Numbers · Published 2025-04-02 · 3 min read
The hard launch and the normal launch
Here is your excuse to tear me apart in the comments. Generalisations are always irritating.
I am often asked about typical campaign launches, typical numbers and common strategies. Yesterday, during a consultation, a client and I worked out two categories of launch for ourselves:
- The normal launch
- The hard launch
The normal launch
You start on a small budget. From the first days you see purchases and trials. Day-one ROAS moves between 40% and 70%.
Strategy for Facebook Web:
- Launch the campaign on worldwide and US. In worldwide I exclude every country whose name ends in "stan".
- Let it learn.
- Keep the best creatives.
- Raise the budget for as long as the trial or purchase price allows.
- Experiment with bid cap. Scale through new creatives.
Many of my clients run profitably on the US alone, without even trying other countries. That is a sign of a normal launch.
The hard launch
Usually in competitive niches. Day-one ROAS on an untrained campaign is below 20%. There are neither purchases nor trials from the first days.
Strategy for Facebook Web:
- We launch the campaign on worldwide and US. Zeroes everywhere.
- We start separating out geos. For those geos we make localised ads.
- One campaign per geo. Each campaign spends little because of the low bid, but there are many campaigns and they drip in a little at a time.
- If organic traffic appears within a geo, we start counting ROAS per geo rather than per campaign.
Some thoughts
- When you form a strategy you have to understand which of the two you are looking at. For example, if yours is a normal launch, do not think about localising ads, especially early on. Or, more generally, do not think about fine tuning: you already have plenty of freedom within start, stop, budget, bid, duplicate, creative.
- And the other way round: if yours is a hard launch, you have to treat setting up the campaign as a startup project in its own right. For instance, you spotted a competitor's Spanish-language creatives. Or you used to work at a company and know that LTV for apps like this is unusually high (that is theft, essentially; I am not recommending it, but it happens).
- A normal launch does not mean you will make money on it straight away. The road from 70% ROAS to 120% ROAS can be very long. There are examples of people walking it for years.
- For large agencies and publishers the hard launch is not an option: they have no resources to dig into small details. You will see that for yourself when you talk to publishers. But for an indie developer who spent a year building an app at night, this route works. If there is no other app to switch to, they can bury themselves in the analytics and dig out some Slovenia where the app catches on and brings them a small income.
But as my client rightly noted yesterday, that is not a scalable story.
Write to me if anyone needs help with a launch. I will help.