Why paid social suits a launch, and its limit
A launch needs many of the right people to hear about something new within a few weeks. Earned coverage and creator posts can do that, but neither can be scheduled to the day or aimed at a defined audience. Paid social can: it buys reach among the segment the launch was planned for, at a frequency the plan chooses, starting on the morning the embargo lifts. It also does something no other bought channel does as cheaply before launch: it can build an audience of people who have already shown interest, by watching a teaser or joining a waitlist, so that launch day starts with a warm group rather than a cold one.
The limit is that the delivery systems behind these ads learn from history, and a launch has none. There is no purchase data for the new product, no customer list of its buyers and no social proof for the ads to carry. An account that treats the launch like a sale of an existing range, as the guide to planning Meta ads around a sale or peak does for proven products, will ask the system to find purchasers it has never seen. The tactics below are the launch version: they create the history the channel needs, in the weeks it has to create it. How paid social is phased against the other channels across teaser, burst and sustain belongs to the media planning guide for launches; this page covers what happens inside the channel.
Paid social tactics for a product launch
| Tactic | What it does for the launch | Playbook |
|---|---|---|
| Build the launch audience before the reveal | Runs teaser video and waitlist ads before launch day so engaged viewers and sign-ups exist as an audience on day one | Custom audiences |
| Optimise for an event the new product can reach | Picks a frequent step, such as a waitlist sign-up or add to cart, while the product has no purchase history, and judges on first orders | Conversion events across platforms |
| Finish learning before launch day | Starts the launch ad sets early enough to settle, then opens the window by adding launch creative rather than rebuilding | The learning phase |
| Lead with what is new, then switch to proof | Opens on the change the product makes for the buyer, then replaces claims with early buyers' and creators' evidence | Finding creative angles, Spark Ads |
| Staff the comments in launch week | Answers questions under launch ads quickly and in public, because those threads become the product's first FAQ | Community management |
| Hand over to always-on after the window | Moves winning creative and audiences into the evergreen sales structure and changes the optimisation event only once the window closes | Meta Ads for online sales |
Build the launch audience before anyone can buy
The weeks before launch are the only time the channel can work on the audience without the product being available, and they decide how warm launch day is. Two kinds of pre-launch ad do the work. Teaser video shows the problem the product solves, or a glimpse of it, to the target segment at low cost per view; everyone who watches a meaningful share becomes an engagement audience. Waitlist ads ask for an email address or a notification opt-in in exchange for first access, using an on-platform form or a short landing page; everyone who signs up becomes both a customer-list audience and a launch-day email. The custom audiences guide explains how video viewers, form openers and site visitors are collected and how long each pool lasts, which matters here: set the retention window so the earliest teaser viewers are still in the audience on launch day. A launch that starts paid social on the day itself skips this step and pays prospecting prices for every first order.
Choose an optimisation event the product can reach
Every network optimises towards the event the ad set is told to find, and it learns from how often that event happens (Optimisation EventThe conversion event an ad set or campaign is told to optimise delivery towards, such as a purchase, lead or install.). A new product's purchase event has no history at all, and in the first days of the window it may fire too rarely for delivery to learn from it. The conversion events guide sets out the trade between closeness to value and frequency; for a launch the usual answer is a step that happens often in the window and reliably precedes an order: a waitlist sign-up before launch, then a pre-order or add to cart in the first weeks. Two rules keep this honest. Judge every ad set on first orders from the business's own records, never on the proxy it optimises for. And write down when the event will change, because Meta lists changing the optimisation event among the edits that restart learning: switching to purchase optimisation in the middle of launch week resets delivery at the worst moment, so schedule it for after the window.
Let the ad sets settle before the window opens
The Learning PhaseThe period after launch or a significant edit when an ad set's delivery is still being optimised. is the period after an ad set is created or significantly edited in which delivery is unstable and results cost more, as the learning phase guide explains from Meta's documentation. A launch that builds its main ad sets on launch morning spends its most valuable days in that state. Build the window's structure during the teaser period instead, consolidated into few ad sets so each collects events quickly, run it on teaser creative until it has settled, and open the window by adding the launch ads to those ad sets in one edit. Adding a new ad is itself listed among the significant edits, but it is usually the mildest of them, and it is far cheaper than creating fresh ad sets on the day attention peaks.
Open on what is new, then let others make the case
Launch creative has a job that ordinary creative does not: it has to say what has changed before it says why to buy. The first seconds of each ad should show the new thing doing what the old way could not, in the words the launch's positioning statement settled, because people scrolling past a familiar brand assume they already know what it sells. At launch there is no customer evidence to borrow, so the angles come from the research behind the launch rather than from reviews; the guide to finding creative angles covers how to turn that research into concepts. Plan the second wave before the first runs: as soon as early buyers post, reviews arrive and seeded creators publish, replace brand claims with their evidence. Formal partnership formats such as Meta's partnership ads and TikTok Spark Ads run creator posts as ads, so agree the rights before launch; the influencer launch page covers the seeding that produces those posts.
