My first breakout session at Inbound 2019 started with the benefits and the downsides of LinkedIn Adverts. One of the key benefits is that LinkedIn can do highly accurate targeting, and this makes it a particularly popular channel for B2B. LinkedIn, however, isn’t perfect, and the presenter highlighted the relatively high cost-per-client, the lack of decide targeting, the inability to schedule ads to particular times of day and the fact that although LinkedIn uses a relevance score in a similar way to Google Search Adverts, this score is not visible to the user.

LinkedIn Advertising Formats

Sponsored content is the most versatile format and the speaker recommends it to about 90% of his advertising clients. This is very much in line with our experience, where we find sponsored content to generally be the best format for almost all our clients. One important factor to remember is that the vast majority (the presenter claimed it was probably over 80%) of sponsored content adverts are viewed on mobile. If you don’t have a website that looks great on mobile maybe you should consider one of the desktop formats.

Text ads are an alternative that means you don’t need to first create a post. Although the click-through rate is lower than sponsored content, they can be good for branding campaigns as you can get a lot of impressions. One challenge with text ads is that they are a desktop-only format, and with more and more people interacting with LinkedIn on mobile, this can be an issue.

Sponsored InMail is a high-quality channel but has a very expensive CPC. The presenter recommended this format for offers such as early access or sneak peaks of new products, job opportunities or a VIP event invite. This isn’t a format for your content offers! Despite the very high CPM and the typically high CPC, it is possible to run campaigns that do achieve low CPCs if the offer is very specific and high quality.

Dynamic ads are the format the speaker never recommends. They are desktop only and result in high CPCs. This doesn’t correlate with our experience, where we have had some good results from a few campaigns where the format could be used to really personalise the advert.

The following benchmarks were presented as being typical numbers for LinkedIn formats that are trying to drive clicks:

FormatAverage CTRTypical CPC
Sponsored Content4%$6-9
Text Advert0.025%$3-5
Sponsored InMail50% open, 3% CTR$23-55
Dynamic“low”$12-15

There are some other ways to advertise on LinkedIn. Lead generation ads, which route the user to a form that is automatically populated with their details if they are logged on seem to be a great idea. The presenter, however, cautioned against their use as you only have 150 characters of text to promote the offer, and the automatic form-fill can mean people who aren’t really interested will send their details. This leads to poor quality leads, so it’s a good approach for generating a large number of leads, but not good at targeting the most valuable prospects.

Video adverts are very expensive when compared to Facebook – the other major platform offering this format, costing from $0.06 to $0.14 for every play (and don’t forget a couple of seconds of viewing will count as a play). One important thing to note is that most viewers will play the video muted, so subtitles are essential if the video is going to be effective. The speaker strongly recommended that all video ads are tried on YouTube first, which will be much cheaper, and only rolled out to LinkedIn if they prove effective.

One important change that LinkedIn has promised is coming is targeting by engagement. For video, this is really important as the onus is on the viewer to then take action after watching a video: if you could retarget everyone that engaged with your videos then that would be a much more effective approach. In fact, this might be where the auto-filled forms of the lead gen adverts start to become better value and deliver higher-quality leads: if you could just target them at people who engaged with your video then this might result in better leads. At this stage, however, all we can do is wait for LinkedIn to roll-out engagement targeting; today there is no way to retarget someone who engaged with your content on LinkedIn but didn’t click through to your website. The ability to retarget people once they hit your website is another reason why routing someone to a form on your website will generally generate better results: retargeting through a service like Google of AdRoll is a fraction of the cost of advertising on LinkedIn.

Carousel ads are simple multiple adverts grouped together. Although the presenter had used it when a brand has a story that can’t be told in a single frame, in general he finds no benefit using them over individual adverts.

Targeting on LinkedIn

Targeting is LinkedIn’s superpower, but it is not well understood by many advertisers. It’s important to know that the selections you use will impact the CPC you pay. The more selections you make (and therefore the more targeted the audience) the higher the CPC.

A good example is job title. This is the most competitive targeting criteria, so it has a significant impact on cost, although it’s the criterion that most advertisers start with. Job function targeting typically reduces the CPC by around $1 compared with job title targeting, although it is broader. Skills are also a very cost-effective way of targeting. For some campaigns, the presenter recommended using a broader selection with these cheaper criteria.

Groups are a very good way to target people on LinkedIn because of the effort that is required to join a group – you’ve got to search for and find the group first. The presenter suggested this made it an excellent criterion for targeting as group membership is a strong indication of interest in a topic.

The following helpful table was presented showing the most common ways of targeting, although as with many generalisations performance can vary so the presenter recommended that advertisers experiment to ensure that their campaigns behave in the same way.

 

AccuracyVolumeCost
Job TitleVery PreciseLow$$$
Job Function and SeniorityVery BroadHigh$
Skills and SeniorityBroadHigh$$
Groups and SeniorityVery PreciseLow$$

LinkedIn Targeting Tips and Tricks

The presenter had some good ideas for improving targeting in different situations:

If you are targeting based on skills and groups, it’s often a good idea to exclude sales, biz dev and marketing job functions. In any group, a percentage of the members will be people who aren’t customers but are trying to sell to the audience. Of course, this doesn’t work if you are an agency trying to reach marketing or salespeople!

When targeting SMBs it’s important to remember that a lot of micro business don’t have a company page, and so large numbers of your target audience will be omitted if you pick specific company sizes. It’s better to select your audience by excluding large companies, rather than including small company sizes.

