Paid Social Advertising · Campaign structure, objectives, audiences and automation · lesson 1 of 17 · 14 min
Campaign structure and objectives across platforms
The three-level structure
Almost every paid social platform uses the same hierarchy, even if the names differ:
| Level | Meta (Facebook/Instagram) | TikTok | LinkedIn | X | |---|---|---|---|---| | 1. Campaign | Campaign | Campaign | Campaign group | Campaign | | 2. Targeting and budget unit | Ad set | Ad group | Campaign | Ad group | | 3. Creative | Ad | Ad | Ad | Ad |
- Campaign level: the objective (what you want the platform to optimise for) and sometimes the budget.
- Ad set / ad group level: audience, placements, schedule, optimisation event, bid strategy and often budget.
- Ad level: the creative, copy, call-to-action button, destination URL and tracking.
Understanding this matters because each decision lives at a specific level. If you want to compare two audiences, they need separate ad sets. If you want to compare two videos for the same audience, they go into the same ad set (or a formal A/B test).
Objectives: tell the algorithm what success looks like
The objective is the most important setting you choose. Delivery systems find people most likely to take the action you optimise for. Choose "Traffic" and you get people who click; choose "Sales" and the system looks for people likely to buy. Choosing the wrong objective is one of the most expensive beginner mistakes.
Meta groups objectives into six options: Awareness, Traffic, Engagement, Leads, App promotion and Sales. TikTok offers similar goals, such as Reach, Traffic, Video views, Community interaction, App promotion, Lead generation and Sales. LinkedIn offers objectives including Brand awareness, Engagement, Video views, Website visits, Lead generation and Website conversions. X offers goals such as Reach, Video views, Engagements, Website traffic, App installs and Followers. Names and options change periodically, so check each platform's current list.
Matching objective to goal
| Business goal | Suitable objective | Optimisation event | |---|---|---| | Launch awareness for a new brand in Dubai | Awareness / Reach | Reach or ad recall | | Grow a creator's following | Engagement or Followers | Follows / profile visits | | Collect B2B leads in the UK | Leads (LinkedIn Lead Gen Forms or Meta instant forms) | Lead submitted | | Sell products on a Shopify or WooCommerce store | Sales | Purchase | | Drive WhatsApp enquiries in Pakistan | Engagement or Leads with a messaging destination | Conversations started |
Rule of thumb: optimise for the deepest event that happens often enough for the algorithm to learn. If you get very few purchases, you may temporarily optimise for "Add to cart" or "Initiate checkout", then move to purchase as volume grows.
Automated campaign types
Automation is now the default rather than an option. Meta renamed Advantage+ shopping campaigns to Advantage+ sales campaigns, extended the approach to leads and app campaigns, and has moved campaign creation so that Sales, Leads and App promotion campaigns start with Advantage+ settings (audience, placements and campaign budget) switched on – you can still switch individual settings off. TikTok's Smart+ automates targeting, bidding, placements and budget with module-by-module controls; LinkedIn offers Accelerate; Pinterest has Performance+; Snapchat has Smart campaign solutions. They work best with accurate conversion tracking and plenty of genuinely different creative. The next-but-one lesson shows how to steer them; the principle here is that structure still matters: automation needs fewer, better-funded campaigns with clear objectives.
A simple, scalable account structure
For a small or medium advertiser:
- Prospecting campaign (new customers): one or two ad sets with broad or lightly guided targeting, several creatives each.
- Retargeting campaign (warm audiences): website visitors, video viewers, engaged users – excluding recent purchasers.
- Testing campaign (optional): a controlled space for new creative concepts before moving winners into prospecting.
Avoid over-segmentation. Ten ad sets each spending a tiny budget will struggle to exit the learning phase (covered later). Fewer, better-funded ad sets usually outperform many small ones.
Naming conventions
Use consistent names so reports are readable by anyone on the team:
2026-10_Sales_Prospecting_KSA_Broad_Video for an ad set, and Hook-Problem_CreatorAmina_v2 for an ad.
Common mistakes
- Choosing Traffic or Engagement when the real goal is sales.
- Putting creatives you want to compare into different audiences.
- Too many small ad sets.
- Unclear naming that makes analysis painful.