Answer the comments as if they were the FAQ
Launch ads attract questions in a way evergreen ads do not: what does it cost, does it work with what I have, when does it ship, how is it different from the old one. Those threads are public, sit under every impression the ad receives and are read by people who never ask anything. An unanswered question about compatibility or delivery dates under a heavily funded launch ad does more damage than a weak headline. Put a named rota on launch-week comments with the agreed answers to the expected questions, escalation rules for complaints and the same tone as the brand's organic accounts, as the community management guide sets out. Feed the recurring questions back into the creative and the launch page within days.
Hand the winners to the always-on plan
A launch campaign is temporary by design. When the window closes, move the creative that produced first orders at an acceptable cost, and the audiences the launch built, into the evergreen structure the business uses for sales, set out per network in Meta Ads for online sales. This is also the moment to change the optimisation event to purchase, since the product now has history and the window no longer depends on stable delivery. Retire the teaser and launch-claim ads rather than leaving them running beside proof-led creative, and keep recent buyers excluded from prospecting so the always-on budget is not spent on the launch's own customers.
Budget posture and KPI targets
For a consumer launch, paid social usually carries the largest paid share of the burst, because it is the one bought channel that can be aimed at the launch segment from the first hour. Commit it as a window budget rather than a daily habit: a fixed amount for the window, a ceiling on what a first order may cost, and a smaller pre-launch amount for teasers and waitlist ads.
Set the ceiling from the launch target rather than from the always-on account. Work out what a first order is worth over the period the business plans for, using margin and expected repeat purchase, and treat that as the most the window may pay on average; a launch can sensibly pay more than an established product's target, but not without limit. Keep the pre-launch spend as the test allowance: it is where teaser concepts are compared on cost per engaged view and cost per sign-up before the main money is committed. Money moves out at two points: when the window closes, into the always-on sales plan, or earlier when an ad set's cost per first order stays above the ceiling for a week after learning has settled. The reach and frequency calculator helps size the window's reach goal, and the CPM calculator compares what each network charges for the segment.
The tactics in a worked plan
Pellowmere Audio, an invented headphone maker
Pellowmere sells headphones online and is launching open-ear sports earbuds. The launch window is five weeks. Paid social has £60,000: £6,000 for four pre-launch weeks and £54,000 for the window, with a ceiling of £38 per first order worked out from margin and expected repeat purchase.
| Tactic | Plan |
|---|---|
| Launch audience | Teaser videos of runners hearing traffic through the earbuds, plus waitlist ads offering first access; 9,400 sign-ups at about £0.64 each |
| Optimisation event | Waitlist sign-up before launch, add to cart in the window, purchase only after week five |
| Settled ad sets | Three ad sets built in pre-launch week two; launch ads added to them in one edit on launch morning |
| Creative | Launch-day ads open on the open-ear design; runner and creator clips replace them from week two |
| Comments | Two people on a daily rota with answers on price, fit and phone compatibility |
| Hand-over | Proof-led ads and the waitlist audience moved to the always-on campaign after week five |
The window spends £48,000 and records 1,450 first orders at about £33 each, 610 of them from people on the waitlist. A cold prospecting ad set runs above £38 for a week after it settles, so £6,000 moves to the always-on plan in week four, ahead of the rest.
The plan shows the posture at work: a small pre-launch budget buying the audience the window needed, a ceiling set before launch and used during it, and the optimisation event changed only once the window was over.
KPI set for paid social launches
| KPI | What it tells you | Read how often |
|---|---|---|
| Reach and frequency in the target audience inside the window | Whether enough of the launch segment met the product while it was news, at a useful frequency | Daily in launch week, then weekly |
| Pre-launch sign-ups and their cost | Whether the teaser period built a launch audience at a price worth paying | Weekly before launch |
| Cost per first order, trial or pre-order in the window | Whether reach turned into the first step at a cost under the ceiling, from the business's own records | Daily in launch week, then weekly |
| Launch-day audience size | How many engaged viewers and sign-ups launch day could reach again | Once, the day before launch |
| Product-name search against the pre-launch baseline | Whether the ads made people curious enough to look the product up | Weekly in the window |
| Share of comment questions answered the same day | Whether the threads under launch ads are answering buyers or putting them off | Daily in launch week |
Read ReachUnique people who saw an ad. and FrequencyAverage times a person saw an ad. together and against the segment the launch plan chose, not against everyone the ads reached. Platform-reported conversions in a launch window count people the ads touched, some of whom would have bought after seeing coverage or creator posts anyway; where the budget allows, a regional or audience holdout gives a cleaner read of IncrementalityConversions caused by marketing.. The product launch playbook shows how these readings feed its go and no-go gates, and paid social for brand awareness covers the reach and frequency reading when the goal is memory rather than a launch.
When paid social is the wrong route for a launch
Paid social is a poor lead channel when the launch audience cannot be reached as a segment on the networks, such as a small group of specialist buyers better reached through trade press, events or account-based work. It is also the wrong place to start when the product cannot be understood in a feed: something that needs a demonstration or a conversation will collect views and little else. And it is a weak choice for a minor release whose buyers are existing customers, where email, the product itself and a release note reach everyone who cares at almost no cost. In each case the launch hub lists the channels that carry the window better, and paid social becomes a supporting line rather than the largest one.