Age and years of experience seem like good proxies to determine the level of seniority, but generally don’t produce as good results as selecting seniority as a criterion

A final tip was to avoid audience expansion. With this targeting, the data is messier, and there is no reporting split showing the impact of the expanded audience vs the selected audience. The presenter questioned whether it is ever beneficial, and this matches our experience: we just don’t recommend audience expansion as an effective way of targeting.

Finally, and perhaps most importantly, the more criteria you use, the more expensive your clicks will become. So, simplify your selection whenever you can: fewer targeting items means that base cost is reduced.

LinkedIn Advertising Bidding: The Professional Approach

Rather amusingly, the presenter’s recommendation for bidding was basically to ignore everything LinkedIn tells you. His approach is to begin with an artificially high daily budget, and the to bid low for the clicks – $2 was recommended. If you do this, LinkedIn won’t accept your bid, and will tell you the minimum bid. Bid this amount. Don’t bid higher than the minimum bid, and definitely don’t take any notice of the values that LinkedIn says other people are bidding.

This is interesting advice, and I’m going to run experiments to see if this is the right approach. At Napier we generally bid low: let’s face it there isn’t a competition to be number one like there is for AdWords, so why bid more to be the first advert shown in a feed?

Although there is obvious logic and the approach is similar to how we bid on AdWords, I do feel that there is an issue with always bidding the minimum amount if you have a small, highly targeted audience. If a lot of people are competing to show ads to this audience, and you bid the minimum, the number of impressions you achieve will be reduced. If you are marketing a high-value product or service to a small and influential audience, you might decide that CPC isn’t the only metric, and paying a bit more to get as many people in the audience see the advert is a worthwhile move. The speaker did say that he generally recommends 20K-80K audience sizes, so he does seem to be targeting much bigger groups than some of our very focussed targets, particularly our ABM campaigns.

Finally, don’t use auto-bidding. The speaker claimed the above strategy delivered a CPC that was at least $1 less than auto-bidding. Auto-bidding uses a CPM, and bidding by CPM is not recommended until you know whether you have a good CTR or not (see below).

CTR and Bidding on LinkedIn

The presenter recommended always starting by bidding for clicks. Of course, CTR has an impact on cost per click: the CTR is really the best proxy you have for a relevance score.

If you have less than a 0.35% CTR the presenter suggested that this is when your CPC will be negatively impacted. 0.35% to 1% is good, and when you get to more than a 1% CTR then it’s time to switch from bidding for clicks to bidding on a CPM basis as you’re likely to get better value.

What Content Should I use for LinkedIn Advertising?

Like most things in advertising you need to be in the Goldilocks zone: the content must not present too much or too little friction to the audience. High friction offers like product demos that require a significant commitment from the audience are likely to result in low CTRs that drive high CPCs, and they risk being disabled by LinkedIn for poor performance.

Offers with too low friction, like a blog post or Infographic don’t tend to deliver the value that is demanded from the relatively high CPCs associated with LinkedIn advertising. So, the “just right” offers include things like a guide, white paper, eBook and webinar.

Exclude Competitors from Your Campaigns

The presenter recommended creating a competitors list that is run as an exclusion for ALL campaigns. This not only avoids wasting money on ads that reach competitors, but also helps keep what you are doing secret from competitors.

A pro-tip for competitive analysis is that you can see the adverts that your competitors are running on LinkedIn. Just go to their company page and click on Ads: it will show you any adverts that they are running on the platform. You can’t, unfortunately, see to whom these adverts are targeted.

Testing Strategies for LinkedIn Adverts

The speaker suggested that there are just two things you need to do when testing: firstly you need to get a click, so the initial step is to make sure that you have an advert that has a good CTR. Once you have a reasonable CTR, then look at how well your landiong page is converting.

The presenter had an interesting analysis of the impact of making changes to different element of the campaign. Modifying advert copy typically only has a 5-15% impact on return on investment, landing pages changes can result in 50-150% improvements while simply tweaking the title of the content being offered can increase conversions by two to three times. Although this is impressive, he sees sales enablement improving the follow-up preocess as having the opportunity to increase sales by 10 to 20 times!

Although we’ve seen very different impacts from different campaigns, we do agree with the basic promise: the closer to the sale, the bigger impact you have. So focussing on advert CTR is not the best approach.

One thing you do need to remember is that adverts tend to saturate after running for about 30 days: generally CTR drops as more and more people have seen the copy. So, although spending a lot of time optimising the advert is not recommending, refreshing on a roughly monthly schedule is recommended. An easy way to do this is simply add a new image to the post.

One thing that did surprise us was that the speaker felt that the image used for ads didn’t have much impact on the CTR. This isn’t something that has been reflected in our experience: we have found that a poor image can have a dramatic negative effect on the CTR, so we would recommend spending some time creating great images. It’s not hard, and there are lots of cheap tools to help you.

Three LinkedIn Advertising Insights

For me, the top three insights from this presentation were:

  • Sponsored content is usually the best advertising format, but you must have a website that looks good on mobile
  • Always bid low and ignore the recommendations for CPC bids provided by LinkedIn
  • Focus your optimisation efforts on the bottom of the funnel where you can have the biggest impact on RoI

Source

These notes were taken from the Advanced LinkedIn Ads for the B2B Marketer, presented at Inbound 2019 by AJ Wilcox, Founder, B2Linked.com.