LinkedIn vs Facebook: the Ultimate B2B Lead Gen Battle [case study]

We get it all the time. B2B client wants to reach their target audience, search has reached its potential and they are now looking for higher funnel opportunities.

So where can they find those potential leads who are not actively looking for them on search engines?

The answer for B2B clients looking for direct lead gen opportunities is typically LinkedIn.

After all no one can match the amount of data LinkedIn has on everyone’s professional status, experience and activity. No one can match the unparalleled capabilities to target and layer by job title, employer or experience. Not even Facebook, despite trying, can match this and therefore LinkedIn is always top of the list.

At Katté & Co we like to challenge assumptions and run tests to see what is really true. Let the data do the talking. Can another channel beat LinkedIn at this game?

Experience suggests that Facebook could drive more leads but quality would be worse due to targeting inefficiencies, whilst LinkedIn would be (much) more expensive but would usually lead to much better quality leads.

So we decided to run an internal lead gen campaign to test the hypothesis, considering Facebook’s massive improvements with machine learning based targeting optimisation.

In the midst of the coronavirus crisis, we created a PPC Playbook to discuss ways in which businesses could still make things work for them through advertising during this crisis.

Ads for LinkedIn and Facebook were not exactly the same due to size restrictions on LinkedIn but the general message was the same.

The playbook would be distributed as a PDF and accessible through Native Lead Forms, both available through LinkedIn and Facebook. Of course both lead forms are not identical due to the options and layouts available but close enough for the purposes of this test.

Based on our experience of Facebook’s lead quality, we opted for a feature that allowed us to add more complexity to the form, adding additional steps and prompts to confirm each step, which is designed to ensure those going through the process are fully engaged and and in theory better qualified. This feature is not available through LinkedIn but in theory would decrease the volume of leads but improve their intent/quality.

Questions were the same for both platforms:

• How much is your media spend (select option)
• Work Email
• Mobile Number

LinkedIn Lead Form:

• The image of the ad is added automatically to the top of the form with no way to change this.
• Compared to the FB form, the LinkedIn form was faster to complete; in a single screen.
• Personal questions were first and questions about media spend was second (cannot be changed).
• One step submit click.

Facebook Lead Form:

• Image from ad also added on top / as background.
• Form divided in 3 ‘pages’ / steps:
  • First step: select media spend
  • Second step: email and phone number
  • Third step: review and confirm

Considering LinkedIn’s capabilities we had to create an audience based on a mix of job titles within a set of target industries we are interested in working with. But we were mostly excited about using the lookalike function in LinkedIn and being able to replicate this on FB as the only real targeting alternative which would keep us precise.

Total test budget was £1,000.

Considering LinkedIn is more expensive and our preferred channel (based on initial assumptions) we decided to allocate much more budget into LinkedIn and let Facebook run in the background. Total spend £800 LinkedIn, £200 Facebook. This would allow us to get similar volumes.

We ran both channels for 10 days with very surprising results. LinkedIn started slow and picked up later on with a few slow days in between. Facebook on the other hand kicked off straight away generating a higher number of leads and more consistently:

We can see that Facebook performed at a massively lower CPL converting 5X more leads with almost 1/3rd of the spend.

Of course, we expected that Facebook lead quality would probably not be as good despite the higher numbers. However, we were surprised to see, after our qualification process that Facebook’s leads were of higher quality on average and we actually started a few conversations with potential clients thanks to those Facebook interactions. So more, cheaper leads of good quality vs expensive leads which did not really translate into much business benefit.

So what kind of conclusions can we gather?

Facebook may not have the detailed targeting capabilities that LinkedIn has around professional criteria, but can find and hone down to the same people if you have a decent customer/prospect database to build from.

In this case having a database of people who we know are the kind of people we want to talk to means we can ask Facebook to find more of them, bypassing the need for detailed targeting around job titles and all the rest.

Volumes for LinkedIn and Facebook were actually quite similar but 1 key metric made all the difference:

• Both Facebook and LinkedIn reached the same amount of people but it cost around 3 times more to reach someone on LinkedIn that it did on Facebook.
• Despite CTRs being a bit lower on FB (which usually correlates with higher CPCs), the CPCs are still much lower than what we would get in LinkedIn under the most exceptional conditions.

The most interesting part is the performance of the lead forms. Facebook has improved the UX of its forms recently which, although other things may be involved, has translated into a whopping 46% conversion rate from click. This in contrast with a still respectable but dramatically lower 8.11% for the LinkedIn form, which to be honest looks a bit dated in comparison to Facebook’s version.

• CVR is almost 4 times higher on Facebook Lead forms despite adding elements to the complexity of the form vs LinkedIn which just uses a one pager.
• Potentially the step by step approach that FB has introduced could be helping CVR by offering a more streamlined and less intimidating phased approach.
• And as mentioned before, this actually led to some very good quality leads which is the biggest highlight.
• One element that we could not quite control was the order of the questions and the fact that FB shows all possible answers instead of a dropdown that LI uses. This could very well have an impact on the form’s CVR.

We mentioned that we used 2 audiences on LinkedIn. A look-a-like one and another based on job titles and industries which was meant to be more precise. However we actually generated more conversions through the look-a-like audience.

• The look-a-like generated more leads at a much lower CPL than the detailed targeting one using Industries and Job titles.
• This is also an important reminder that sometimes our biases may affect the way we inform targeting and that our database may not really reflect what we think about our clients.
• In this case, the LinkedIn algorithm managed to create a better quality targeting list from our database than us trying to use their own targeting tools.

Of course it’s important to mention that this is not an absolute test. It’s a reflection of a specific piece of content during a very particular time and we are keen to see if we can replicate it across other clients.

Are you willing to put your lead gen machine to the test? Contact us for a free consultation.