Module 3: Who Sees Your Ads
Targeting is where beginners either waste money talking to everyone or strangle a good campaign with an audience of 4,000 people. This module walks through defining your customer, choosing between interest and broad targeting, and using your warmest audiences well.
Defining your customer on paper
Before you touch a single targeting setting, get a blank page and write down who you're actually trying to reach. Not "everyone who might buy" — one specific person. Demographics first: age range, location (how far will they travel or do you ship?), and anything else that genuinely matters, like homeowner versus renter for a roofer. Then the part most people skip: what problem are they trying to solve right now, what have they already tried, and what's stopping them from buying?
Finish with this exercise: write three sentences your customer would actually say out loud. "I need someone who won't butcher my dog's face." "I keep meaning to book but I forget." "Is $85 crazy for a haircut for a dog?" Those sentences become your ad hooks in Module 4. The pitfall here is defining people only by demographics and interests while ignoring the problem layer. Two 35-year-old women in the same zip code can be in completely different buying modes — one is researching, one has her credit card out. Your targeting and your copy should aim at the one with the card out.
Ideal customer profile — copy and fill in:
Age range: ___
Location: ___
What they want (in their words): "___"
What frustrates them now: ___
Where they spend time online: ___
What would make them buy today: ___
Build brief
What's the exact objection that has stopped your best customer from buying so far? Write it in their words — that's your ad angle. Now turn it into one headline.
Saved
Interest targeting vs broad targeting
Interest targeting means hand-picking who sees your ads: people interested in "meal prep" and "CrossFit," say, aged 25 to 44, in your metro area. Broad targeting means setting only the basics — location, age, gender — and letting the platform's system find the buyers using your pixel and conversion data. Both are legitimate. The honest truth in 2026 is that broad usually wins once the system has conversion data to learn from, while interests are useful when you're starting cold or your niche has unusually clear signals.
A few guardrails. On Meta, avoid narrowing an audience below about 500,000 people unless you're retargeting — tiny cold audiences burn out fast and the algorithm can't learn. On Google Search, "targeting" is mostly the keywords themselves, so broad versus narrow plays out in your keyword match types instead (more on keyword structure in the launch module). And resist the urge to exclude aggressively: excluding competitors' fans, excluding everyone who ever visited your site, excluding three age brackets — each exclusion shrinks the pool and raises your costs. Exclude only when you have a real reason, like keeping existing customers out of a new-customer offer.
Targeting everyone 18–65 in the whole country. Broad targeting with a small budget means your ad shows to thousands of wrong people and learns nothing. Start narrow, prove the message, then widen.
Build brief
What existing audiences do you have to work with — past website visitors, an email list, social followers, video viewers? List everything you've got.
Saved
Retargeting and lookalike audiences
Lookalike audiences go the other direction: you hand the platform a "seed" list of your best customers, and it finds new people who resemble them. The seed quality decides everything. A lookalike built from 2,000 past buyers will outperform one built from 50,000 casual page viewers every time — buyers resemble buyers, browsers resemble browsers. On Meta, start with a 1 percent lookalike of your country (about 2 million people in the US), built from purchasers or qualified leads. On Google, the equivalent is similar audiences and Customer Match lists uploaded from your email subscribers.
Structure matters: keep cold and warm audiences in separate ad sets or campaigns with separate budgets. If you mix them, the platform will spend most of your money on the easy retargeting wins, your cold prospecting starves, and you learn nothing about new customer acquisition. Give retargeting a smaller budget — those audiences are small and don't need much. Watch frequency on retargeting like a hawk: when the average person has seen your ad 8 or 10 times, you're not persuading anymore, you're annoying. Rotate the creative or widen the window before it gets there.
Know your audience
Warmer audiences convert better
Example for illustration — not real results.
Your path
Website visitors
Past visitors from the last 30, 60, and 180 days as separate groups.
Video viewers
People who watched 50 percent or more of your videos.
Social engagers
Engagers on your Facebook or Instagram profiles.
Lookalikes
1 percent of your country, built from purchasers or qualified leads.
Build brief
Are you starting cold with no conversion data, or do you have purchase history to feed the algorithm? How does that change your targeting plan?
Saved
Write a one-paragraph description of your ideal customer — age, location, what they want, what keeps them up at night. If you can't, you're not ready to target.
Which seed list makes the strongest lookalike audience?
Your turn: Define your first test audience — age range, location, three interests, and one sentence on why they'd buy.
Saved.