Paid Media Strategy: How to Build One That Pays Back

Advertising

Reading Time: 10 minutes

Share this article

A paid media strategy is a written decision about which business outcome you are buying, from whom, on which channels, with how much money, and how you will know it worked. If your plan starts with “we’ll run Meta and Google”, you have a shopping list, not a strategy.

Most Australian brands we meet have the list. Few have the other four parts. This guide gives you all five, with the numbers we can source and a plain note where nobody has a number.

Key takeaways

  • Paid media is anything you pay to place in front of an audience, from a TV spot to a Google search click. It stops delivering the day the spend stops (Media Federation of Australia, 2024).
  • Set one business objective first. “Sales now” and “change how people see us” need different channels, creative and metrics.
  • Search and video took most of the growth in Australia in 2025: search hit $8.0 billion and video $5.4 billion (IAB Australia, March 2026). Big markets are not your market. Test where your buyers are.
  • Budget by the job each channel does, then pace it across the flight. Do not split it evenly because that feels fair.
  • The click is half the job. Weak landing pages waste good traffic, and platform-reported conversions overstate what the ads caused.

Table of Contents

What a paid media strategy is (and what a paid media plan adds)

The strategy is the argument. The plan is the schedule that carries it out.

The strategy answers four questions: what outcome, for which customer, through which channels, and under what rules for judging success. The plan turns those answers into a calendar: which campaigns run in which weeks, at what daily spend, with which creative, and who checks what on which day.

Paid media strategy mapped to its execution plan

Most thin guides skip this split, and the cost shows up later. A plan with no strategy produces busy dashboards and no decisions. A strategy with no plan produces a good slide deck and an under-spent account.

What counts as paid media

The Media Federation of Australia defines it as media you pay to place an ad in, from buying a spot on a TV network to paying when someone clicks a Google search ad. It sits next to earned media (PR and editorial coverage), shared media (people commenting and sharing) and owned media (your website, email, packaging and store signage) (MFA Guide, 2024).

The MFA’s own view of the trade-off is useful. Paid media gives you control over who sees what, is easier to measure than the other types and can be tuned while live. It is also trusted less than earned exposure, and it goes quiet when you stop paying.

Sit with that last point. Paid media is rented attention. That is fine, as long as the rent buys something your owned channels can keep: an email address, a repeat purchase, a customer who searches your name next time.

Step 1: Pick one objective and put a number on it

The Australian IAB’s effectiveness playbook starts in the same place: define the business challenge before you pick benchmarks or metrics, for example driving an immediate response versus shifting perception (IAB Australia, Digital Ad Effectiveness Playbook, 2017). The paper is old. The logic is not.

Turn it into a sentence with a number, a time and a constraint:

  • “Acquire 1,200 new customers in 90 days at a blended cost per acquisition under $48.”
  • “Lift first-purchase conversion on paid traffic from 1.6% to 2.2% by the end of next quarter.”

Those figures are examples of format, not benchmarks. Your numbers come from your margin. Work backwards: average order value, gross margin, the repeat rate you can prove, and the most you can pay to win a customer and still be in profit by month three or six.

Most vague practices start here. “Brand awareness” with no measure, no time window and no link to revenue is a wish. We say so to clients in the first call.

Pick the audience from your own data

The ACCC’s digital advertising inquiry describes the standard approach: first-party data such as site visits, past purchases and loyalty membership, often topped up with third-party segments (ACCC, Digital Advertising Services Inquiry interim report, 2020).

The practical version: export your last 12 months of orders. Find your top 20% of customers by lifetime value. Ask what they bought first, where they came from and what they have in common. That group, not a demographic guess, is your seed audience.

Step 2: Choose channels by the job they do

Online options include search, social, sponsored newsletters, influencers, native placements and ad networks (Digital Transformation Hub). Shopify Australia lists search, social, display, video, retargeting, native, shopping and automated campaign formats (Shopify AU, April 2026). Offline, you also have TV, radio and out-of-home.

A list of channels does not help you choose. A list of jobs does.