Hands-on: a starter account blueprint
A copy-and-adapt structure for a small e-commerce brand spending roughly USD 3,000–10,000 a month (adjust currency and scale):
META
Campaign 1: 2026-10_sales_prospecting_advplus Objective: Sales (Purchase, value)
Ad set: broad_uae-ksa_18-54_advplus-audience Budget: campaign-level
Ads: 6-10 genuinely different concepts (UGC demo, founder story, testimonial, offer static, carousel)
Campaign 2: 2026-10_sales_retargeting_manual Objective: Sales (Purchase)
Ad set: visitors-30d_atc-14d_excl-buyers-30d Budget: 10-20% of Meta spend
Ads: 3-4 objection-handling ads (reviews, delivery/returns, COD, bundle)
Campaign 3: 2026-10_creative-testing Objective: Sales
Ad sets: one per concept test, ad-set budgets, run 7 days
TIKTOK
Campaign: 2026-10_sales_smartplus Objective: Sales (Complete payment)
8-15 vertical videos incl. Spark Ads from creators
Naming rule: yyyy-mm_objective_role_detail for campaigns, audience_geo_age_setting for ad sets and concept_hook_creator_version for ads, all lowercase. Build names in a sheet with =LOWER(TEXTJOIN("_",TRUE,A2:D2)) so everyone uses the same pattern – and because dynamic UTM parameters copy these names into your analytics.
Worked example 2: a UK B2B course provider
A London provider of compliance training sells a £450 online course to HR managers and runs a free webinar to generate leads.
- LinkedIn: Lead generation objective with a Lead Gen Form (webinar registration), audience of HR job functions at companies with 50–1,000 employees in the UK. Optimisation event: lead.
- Meta: Leads objective with an instant form containing a qualifying question ("Company size?"), Advantage+ audience on, creative that calls out "HR managers" explicitly.
- Retargeting: webinar registrants who did not buy, on both platforms, with a Sales objective to the course page.
Result pattern to watch: Meta often delivers cheaper leads, LinkedIn more qualified ones. The right comparison is cost per sale from the CRM, not cost per lead.
How to measure success
- Every campaign has one objective that matches the business goal, and the optimisation event fires at least a few times a day per ad set.
- Budgets concentrate in a few ad sets rather than many starved ones.
- Anyone can read a report and tell what each campaign, ad set and ad is from its name alone.
Video lecture: Campaign structure and objectives: telling the machine what success looks like
Lecture coming soon · 11 chapters · about 8 minutes. Read the full transcript below.
- Structure and objectives
- Three levels
- The objective = the destination
- Optimisation event rule
- Example 1: Riyadh skincare (illustrative)
- Automation by default
- Starter blueprint
- Naming convention
- Example 2: London B2B course (illustrative)
- Judge on real outcomes
- Recap and try this now
Lecture transcript
Structure and objectives
Here's an expensive mistake I see all the time. A brand wants online sales. They open Ads Manager, pick the Traffic objective because clicks sound good, and a week later they've got thousands of cheap clicks and almost no purchases. The ads did exactly what they were told. They found people who love clicking. In this lecture you'll learn the three-level structure every paid social platform shares, how to choose the objective and optimisation event that match your real goal, what the shift to automation like Advantage plus and Smart plus means for structure, and a simple account blueprint you can copy today.
Three levels
First, the structure. Almost every platform uses three levels, even if the names differ. The campaign level holds the objective, what you want the platform to optimise for, and sometimes the budget. The ad set level, called an ad group on TikTok and X, and a campaign on LinkedIn, holds audience, placements, schedule, optimisation event, bid strategy and often budget. And the ad level holds the creative, copy, button, destination and tracking. Why does this matter? Because every decision lives at a specific level. To compare two audiences, you need two ad sets. To compare two videos for the same audience, they go in the same ad set or a proper A/B test.
The objective = the destination
Now the objective, the most important setting you'll choose. Think of the delivery algorithm as a very literal taxi driver. Say take me somewhere with lots of people, and you'll end up at a busy roundabout. Say take me to the airport, and you'll get to your flight. The objective is the destination. Choose Traffic and the system finds people who click. Choose Sales and it looks for people likely to buy. Meta groups its objectives into six: Awareness, Traffic, Engagement, Leads, App promotion and Sales. TikTok, LinkedIn, Snapchat, Pinterest and X offer similar sets, and names change, so always check the current list.
Optimisation event rule
Inside the objective, you also choose an optimisation event, the specific action to find more of. Here's the rule of thumb. Optimise for the deepest event that happens often enough for the algorithm to learn. For an online store, that's usually purchase. But if you only get a handful of purchases a week, the system struggles, so you might temporarily optimise for add to cart or initiate checkout, then move to purchase as volume grows. For lead generation, optimise for the lead, and later for a qualified lead that you send back from your CRM. And for WhatsApp sales in Pakistan or the Gulf, optimise for conversations started.
