The time-poor parent
'Scrolling at 9pm, Saturday already full of chores'
- Worried about
- I'm going to lose another weekend to the house.
- The angle
- Lead with the time back. The cleaning is secondary.
- How the ad opens
- Your Saturday, back.
Social media
marketing.
Social media marketing, the way I do it, means paid ads on Facebook and Instagram that reach people likely to buy, and showing which ones actually made sales. I run the ads, not your daily posting. I'm a Meta ads consultant on the Central Coast, NSW, working with businesses across Australia. Meta's system now decides who sees your ad mostly from the ad itself, so the creative does the targeting. Most accounts I look at aren't short of audiences. They're short of different ideas, and they're split into so many pieces that none of them gets enough data to learn.
Written by Dan Picken, founder. Last checked 27 October 2026.8 ad sets
Learning limited
About 6 results each a week. None of them gets enough to settle.
1 ad set
Learning complete
About 50 a week from the same money. Enough for Meta to learn who buys.
Example numbers. Meta's guidance: about 50 results per ad set in 7 days.
Audience first
Target with the ad,
not the settings.
Meta now decides who sees your ad mostly from the ad itself. So the audience work moves out of the targeting boxes and into the creative. I work out who your buyers really are from your own data, give each group its own angle, and let Meta match each ad to the people it fits.
The data I use
- Your customer listWho actually buys and who is worth the most. Only when you can share it, and kept private.
- Meta's breakdownsThe ages, regions and placements your real sales came from, not the ones you targeted.
- Ad comments and messagesObjections and questions in the exact words people use.
- Search termsWhat people type into Google before they buy: the worries they would never say in a comment.
- Reviews and forumsWhat made past buyers say yes, and what nearly stopped them.
- Your sales dataWhich ads brought repeat buyers, not just first orders.
How it turns into pages
- 1
Find the audiences
Group your buyers by what they want and what worries them, not by age bracket.
- 2
One angle each
Each audience gets its own ad idea, so the creative does the targeting.
- 3
Read it back
Check which people each angle actually reached and sold to, using Meta's breakdowns and your sales. Angles nobody responds to get dropped.
Worked example: home cleaning
Same service, two people scrolling for very different reasons. Two ads that open differently:
The landlord between tenants
'Reading the end-of-lease inspection checklist'
- Worried about
- Will this actually pass the final inspection?
- The angle
- Lead with the checklist, room by room. Show the oven and the skirting boards.
- How the ad opens
- Built around the end-of-lease checklist, room by room.
One generic ad half-suits both. Two angled ads let Meta find each person, and the results show which audience is worth more to you.
From my audits
Where most accounts
leak money.
Meta will happily spend whatever you give it. In the accounts I review, this is where it goes without coming back.
- 1
Too many ad sets
The budget is split across lots of audiences, so no ad set gets enough results to learn. Everything stays 'Learning limited'.
- 2
Ten ads, one idea
Different colours and headlines on the same message. To a buyer, and to Meta, that is still one ad.
- 3
Optimising for the cheap thing
Clicks or form fills get reported as wins. The system learns to find people who click and fill in forms, not people who buy.
- 4
The same ad for months
One ad that worked in March is still running in August. The people who'd respond to it already have.
- 5
Leads nobody follows up fast
Instant form leads come in, sit in a spreadsheet for two days, then get called. By then the person has forgotten filling it in.
- 6
Trusting Meta's own numbers
Meta reports the sales it thinks it caused. Without checking against real sales, there's no way to tell what was extra and what would have happened anyway.
How Meta picks who sees your ad
Your ad is
the targeting.
Meta's system first narrows tens of millions of possible ads down to a few thousand for each person, then runs an auction. The winner isn't just the highest bid. It's the ad with the highest total value: your bid, how likely that person is to act, and the quality of the ad.
- 01
Picking candidates
Before any auction, Meta shortlists the ads most likely to suit each person, using what is in the ad itself.
What I do: Make ads that are genuinely different, so each one can find a different kind of buyer.
- 02
Your bid
What you are willing to pay for the result you chose.
What I do: Set it from what a sale or lead is really worth, not what you'd like it to cost.
- 03
Likelihood to act
Meta's estimate of how likely this person is to do the thing you optimise for.
What I do: Optimise for the result you want, with real sales or qualified leads reported back.
- 04
Ad quality
How people respond to the ad, including hiding it or reporting it.
What I do: No clickbait, no withheld information. It costs you in the auction.
