16 Strategies to Split Test Your Way to Facebook Ad Success

If you’re running (or thinking about running) a Facebook ad campaign, this is a post you cannot miss! Einstein’s got 16 Ad optimising tests that are sure to boost your stats…

It’s the end of another working day…

…your usual route home is jammed with traffic. There’s an alternative, but you don’t normally take it because it’s always busy…

…which way would be the fastest route home?

No matter what the sat nav says, you’ll never know the answer for sure.

Unfortunately, when we’re faced with a split decision, it’s almost impossible to know if we’ve made the correct choice…

…but if we did, we’d be able to make future decisions based on those answers.

The same can’t be said of the digital world.

We can split test until we turn blue, track results and OPTIMISE our campaigns.

As an agency Einstein spends millions on Facebook ads every year. Unless you’re spending more, pick up a notepad, and take these 16 split test ideas on board.

Let’s do this:

#1 Headlines

There’s no better place to start than the title.

This is the first thing your audience will read when they see your post. And your first (and sometimes only) opportunity to TANTALISE them into clicking.

You should try running ads with at least two different headlines, testing one (or all) of these features-

-Length

-Power/emotive words

-Use of numbers

-Positive vs Negative headlines

-Scarcity

-Question v Statement

-Swapping headline and description around.

EINSTEIN MOMENT –When split testing headlines, don’t create headlines with only slight differences. Make them completely different, find the best and test again around the winner.

#2 Description

The description is the text BENEATH your headline (usually in grey). It’s called into action when the headline has interested the reader, but not totally convinced them to click.

Einstein suggests making descriptions one, or maximum two sentences.

Use these as SECONDARY HEADLINES. In other words, the headline you really liked but didn’t quite make the cut.

#3 Ad Images

It seems PAINFULLY obvious, but you’d be surprised by how many businesses don’t run image tests.

The image in your ad is the attention grabber…

…and it can give you a LOADS MORE CLICKS.

We suggest running ads with 3 or 4 different ad images.

EINSTEIN MOMENT -If you’re no good at Design, and are sick of paying over the odds for Freelancers, use a service like DesignPickle. All of Einstein’s images come from them! 

#4 Predominant Colours

Another biggie…

…colours trigger different reactions in an audiences’ psyche.

If you’re struggling for ideas with different images, change the background colour.

Decide on the objective of the ad, research colour triggers (on Google, there’s tons of stuff), and test the most promising on your ad.

EINSTEIN MOMENT –When testing images, always make sure your images are placed in the same AD format. This will give you an accurate reading on which performs better.

#5 Photo vs Illustration 

Another nod towards the effects of your ad creative…

…photos can activate memories and make your product seem REAL.

But an illustration can make your brand feel friendly, fun and exciting.

Tests should be run on photo vs illustration, but only if you’ve got HIGH-QUALITY versions of each.

#6 Text in Images vs No Text in Images

If you’ve got some clear space in your ad creative…

…split test that same image with text.

Einstein warns you to be careful with text size though. Facebook doesn’t like images with text that takes up more than 20% of creative space.

We’d suggest testing text like:

-the headline

-description

-question (with YOU included)

-Hyper-shortened angle on content

#7 CTA Button

If you’re new to Facebook Ads, (we’ve all been there, don’t sweat it), CTA stands for Call-To-Action.

In this case, our CTA button is the clickable blue button at the bottom right of our ad.

Facebook’s Business Manager offers a range of different options, including:

-Sign up

-Contact us

-Learn more

-Download

Find the top two buttons that match up with your campaign goals and test them.

#8 Social Proof

Social proof should be used at every opportunity.

But, what type of social proof should you use? And how should you present it?

Run split tests on numbers vs percentages. E.g. 180 out of 200 people… or 90% of people…

Don’t restrict your social proof testing to just this though. Try different types!

Social media numbers vs (happy) customers

Reviews vs (quantitative results of) product effect

Learn more about social proof with our guide (we’ll link it at the bottom of this article too!)

#9 Angle 

What’s the chosen strength of your product? And how are you presenting it?

This is your ANGLE and can be a real make or break for your ad.

The angle should be pushed in every part of your ad, headline, ad copy etc.

To find your most effective angle, use your customer avatar to discover your target market’s biggest pain points and work around this…

…Einstein has some AWESOME guides with different psychological cues that guarantee to HOOK audiences.

Check them out HERE, or wait until the end of this article.

#10 Copy Length

Long v short? Descriptive v teasing? Simple v technical?

Copy length is MASSIVE. It can flip audience’s that are on the fence, right into your lap.

Einstein suggests running split tests between one sentence copy, and one (quite long) paragraph copy…

…we’ve found copy length is very much product and industry dependent. If you haven’t found the sweet spot already, TEST!

#11 Copy Style

Informative? Friendly? Chatty? Formal?

Your copy style should represent your brand, fit with your target market and TEMPT your audience into clicking.

Just with all our Facebook ad split tests, start with very different copy styles and narrow until you’ve found your perfect mark.

#12 Time of Day

What time are you running ads? Are your budgets running continuously?

If you’re on a tight budget, it might be worth pushing all your budget into certain times of day.

But what times?

That depends on your industry, product and target audience’s working hours and leisure time…

…so, SPLIT TEST!

Check out Einstein’s guide to social media posting times for more help.

#13 Days of the Week

If your budget is thinly spread over 7 days, why not pick certain days of the week instead?

Your target audience’s usage will vary all week, depending on a ton of different factors…

…try measuring weekends against weekdays or Monday’s against Fridays…

…one day can make a big difference.

Check out Einstein’s guide to social media posting times for more help.

#14 Video

Video marketing is just too big to ignore…

…and it’s still growing.

Facebook loves video’s because it entertains their users ON their domain. This makes it super cheap to get views.

Sounds amazing right?

Almost.

Getting videos right isn’t easy, especially if you’re selling a product.

If you’re new to videos, Einstein suggests split testing your videos against image ads.

See what comes out on top, and go again.

#15 Emoji vs No Emoji

To winky face, or to not winky face, that is the question…

…a modern-day Hamlet would surely ask.

And you should follow suit.

Emoji’s have become so popular, they’ve established themselves as a crucial part of our language.

And they’re UNIVERSAL.

Emoji’s definitely catch the eye, and can trigger certain emotional reactions from your audience…

…but they can also look cheap and very informal.

Test a few within your AD COPY, and see how it goes.

#16 Ad Type

Facebook has more than 10 different ad types.

That’s a lot to choose from.

If you’re running one type, just because it works and you’ve never tried anything else, SPLIT TEST AD TYPES!

You might find an extra 5% in conversions that you never thought possible.

This is especially relevant for people who run tons of ads and have got lazy, using the same one time and time again.

Facebook Ad Split Test Ideas 

Whether you’re just starting out on your Facebook Ads journey…

…or you’ve been doing it for ages…

…it’s CRUCIAL to run tests.

We don’t suggest running all 16 split tests at the same time (unless you’ve got far too much time on your hands) but running one to three tests can give you invaluable results.

When you’ve got these answers, test the next part of your ad until…

…it’s perfect.

Every industry, product, and brand has its own advertising sweet spot.

There are no exact guidelines that will work for everyone.

Go find yours.

Did you enjoy this blog? What do you think about our Facebook ads split test ideas? Leave a comment or send us a message and we’ll reply as soon as our social media manager has stopped fantasising about the ‘sandwich lady.’

 

RELATED POSTS

The Hook: Attract More Readers 

Social Media Posting Times 

Facebook Audiences

Facebook Ads: The 3 Questions You Need to Answer

Social Proof