Example 1: Riyadh skincare (illustrative)
Let's do a simple worked example. A skincare store in Riyadh runs a Traffic campaign. Cost per click looks great: about half a riyal. But purchases are tiny, and the cost per purchase is terrible. They switch to a Sales objective, optimising for purchase with order value. Same budget, same creative. Cost per click roughly doubles, and the team panics for a moment. But purchases more than double, and cost per purchase falls. Here's the key idea. A higher cost per click can be good news, because the system is now buying the right people instead of the cheapest clicks. Judge campaigns on the goal, not the cheapest metric.
Automation by default
Now, automation. It's no longer an option you try. It's the default. Meta renamed Advantage plus shopping campaigns to Advantage plus sales campaigns, extended the idea to leads and app campaigns, and now starts Sales, Leads and App campaigns with Advantage plus audience, placements and campaign budget switched on. You can still switch individual settings off. TikTok's Smart plus automates targeting, bidding, placements and budget, with controls for each module. LinkedIn has Accelerate, Pinterest has Performance plus, and Snapchat has Smart campaign solutions. So does structure still matter? More than ever. Automation needs fewer, better-funded campaigns, clean conversion data and lots of genuinely different creative.
Starter blueprint
Here's a starter blueprint for a small e-commerce brand. On Meta: one prospecting campaign using Advantage plus, with a Sales objective optimising for purchase value, and six to ten genuinely different concepts. One retargeting campaign for recent visitors and cart abandoners, excluding recent buyers, with ten to twenty percent of the budget and ads that handle objections. And an optional testing campaign for new concepts. On TikTok: one Smart plus sales campaign with eight to fifteen vertical videos, including creator Spark Ads. That's it. Avoid ten ad sets each spending pocket change. They'll never exit the learning phase, which we cover later.
Naming convention
And name everything consistently. A pattern like year and month, objective, role and detail for campaigns; audience, geography and age for ad sets; and concept, hook, creator and version for ads. All lowercase, separated by underscores. You can build the names in a spreadsheet with one formula, which is in the lesson text. Why bother? Because anyone should be able to read a report and know exactly what each line is. And because dynamic UTM parameters copy these names straight into your analytics. Messy names in Ads Manager become messy data in GA4.
Example 2: London B2B course (illustrative)
Now a realistic B2B scenario. A London compliance-training company sells a four hundred and fifty pound online course to HR managers, using a free webinar to generate leads. On LinkedIn: a Lead generation objective with a Lead Gen Form, targeting HR job functions at companies with fifty to a thousand employees. On Meta: a Leads objective with an instant form that includes a qualifying question about company size, Advantage plus audience on, and creative that calls out HR managers directly. Then retargeting for registrants who didn't buy, using a Sales objective. The pattern they see: Meta leads are cheaper, LinkedIn leads more qualified. So what's the right comparison?
Judge on real outcomes
The right comparison is cost per sale from the CRM, not cost per lead. In this case, LinkedIn's pricier leads turned into customers at a much higher rate, so its cost per sale was competitive, while some cheap Meta leads never answered the phone. That's why objectives, events and structure all connect back to real business outcomes. Common mistakes to avoid: choosing Traffic or Engagement when the real goal is sales; putting creatives you want to compare into different audiences; too many small ad sets; and names nobody can read. How do you measure success? Each campaign has one objective that matches the business goal, and its optimisation event fires several times a day.
Recap and try this now
Let's recap. Every platform shares a campaign, ad set and ad structure, and each decision belongs at a specific level. The objective is the destination, so match it to your real goal and optimise for the deepest event that happens often enough to learn from. Automation is now the default, which makes a simple, well-funded structure and clean names more important, not less. Here's your try this now. Sketch an account for a real or imagined brand using the blueprint in the lesson text: name every campaign, ad set and ad with the convention, and write the objective and optimisation event next to each.
Key takeaways
- Campaign → ad set/ad group → ad is the shared structure; each decision belongs at a specific level.
- The objective tells the algorithm whom to find, so match it to the real business goal.
- Optimise for the deepest event that occurs often enough to learn from.
- Keep structure simple: prospecting, retargeting and optional testing, with clear naming.
Try it
Sketch a campaign structure for a real or imagined brand: name the campaigns, ad sets and ads using a consistent convention and state the objective and optimisation event for each.