- 05
Learning
Each ad set needs enough results each week for delivery to settle.
What I do: Fewer ad sets with more budget each, and fewer big edits.
Worth knowing: Meta's guidance is about 50 results per ad set within 7 days to finish learning, and big edits like new ads, new targeting or big budget changes restart it. The tinker-every-day approach keeps an account learning forever.
Method: Spend Curve
Is your budget
stuck learning?
Meta's guidance is that an ad set needs about 50 results within 7 days to finish its learning phase. Below that, it stays 'Learning limited' and delivery is less stable. The usual cause isn't too little money. It's the same money split across too many ad sets. Put in your numbers.
Cost per result: use what you pay now for the thing you optimise for. If you don't know it, start with your cost per lead or per sale from last month.
Results per ad set, per week
7/ 50
- Daily budget each ad set needs
- $179
- Ad sets this budget can feed
- 0
Learning limited. Each ad set gets about 7 results a week. Cut to 1 ad sets, or optimise for an earlier step.
Meta's 50 is guidance, not a hard cut-off. Useful as a sense check, not a law.
Consumer psychology
The psychology
of a scroll.
Nobody goes on Instagram to see your ad. Every Meta ad is an interruption, and whether it works comes down to a few split-second reactions that have all been studied. Here's what the research says at each moment, and what I do about it.
01The first frame
The quarter-second test
Facebook's own 2017 research found people can recall mobile feed content after just 0.25 seconds, and scroll 41% faster on a phone than a desktop.
Most people never see past the first frame. If the idea is not there, it is not anywhere.
What I do: Put the brand and the one idea in the first frame, readable with the sound off.
02Stopping the scroll
The isolation effect
Since von Restorff in 1933, studies have found that things which stand out from a sequence are remembered better than the things around them.
A feed is a sequence. An ad that looks like the posts around it disappears into them.
What I do: Design to stand out from the feed, not just from competitors. One clear break in pattern per ad.
03Seeing it again
Familiarity, with a ceiling
A 2017 review of the research found liking rises with repeat exposure, then falls: an upside-down U.
Seeing your ad a few times helps. Seeing it every day for a month works against you.
What I do: Watch frequency and results together, and bring in new creative before the old one tires.
04Deciding to trust it
Social proof, the specific kind
In the original hotel study, guests reused towels most when told what guests in that same room did. A later replication found the effect less consistent, so treat it as worth testing, not a sure thing.
Proof from people like them beats general praise. Local and similar beats famous.
What I do: Show proof from the same suburb or the same kind of customer. Never in health ads, where testimonials aren't allowed.
05Coming back to buy
The endowment effect
Kahneman, Knetsch and Thaler found people value something more once it feels like theirs, asking far more to give up a mug they'd been given than others would pay for it.
Someone who left a cart is weighing what they'd give up against what they'd pay.
What I do: Remind them what they picked and answer the reason they stopped. Real deadlines only, never fake countdowns.
I'm a marketer, not a psychologist. This is published research, linked so you can read it yourself, applied to ads the way I'd apply it to yours.
Method: Buyer State Mapping
Nobody searched.
So the ad has to start it.
Meta ads interrupt people who weren't looking. What the ad needs to say depends on how far along they are, and most accounts show everyone the same thing.
1Not looking yet
'Scrolling Reels after work'
Health
- What they're thinking
- My back's been stiff for weeks. It'll sort itself out.
- What the ad should do
- Name the feeling, not the treatment. Health ads can't use outcome testimonials, so show the clinic and the people instead.
2Thinking about it
'Saw the ad twice, checked the page'
Hospitality
- What they're thinking
- Is this place actually good? Will it suit a group of ten?
- What the ad should do
- Show the room, the menu and the booking button. Answer the group question before they ask.
3Nearly bought
'Added to cart, closed the app'
E-commerce
- What they're thinking
- Is shipping going to blow out the price?
- What the ad should do
- Answer the reason they stopped: delivery cost, delivery date, returns. Not a generic 10% off.
Campaign types
Which campaigns,
and when.
Each Meta campaign starts with a goal, and the goal decides who Meta shows your ad to. Pick the wrong one and it will find exactly the wrong people, very efficiently.
Sales
Finds people likely to buy, using sales reported back from your website or store.
Best for: Online stores, and anyone who can report sales back to Meta. Usually the main campaign.
Leads
Collects enquiries, either on your website or with a form inside Facebook and Instagram.
