7 Tips Before Starting Facebook Ads

Business Manager? Ready.

Ad Account? Ready.

Campaigns & Ads? Ready.

By this point you should have everything ready to go. Before you hit the switch though, and set everything live, take a look through these 7 tips I wish I knew before starting out on Facebook Ads.

Tip 1: It's only going to get better

Rome wasn't built in a day, and your ads are unlikely to become profitable in one either: Facebook Ads takes time.

Strictly speaking that isn't true. Ads don't take time to get better, they take data, and data is money. You're going to have to put some spend through your ads in order for Facebook to learn who to deliver them to.

You can't spend a few dollars and expect to see instant results. Depending on what you're trying to achieve, it can take anywhere between $50-$100 to even begin to understand if an ad is working, and how well it's likely to work. At this point it will have generated enough results that Facebook can start optimising the ad'd delivery more effectively.

If this seems like a high barrier to entry, it's because it is. If anyone could see results after $1 of spend, then everyone would be running Facebook Ads.

Tip 2: Test. Test. Test.

Testing capacity refers to the ability to learn by testing things within your account. If you're ever running 1 variant of anything, for example if you're only running 1 ad, you're wasting testing capacity. You should always, where possible, be running multiple variants of things. This could be running different ad variants, or ad sets with different targeting. If you can change anything, you can test it.

The reason that testing is so important is that you have to assume all your competitors are doing it to. If they're running tests and improving their ads, this is only going to hurt you. The way to counteract this is to test faster and more intelligently than they can. Which brings me onto my next point:

Tip 3: Test intelligently

We've all heard apocryphal stories of how some famous company tested a million different shades of red on their site's purchase button, just to see which resonated with people the most. As interesting as these stories sound, don't try to emulate them. The only brands which can run these tests are those with millions of customers. If you try to run ad tests that subtle, with little volume, you won't get reliable data.

Instead, focus on the tests where you expect there to be a big difference in outcomes. For instance, will I get better results if I only target women versus targeting both genders? or do video ads get me more sales than image ads? These are the sorts of tests you can conclude without spending tens of thousands of dollars on, and will likely have a bigger impact on your bottom line.

Tip 4: Keep it broad

It's very tempting to think we know our target customer so well that we know exactly what they look like. "Of course", we say, "the only people interested in my product are 34 year old married women in Oregon who are interested in pottery but not in motoring". If you try to create ad sets with targeting as detailed as this, you just won't have enough volume for Facebook to be able to optimise from.

Go as broad as it makes sense for you to, whether that's targeting an entire state, or everyone in the country who likes swimming. Facebook's targeting systems are so advanced now, that they'll figure out the rest for you.

If you're unsure how broad is broad enough, then as a rough guide I'm suggest making sure your target audience is between 250,000 - 1,000,000 people. This will depend on your budget though; for larger budgets go higher, and for smaller budgets feel free to go down to as low as 50,000 people (you can see your target audience size in the ad set settings screen).

Tip 5: Keep your eyes on the prize

Why are you using Facebook Ads? Is it to drive sales? In that case, make sure you're actually optimising towards sales. For this you'll need to make sure your campaign is set up as a conversions campaign, and you have the right purchase event selected at ad set level.

Don't try to game the system by running a traffic campaign. Facebook will think you don't actually want purchases, and so it'll de-prioritise showing your ad to the people who actually buy stuff from Facebook ads (the conversion audience).

Because these people are so much more sought after by advertisers, meaning they cost more to reach, you won't end up showing to them at all. The only people who will see your ad are the sort of people who click all day on Facebook ads without buying anything.

There is one exception to this piece of advice though. If your sales are few and far between, it's perfectly fine to set your ad sets to optimise for something that happens earlier on in the sale journey.

An example of this would be optimising towards people adding items to their cart. Because this will always happen more frequently than sales, it can give Facebook more data to learn from, ultimately improving performance.

Tip 6: Keep an eye on your competitors

If you're advertising on Facebook, chances are that others are advertising very similar products. It's always useful to keep tabs on what ads they're running. Doing so can give you inspiration for your own ads, and can help explain changes in performance that you see (if they run a cut-price sale, for example).

You can see the ads a competitor is running by going to their Facebook page, clicking page transparency, and then Go to ad library. You can filter by location at the top, meaning you can see not just what ads a competitor is running, but where they're running them.

Tip 7: Reach out

There are plenty of others out there running Facebook ads, and plenty of others who have the same troubles as you. Reach out to them and canvas opinions. Good places to do this are reddit, Quora, or Slack.

Don't expect to pay for one-off advice, there are plenty of people out there who can help you with quick questions. This will save you a ton of money in the long run.

That's it.

That's everything you need to know to get started; you're ready to set your Facebook ads live!

That's it for basics

If you want to be notified the next time I write something, leave your email here. No spam. No sales pitches. Just good advertising stuff every couple of weeks.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.