JobTypical channelWhat it buysMain risk
Catch existing demandGoogle Search, ShoppingBuyers already typing the productCapped by how many people search
Create demandMeta, TikTok, YouTubeNew buyers who did not know they wanted youSlower, harder to measure
Win back the almost-buyerRetargeting, email and SMSCheap repeat exposureOverstated credit, since many would have bought anyway
Build reach at scaleConnected TV, broadcast, radio, outdoorMass familiarityNeeds budget and patience

Why search alone stalls

Search is the biggest slice of Australian online advertising: $8.0 billion in 2025, up 11.5% (IAB Australia, March 2026). But search only harvests people who already know what they want. When you have squeezed every profitable keyword, the only way to grow is to create more people who search. That is the job of video and social.

The same report puts video at $5.4 billion, up 19.8%, and social video at 40% of video investment. Money is moving to where attention is. That tells you where the competition is getting crowded. It does not tell you where your customer is.

Pick two, not seven

A new account should run two channels properly before it runs five badly. Each channel’s algorithm needs enough conversions to learn from, and a thinly spread budget starves all of them. If you can only afford two, pair one demand-capture channel (Google) with one demand-creation channel (Meta or YouTube). If you are deciding how to split a Google budget between automated and manual campaign types, our breakdown of PMax vs Search budget splits covers the mechanics.

Step 3: Allocate the budget

Shopify’s media planning guide sets out the sound order of operations: allocate by objective, audience, channel role, timing and expected return, then pace spend across the flight so it is not used up early, and move money when live results show a better home for it (Shopify AU, 2026).

Here is how that looks in practice.

  1. Fix the total from the unit economics. Take your target cost per acquisition times the customers you want. That is the spend that hits the objective, whether or not it fits last year’s budget.
  2. Hold back a test reserve. We suggest putting a defined slice of the total into experiments: new audiences, new creative angles, a new channel. Call it a rule of thumb, not a law. Nobody has a universal right number, and anyone quoting one for all brands made it up.
  3. Split the rest by job. Fund demand capture to the point where the marginal customer stops being profitable. Put the remainder into demand creation.
  4. Pace it. A $60,000 quarter is not $20,000 a month by default. If your sales peak in November, front-loading the spend in September wastes the money when it is cheapest to learn and starves you when buyers are most ready.
  5. Set a reallocation rule before launch. For example: “If a campaign runs 14 days over target cost per acquisition with at least 50 conversions, cut its budget by a third and move it.” The numbers are yours to set. The point is that the rule exists before emotion does.

Step 4: Fix the creative and the landing page before you scale

This is where most plans quietly fail. The ad earns the click. The page earns the sale. If the page loses six of every seven visitors, doubling your ad spend doubles your waste.

Three coordination rules cost nothing:

  • Match the promise. The offer, product and wording in the ad must be the first thing on the page. A mismatch reads as a bait-and-switch and buyers leave.
  • Give each audience a page built for it. A cold Meta viewer needs proof and context. A returning searcher needs price, shipping and a buy button.
  • Test the page on the same cadence as the ad. If you rotate creative weekly but leave the page untouched for a year, you are optimising half the funnel.

We cover the common failure points in why your digital ads aren’t working.

The creative mechanism most teams miss

On Meta, the system tries to show each person the ad most likely to produce a result. If you load ten ads that share one script and one angle, with different thumbnails, the platform has effectively one idea to test. Real variety means different hooks, different customer problems and different formats. That is why we plan creative as a set of distinct angles before we brief a single shoot. For how we scale winners once you have them, see our advanced Facebook Ads scaling blueprint.

(Variety ≠ volume.)

Step 5: Measure what the ads caused, not what they claimed

The IAB playbook says to choose metrics that show whether the objective was met, then examine results to improve (IAB Australia, 2017). The AANA’s submission to the ACCC names the standard set: impressions, viewability, reach, frequency, engagement and conversion (AANA, March 2021). It also backs independent verification of the platforms that buy your ads.

Use that set as a dashboard, not a verdict. Each metric answers a different question:

  • Reach and frequency: are you hitting enough different people, or the same few over and over?
  • Viewability: was the ad on screen at all?
  • Conversion and cost per acquisition: did it pay?
  • New-customer share: how many buyers were new rather than people who already knew you?