Best for: Service businesses. Instant forms are easy to fill in, which also means more low-quality leads.
Engagement and messages
Gets people to message you, call, or interact with a post.
Best for: Businesses that sell in a conversation: hospitality bookings, quotes, local services.
Awareness
Shows your ad to as many of the right people as possible, as cheaply as possible.
Best for: New brands, launches, events. Judged on reach, not sales.
Traffic
Sends people to a page, optimised for clicks.
Best for: Rarely. It finds people who click, which isn't the same as people who buy.
How most accounts run.
How I run them.
Most accountsHow I do it
TargetingLots of ad sets with stacked interests.Broad audiences in fewer ad sets, and different creative doing the targeting.
CreativeOne hero ad in a few colours.A different angle for each buyer worry and each stage.Buyer State Mapping
BudgetSpread thin so everything gets a go.Enough per ad set to finish learning, and a clear point where more stops paying.Spend Curve
GoalCheapest clicks or form fills.Real sales and qualified leads, reported back to Meta from your store or CRM.
ChangesTweaked every day.Changed in planned batches, so the system can learn between them.
ReportingMeta's reported return on ad spend.Checked against real sales, with tests that show what the ads added.Cause Testing
How it works
Four steps.
Nothing hidden.
- 01
Free audit
I go through your ads account and send the top 3 fixes, then we talk them through in a 20-minute call.
- 02
Fix the tracking
Make sure Meta is told about real sales and good leads, from your website and your CRM, not just page views.
- 03
Build the creative
New ads built around what your buyers worry about at each stage, so each one finds a different kind of buyer.
- 04
Run it monthly
Every month: what worked, what's tiring out, and the next ideas to test. In plain English.
What you get
- Ads account audit and rebuild plan
- Meta Pixel and Conversions API checked, with sales or qualified leads reported back
- Audience angles built from your own data, one ad idea per audience
- Account structure that gives each ad set enough to learn
- Ad ideas and scripts for each buyer worry and stage
- Simple illustrated and animated ads made in-house, in a consistent drawn style. For filmed video or photos, your own footage or a creator brought in
- Monthly testing of new creative
- A monthly report in plain English
- Direct access to me, start to finish
Signals this works on
Out of the 8 signals every page and ad sends.
- Pattern interrupt
- Emotional activation
- Relevance signal
- Trust architecture
Questions
people ask.
How much does Meta ads management cost?
There are two parts: your ad spend, which you pay to Meta, and the management fee, which depends on how many campaigns and how much new creative is involved. The free audit shows whether your current spend is set up to work before you add more.
How much should I spend on Facebook ads?
Enough for each ad set to get about 50 results a week, Meta's guidance for finishing the learning phase. The calculator on this page works it out from your cost per result. If the number is out of reach, use fewer ad sets or optimise for an earlier step.
How long before Meta ads work?
Ads can bring results in the first week. Each ad set then needs about 50 results in 7 days to settle, so judge an account on a couple of months, not a couple of days, and don't keep editing it while it learns.
Facebook ads or Google Ads?
Google Ads catches people already searching. Meta ads reach people who weren't looking yet. If people search for what you sell, Google usually comes first. If they buy on impulse or don't know you exist, Meta often does more. Most businesses that can afford both use both.
Are boosted posts worth it?
Usually not for sales. A boost is a simpler campaign with fewer controls, and it often optimises for likes and comments. Fine for getting a post seen. For sales or leads, a proper campaign does more with the same money.
Instant forms or my website for leads?
Instant forms get more leads because they're easier to fill in. That's also why more of them are poor quality. In my experience, if you can call leads back within minutes, forms work. If you can't, sending people to your website usually brings fewer but better leads.
Do I keep my ads account?
Yes. You keep ownership of the ads account, the Pixel and all the data. I work in it as a partner. If you leave, everything stays with you.
What's in the free Meta ads audit?
I check your tracking, account structure, campaign goals, creative and how leads or sales are reported back. You get the top 3 fixes in writing and a 20-minute call.
Do you only work with Central Coast businesses?
No. I'm based on the Central Coast, NSW and work with businesses across Australia, including Sydney.
See what your
account is missing.
Takes a minute. Audits come back within 1 to 2 business days. For anything else, I'll get back to you as soon as possible.
What happens next
- 1I'll ask for read-only access, or just your website for a search audit.
- 2I go through it and write up the 3 biggest fixes.
- 3A 20-minute call to walk you through it. No pitch unless you ask.
Keep going