bid problem.

Copy also shows up in the gap between ad CTR and landing page conversion rate. High-CTR copy that produces low post-click conversion usually means the copy overpromised something the landing page didn't deliver. The click is easy. Earning a conversion on a promise the copy made is harder.

The Australian context

Australian ad copy sits inside a market with stricter consumer-protection rules than the US. The Australian Consumer Law prohibits misleading or deceptive conduct in advertising. Superlatives like 'cheapest', 'best' and 'number one' carry legal risk if they can't be substantiated. Copy that would pass a US review can fail an Australian legal review, and the ACCC (Australian Competition and Consumer Commission) has enforced actively in digital advertising.

Australian audiences also respond less to urgency tactics that feel artificial. Countdown timers, 'only three left' claims and manufactured scarcity read as spam faster here than in equivalent US tests. Directness and plain language outperform pressure tactics in most categories.

Where people get this wrong

Writing headlines that describe features instead of outcomes.People click because they see themselves in the ad. Features are about you. Outcomes are about them.
Using the same copy across every placement and format.Search intent copy and social interruption copy need different voices. A headline that works when someone is actively searching sounds tone-deaf in a passive feed.
Treating the first draft as a final.Ad copy is a hypothesis until the data says otherwise. An A/B test comparing two headlines costs nothing in most platforms and tells you more than any brief does.

Related terms

Common questions

How long should ad copy headlines be?

Google Search allows up to 30 characters per headline. Meta feed ads show roughly 125 characters of primary text before truncation on mobile. Write to the limit that fits the platform, but aim to say the most important thing in the first few words regardless. Scanners read the start, not the end.

How do I know if my ad copy is working?

Click-through rate is the first signal. Quality score in Google Ads and ad quality ranking in Meta both reflect how well copy matches audience intent. But the real test is what happens after the click. High CTR with low conversion usually means the copy promised something the landing page didn't deliver.

Should ad copy match the landing page?

Yes, closely. If the headline promises a free assessment and the landing page leads with a product demo, the visitor feels misled and leaves. Message match between the ad and the landing page is one of the most reliable ways to improve post-click conversion without changing anything else.

How many ad copy variants should I test at once?

Two to three variants per ad set is enough in most accounts. More variants split the data and take longer to reach statistical confidence. Test one variable at a time where possible: two headlines with the same description, or two descriptions with the same headline. Clean tests teach more than messy ones.

Keep exploring

About New Rebellion

New Rebellion is a marketing intelligence consultancy. We build tools, score Australian businesses on how their marketing actually performs, and publish Debrief every day. This dictionary is part of how we work in the open.

How we think →