Paid social on a small budget
Warm audiences first, one offer, and a stop condition written before the first currency unit. The boost button is not advertising.
Paying to be seen on a social platform is a different activity from posting on one, with different skills and a different failure mode, and on a small budget most of what is written about it does not apply.
What a small budget can and cannot buy
It cannot buy a discovery campaign at scale, it cannot support testing many variations, and it cannot absorb a learning period on several audiences at once. Every system here needs volume to optimise, and a small budget does not produce volume.
What it can buy is reaching people who already know you, one offer, consistently, for long enough to learn something. That is a narrow use and it is where small budgets actually work.
Spend it on the warm audience first
- People who visited your site, if you have the tracking in place and the consent to use it.
- People who engaged with your posts or watched your video.
- Your own customer or email list, uploaded where the platform and the law allow it, which requires a lawful basis and a privacy notice that covers the disclosure.
- Only then, people who look like those groups, which is the first step that is genuinely cold.
The first three are small and cheap and convert at a rate the cold audience will not approach. Starting cold with a small budget is how most of this money is wasted.
Run it properly or not at all
- One offer, one audience, one creative concept. A small budget split four ways learns nothing about any of them.
- Leave it alone for long enough to produce data. Changing things every two days resets whatever learning was happening.
- Measure the outcome rather than the click. A platform will happily report cheap clicks on traffic that never does anything.
- Decide the stop condition in advance: spend this much, and if no outcomes, stop. Without it, this becomes a subscription.
- Track what it cost per actual customer, not per lead, which is where small-budget campaigns usually turn out to be unprofitable.
The obligations that come with it
Advertising carries requirements that organic posting does not always make obvious. Claims must be substantiated, prices must be accurate and include what is mandatory in the market you are advertising to, and comparison with competitors is regulated in most jurisdictions.
Tracking visitors for advertising purposes requires consent in a growing number of places, including under European rules on storing information on a device, and the consent has to be obtained before the tracking starts rather than alongside it. If you are running retargeting, that is the part most likely to be non-compliant.
The creative is most of it
More of the outcome depends on what the advertisement actually says and shows than on the targeting, which is where small advertisers spend their attention.
- Make it look like the platform rather than like an advertisement. Something filmed on a phone outperforms something designed, consistently.
- Say the offer in the first second and in the image, because most people will not read the text.
- Show the actual thing. The product, the work, the place, the person.
- One idea per advertisement. Two offers in one is two advertisements you cannot evaluate.
- Change it before you need to. The same thing shown repeatedly to a small audience stops working, and a small budget reaches the same people quickly.
What two weeks of data tells you
On a small budget, less than you want. A handful of outcomes is not a result, and the temptation is to declare a winner between two variations on the strength of four conversions against two.
Be honest about this. At low volume you can tell the difference between working and not working, and you cannot tell the difference between two things that both roughly work. So test big differences rather than small ones, and treat anything that needs statistics to see as beyond what this budget can learn.
When not to advertise at all
If the page the advertisement sends people to does not convert the traffic you already have, paid traffic will not fix it, it will only cost more to discover the same thing. The order is: a page that works, then organic traffic that converts, then paying to send more people to a known quantity. Advertising into an untested page is the most reliable way to conclude that advertising does not work for your business.
Work to do
- Put your budget on warm audiences before anything cold.
- Write the stop condition before the first currency unit is spent.
- Check that your tracking has a lawful basis and that consent, where required, is collected before it loads.