Given our strict CPA target and the high costs of running LinkedIn Ads, the bid strategy remained the same throughout all of our testing, which was to bid at or close to the minimum bid floor (the lowest LinkedIn allows you to bid for your target audience ) on cost per click bidding.
Knowing that LinkedIn bidding takes into account not just how much you're paying but your quality score; how relevant your ads are and your click through rate (CTR), frist database we knew we could win the bidding through clever targeting and really relevant creatives without blowing our budget early on. But first, we needed to find out what audience was going to work for Telephone Systems on this brand-new channel.
Phase 1 - Targeting Based on Research
The first audience we tested was one targeting key decision makers (using LinkedIn's job title targeting option eg owners, presidents, directors etc.) of companies in select industries that we had identified from research that would be relevant for the product.
The results were not promising. Although we had high levels of engagement with our ads, we had poor conversion rates resulting in over double our target CPA (cost per sign up). Also, when digging into our impression and click data, job title targeting wasn't as effective in targeting decision makers as we thought; 67% of our impression share was being served at entry and senior job levels, which generally provide less value for our clients. This meant we weren't able to sell them at the higher prices we needed in order to compensate for higher click costs. This led to the campaign operating at a -50% profit margin.
As a result, we had to re-strategise and find ways on improving both our targeting and creatives (ads).
Phase 2 - Honing our Audience and Creative
In response to Phase 1, we switched from job title targeting to job seniority targeting, including only Director level and above, and excluding job levels below this, addressing the issue we found with Job title targeting in Phase 1. We also removed company industries from our targeting to broaden our audience. Hypothesis here was that by going broad, we would be able to identify – based on actual tracked conversions, rather than research – which company industries and other targeting criteria had the highest conversion rates. We could do this using LinkedIn's demographic breakdowns.
Lastly, we also included relevant company size targeting as a lever to improve our chances of generating business leads that needed 5 or more phone systems – our clients' valuable customers and the biggest revenue tier for us.