Platform numbers flatter the platform

Every ad platform claims credit for conversions inside its own window. A person who searched your brand name, clicked a retargeting ad and bought gets counted by both. Add up the platforms’ own claims and the total often exceeds your actual orders. We have not seen a sourced Australian figure for how large the overlap is, so we will not give you one. Check yours: compare the sum of platform-reported conversions to orders in your store for the same week.

The fix is incrementality testing: switch ads off for a matched group or region, and measure the gap in sales. It takes more effort than reading a dashboard, which is the reason it is trusted. Our write-up of incrementality testing for Meta ads shows two ways to run it.

Caveat. Holdout tests need enough volume to read. A brand with 30 orders a month will not get a clean answer in two weeks. In that case, track blended cost per acquisition (total ad spend divided by total new customers) and watch the trend over 8 to 12 weeks.

Platform numbers flatter the platform — overview diagram

A worked example: $30,000 over one quarter

This is an illustration of the method, not a client result.

An Australian skincare brand has an average order value of $70 and a gross margin of 65%, which is $45.50 per first order. It can tolerate losing money on the first order if repeat purchases pay it back, but the founder wants break-even on the first order within 90 days. So the ceiling on cost per acquisition is about $45.

  • Objective: 600 new customers in 90 days at or under $45 blended. That is $27,000 of spend.
  • Budget: $30,000. The extra $3,000 is the test reserve, a tenth of the total.
  • Channels: Google Search and Shopping to catch existing demand; Meta for demand creation, with four distinct creative angles.
  • Rule: after 14 days, any campaign over $60 per acquisition with enough conversions to read gets cut by a third.
  • Measure: weekly blended cost per acquisition, new-customer share, and one holdout test in month two.

Every number there traces to the margin and the target. That is the property you want in a plan.

Our perspective

We built Thrive around one complaint: agencies that split strategy, creative, media buying and conversion work into separate teams create gaps between them. The strategist writes a brief that the creative team reinterprets. The media buyer gets ads that do not match the page. Nobody owns the number that matters, which is revenue.

So we run it as one team. Brand and creative strategy, integrated campaigns, ad production, paid media and conversion work sit under one plan and one scoreboard. When a campaign underperforms, the person who diagnoses it can change the ad, the audience or the page, not file a ticket with someone else.

Our advice to any brand, whether it hires us or not: pick the business number first, then argue about channels. Be wary of anyone who sells you a channel before they ask about your margin. That is a pitch, not a strategy.

— Harry

Where Thrive fits

If you want this built and run for you, we manage paid media across Meta and TikTok, Google and YouTube, with ad creative production and conversion optimisation in the same team. You can see how that looks on a real account in the Bonaire case study, or start from our full services overview. We do not publish prices, because scope depends on your channels, creative load and goals.

FAQ

What does paid media include?

Anything you pay to place in front of an audience: TV and radio spots, billboards, search ads, social ads, video ads, display banners, sponsored newsletters and influencer placements. The MFA separates it from earned, shared and owned media, which you do not pay a media owner to place (MFA Guide, 2024).

Is SEO or paid search better?

Neither wins in general. Paid search gives you traffic the day you switch it on and stops when you stop paying. SEO takes longer to build and keeps working without per-click costs. Brands that need sales this quarter usually start with paid and build organic alongside it.

How is a paid media plan different from a strategy?

The strategy sets the objective, audience, channel roles and success rules. The plan is the calendar and the budget: what runs when, at what spend, with which creative, and who checks what. You need the strategy first, or the plan has nothing to defend.

How long before I know if paid media is working?

It depends on your volume. Platforms need a run of conversions to learn, so a low-volume account reads slowly. Set the review dates and the minimum conversions per decision before launch, then judge on trend over several weeks, not a single day.

Sources

We'll 3X your investment in us

That's our pinky promise to you! You'll see measurable improvement in results within the first 90 days with better ads and higher conversions.

Keen to have us as your performance marketing partner? Please click the "partner with us" button to see if we're a good fit.

more from our desk