As marketers, you should be grabbing any and all opportunities to optimize your Linkedin ads campaigns to ensure you’re engaging the best audiences with the best content.
LinkedIn is a wonderful tool for your advertising belt, and our LinkedIn ads agency can help you build campaigns and fine-tune your targeting to the most precise B2B audience.
But the real secret to campaign success lies within A/B testing. If you’re not familiar with this concept, this article explains the importance of LinkedIn Ads A/B testing and how to do it right.
What is A/B Testing?
A/B testing (sometimes called split testing) is testing online marketing content to see if different variations affect the metrics. These are conducted in order to optimize said content to improve the metric results and overall return-on-investment.
To A/B test a campaign, you will need to create two copies of the ad that are identical – except for one factor (for example, a headline, an audience demographic, or an image), and then run them both simultaneously and monitor the metric results to see which one performs the best. This can then give you an indication of what, if anything, needs to be changed.
Marketers use A/B testing when asking questions like:
- What imagery grabs the audience’s attention?
- What title is the most thought-provoking?
- Which CTA (call-to-action) option is most likely to be hit by the reader?
To sum it up, A/B testing helps you understand better what should/shouldn’t be included within your ad campaign to get the best results for your budget.
Case Study: How We Helped a Security Analytics Software Company Generate 42% More SQLs Using LinkedIn Ads
Understanding Split Testing
Split or A//B testing helps you create the best possible campaign for the best possible engagement and will help you smash your campaign goals. It helps you break down every element of your ad (such as the images, CTAs, headlines, copy, and so on) to better optimize it.
Oftentimes, it’s what’s required to turn your campaign from an ‘okay’ one to a ‘fantastic’ one – so it’s worth taking the time to do. Here’s the lowdown:
Prepare 2 Weeks in Advance
Don’t leave your split testing to the last minute; instead, get cracking with it at least two weeks before you plan to go live with your campaign. You will also need to create several sample sets to test to ensure every element of your ad is the best it can be.
Leaving it any later than two weeks may not be ample time to get enough testing done.
One at a Time
This means each campaign “set” you create should be identical, bar one thing. Don’t try to rush the process by testing more than one variable per split test. The results won’t be insightful because you won’t be able to identify which variable needs amending based on the metric results.
Again, this is why it’s important to give yourself plenty of time to perform your split testing.
Don’t Change Your Bidding When Testing Targeting
Audience testing is as crucial as testing campaign content because you need to know your ad is meeting the most conversion-likely candidates. You can do this by split testing via audience groups.
But you will need to keep all other elements of the campaign the same (except targeting) – including the bid rate. While there may be small variations in the bid rate owing to factors such as region and audience, etc., try to find a happy medium for your bid.
Test Multiple Creative Variations
If your LinkedIn ad campaign is a Sponsored Content one, you will increase your optimization options and more ad exposure if you run multiple creative variations at the same time. And by ‘multiple’, we mean at least four per campaign.
Once you’ve done this, you will then be able to optimize appropriately based on the performance results of the testing and focus your spending on the best performers.
‘Rotate Variations Evenly’
You can set your campaigns to a function called ‘rotate variations evenly’, which can be found in your Campaign Manager on LinkedIn. There are two options of this feature to choose from:
- ‘Optimize click-through rate – better performing variations appear more often’.
- Serve each variation evenly without weighting for performance. AKA, ‘rotate variations evenly’.
Resist the urge to opt for the first, even though it is a good time-saving option. Instead, click the second for more meaningful and insightful data.
Testing Ideas
So now we understand what A/B testing is and why it matters, let’s look at some of the best campaign aspects to test.
Ad Copy
Your ad’s copy is important because it effectively sells the product/service you’re offering, or coerces the reader into doing what you want (i.e.: downloading your eBook, filling out a lead-gen form, etc) – so it needs to be good!
This means testing different copy styles and finding out which ones the audience relates to the best. You’ll know because of whether or not they engage with it. Good copy is well-written (with exemplary spelling and grammar) and is thought-provoking.
There are 4 different copy styles to try:
- Question-led copy. Naturally, this means opening with an insightful and relevant question. This question needs to address whatever problem is vexing your audience that your brand can fix. Round the copy off with a good CTA for them to reach out.
- Benefit-led copy. This essentially means selling the brand by highlighting the benefits of what’s on offer. The benefits should be exactly what your audience is looking for. End with a good CTA.
- Empathy-led copy. This is copy that lets the audience know that you understand them and their needs – and are here to help. It ends with a CTA for them to be able to reach out.
- Statistic-led copy. B2B customers LOVE statistics because they act as numerical testimonies that encourage them to get in touch or do whatever it is that your ad is teasing them to do. Again, round it off with a good CTA.
Audience Selection
This means finding the best way to identify and reach your target audience. LinkedIn is an extremely beneficial platform when it comes to targeting for campaigns – perhaps more so than any other platform. You can filter various leads via groups, job title, industry, skills, job function, and so much more.
Different audience filters mean different audience sizes, so keep that in mind when building your audience. This is where testing comes in handy; you can find the right combination of filters by playing around with different combos while A/B testing.
The 2 Bidding Strategies
As in, CPC (cost-per-click) or CPM (cost-per-impression).
Don’t be fooled into thinking that your bid amount highlights how much your ad will cost. While it gives you some indication, keep in mind that you’ll be bidding against your competitors, and your CPC or CPM amount may not end up being the same as your initial bid amount.
Monitor & Record
You should always monitor your tests in situ and then keep a record of all results. Failure to do so means you will miss out on invaluable data and information that you need to optimize your campaign.
Aside from that, monitoring will ensure the testing process is working properly and everything is progressing as it should. Keeping a record allows you to view results clearly and concisely and refer back if needed.
Case Study: How We Helped Kodo Survey to Generate 167% More Leads in 3 Months Using LinkedIn Ads
Keep Going With Your Testing
Make A/B testing a permanent part of your campaign building process. Don’t assume the results you got for your first campaign will ensure the best engagement for the next. Treat each campaign for what it is – a different and unique project.
The secret to campaign success is to test all elements of every single campaign. This may seem time-consuming but it will make or break your campaign’s success.
Conclusion
We hope this article has explained the importance of A/B testing for each and every campaign.
While you may think you have winning content/audiences/etc, without the data to back that up, you’re moving blindly with your advertising.
If you’d like to learn more about how we help B2B SaaS and Tech companies grow their MRR through LinkedIn advertising, contact us online or send us an email today at info@getuplead.com to speak with someone on our team.
You might also be interested: