LinkedIn Ads Matched Audiences are the Greatest Feature

Show Resources

Here were the resources we covered in the episode:

Article on LinkedIn Audience Network Measurement Error

Contact us at Podcast@B2Linked.com with ideas for what you'd like AJ to cover.

Matched audiences aren't new in social advertising. But LinkedIn gave us something special that no other platform has. What is it you ask? Listen, find out.

Welcome to the LinkedIn Ads Show. Here's your host, AJ Wilcox.

Hey there LinkedIn Ads fanatics! So Facebook blew us all away by offering custom audiences back in 2013. That allowed us to upload lists of individuals for targeting of ads. It also included website retargeting through the facebook pixel, which was very cool indeed. Then Google Ads released customer match in 2015, it was a lot more limited, but still pretty cool. Ever late to the game, LinkedIn released matched audiences in 2017, with very little fanfare, but this was the most epic release to date. Not only could we upload individuals, which was a total game changer, and we can of course do website retargeting. But we got a feature that no other platform can match, account matching, or company name, uploads, whatever you want to call it. Today, we're going to discuss why this is the most important feature you'll ever use on LinkedIn Ads. This month, I listened to a podcast by a woman named Lisanne Murphy. And it's called The Marketing Matrix podcast. It's really good. If you're into Facebook ads, it inspired me to do my episodes a little bit less like a Wikipedia article where I just toss everything out there that you could ever imagine about a topic and make it so you have to take tons and tons of notes, and instead focus more on individual features that you can walk away and focus on much smaller facets of the LinkedIn Ads platform. So I would love your feedback as you listen to the next few episodes. If this is a little bit easier to listen to a little bit less Wikipediaish. Or if you still found yourself scrambling for a pen. Lee Gannon, who's a friend of mine and a friend of the show, he found out something that was pointed out to him by LinkedIn support, he was asking about skills targeting and when he found the documentation that the support was referring to, it said something about how skills targeting is done. It reads, "skills include those explicitly added by members in their skill section, as well as keywords and phrases mentioned throughout their profile and summary." Okay, just capitalizing on what this means. This means not only are we targeting people with the skills listed in their skill section that they themselves added, but LinkedIn is also apparently looking for keywords and key phrases throughout the profile. I'm imagining in your about section, or maybe even in your headline or title. It goes on to say, "LinkedIn uses modeling to infer skills from a members job title and job description." Alright, this was a big shocker to me. And actually, one that I don't know is actually out there in the wild. I have a feeling that this is actually not something that has been rolled out yet, but probably will in the future. And the reason I say this is because when we have leads come in to one of our clients accounts that are of poor quality, we asked the client to bring us a link to the the client's profile, and we go figure out which campaign was targeting them, and what sort of targeting we were using. And every once in a while we will find skills targeting that brings in someone who was of lower quality, but we haven't yet experienced where when we look for the skill that we were targeting and we look down into their skills that that particular skill is not there. So that leads me to believe that this either happens very, very rarely, or it's just not happening yet. I would absolutely love it. If any of you listening, if you have any evidence of this happening, maybe a skills campaign was targeting someone that you cannot find that skill in their profile, please, please please reach out and let me know that would be really great to know. And then Rishabh Rastogi from India, he put in the LinkedIn Ads support group, which is a LinkedIn group. If you're not already a member of it, I highly recommend it. I'll link to that in the show notes. He pointed out an ad format that looks like a text ad, but it was actually in his newsfeed. And he provided a screenshot of it. It was in the newsfeed right in between two posts. And it was actually right above a promoted post, a sponsored content. And because we know sponsored content only goes at max one out of every five slots in the newsfeed and this was right above it. We know this wasn't sponsored content. So he reached out about it. And this one in particular, the headline says LinkedIn ads drive business results reach buyers with the power to act. And then it does have an image and it's around image and the call to action says Create Ad so this is obviously one that LinkedIn is using for themselves to market their own LinkedIn Marketing Solutions products. This was pretty interesting. And I would absolutely love it if we got this kind of an ad format in the future, because I don't know about you, but text ads are one of my favorite ad formats on LinkedIn. They're inexpensive, they're really good for branding, they do such a good job of propping up all the other ad formats that I'm running at the time. And the only problem is that because they're way over in the right rail, and they're on desktop only, so most users aren't seeing them that they get clicked on very, very little. So if I could have a text ad that was in the newsfeed, I think that would probably be really, really powerful. And then Tamas Banki from Budapest, he shot me a private message. He saw this new ad format and wondered what it was. So what it is, it's a sponsored content post that says it's from LinkedIn. And it does say promoted. And then it asks, How familiar are you with and then lists the company name. And then there's some, it's like a poll, like a sponsored poll where they can click very familiar, somewhat familiar, I've only heard the name, and not at all familiar. And he hadn't seen this anywhere. So he was wondering, is there any way that this is a new ad format. And the truth is, this is an ad format that everyone can get access to, if you're spending at least $90,000 per quarter. And maybe you've even seen it inside of campaign manager, there's a new heading right at the top there. It's right in between campaign performance and website demographics. And it's called testing. When you click it, it will come up with the ability to create a test. Now we've been able to run these tests if you are spending high budget and had a LinkedIn Rep for quite a while. But this is now right on the front end where anyone can run these. The way it works is if you're spending at least $90,000 per quarter, as you spend, LinkedIn is going to ask people questions about your brand. There are six different studies that you can run and if you want to run all six of them, you just have to make sure that you have a minimum budget of $270k for the 90 day period. So immediately the small spenders are out. But you large spenders, this is actually really fun to run. The first is a brand lift test. And then there's also an aided awareness test, a brand familiarity test, a brand favorability test, a brand recommendation test, and product consideration test. And what it does is as you are advertising at obviously high volumes, it's asking people how familiar they are with your brand. And by doing that, you'll get data back about how effective your ads are from a branding perspective. It's really cool.

Next, just this week, LinkedIn started notifying their customers who were using the LinkedIn audience network. There was a measurement issue we had to do quite a bit of digging for because if you go to the Adweek article that LinkedIn published, there is like zero information about what actually happened. It just says that this mis measurement applied to roughly 8% of LinkedIn Marketing Solutions customers, and for 66% of them, the impact was less than $25. So of course, LinkedIn is crediting that back, some of our clients were getting refunds in the $1,000s of dollars, which obviously made us really curious about what caused this. So in doing a little bit of digging, a source at LinkedIn responded that the effective timeframe is a little over two years, with a heavier impact occurring during the two months leading up to the discovery of the issue. If you want to calculate the impact, basically, you look at the total number of LinkedIn audience network clicks, and basically discount 1.21% of them for any date that was from June of 2021 to before that. This is the average impact that could be on some smaller campaigns. They said this is the average impact so some accounts could see a much smaller effect, and some could be higher. And the effect was higher here in May and June. It seems like every time an ad platform comes out and admits that they had some kind of a mistake, and they go to credit their advertisers, which I think is really awesome because they could just as easily just not tell anyone and let it go. But they do they more or less fess up to it. They give people refunds, which I think is super honorable, but then some jerk or some set of jerks. always end up suing them over it. So please, please, please do not join a class action lawsuit against LinkedIn Marketing Solutions. I would reserve those kinds of actions for companies who do terrible things to you and then don't admit it and you have to find it out yourself. Okay, that's it for the news.

Then we've got a couple of reviews that came in one by the user wixfi. It says, "The authority and LinkedIn ads. AJ is a great voice on running LinkedIn campaigns and doesn't hold back. He gives strategic and tactical tips on how you run the ROI positive campaigns for your business." Wixfi, thank you so much for saying that. I really do try to not hold back and just share everything I know. And of course, our goal is always ROI. And then Betsy Hyndman, from Nashville wrote, "AJ is the real deal. Great podcast, super guy, very knowledgeable. I've learned a ton from AJ generously sharing his deep expertise." Betsy, I'm glad to call you a friend. Thanks so much for leaving such a kind review. All right for you right now who are listening. Yes, you! You're listening, you haven't left a review, please go do it. Honestly, it helps the show a lot. It's going to get more people to listen to the show. And plus, I get to shout you out and tell people how awesome you are for saying such awesome stuff about us. So here we go. But honestly, please go do leave a review, I would love to feature you.

Okay, with that being said, let's hit it. We're getting right into the meat of the show today. So we're talking about matched audiences. And my favorite part about matched audiences are the list uploads. And so there are two different kinds of list uploads you can do one is a contact list, which you think of as maybe email matching, but I'll tell you why that's not, it's a little bit of a misnomer. And the next one is the company name list upload or the account list. Sometimes you'll hear it referred to as the ABM account targeting list. It's important to understand that with all of these list uploads, there are three things that you can do with it. You can either include that list in your targeting, just targeting those who are on your list. You can exclude those from your targeting. And then you can also use that to create a look alike audience. As long as you're listening to this since 2019, you've been able to do that. So let's touch on the the individuals the contact list upload, there are so many different uses that you can have for uploading a list of individuals. You can use this to target individuals who've joined your email list, that's pretty cool. You can use it to exclude your current customers from seeing your ads. How awesome is that, that you can avoid showing ads to someone and having them pay or you know, charge you, to click on your ads when they're already a customer. So lots of different uses. Now, it's a little known fact that you don't actually need an email address for targeting here, more info on that later. Then you've got the company name list upload. And this is, like I said in the intro, my favorite part of LinkedIn Ads in general. What this allows us to do is upload lists of accounts for inclusion or for exclusion. So inclusion would be like if I had a list of companies on my account based marketing list, and I wanted to fire out some ground cover across those audiences so that when they see our next ads, they're much more likely to interact. They're one of our target accounts, we want them. You can also do the same thing of uploading a list of let's say, your competitors, the competitors by company name, and then you could exclude that list from all of your campaigns. And now all of a sudden, your competitors have no idea what you're advertising, you're flying completely under the radar. That's pretty cool. You can also exclude your current customer list, the companies they work at, or how about you could include a list of all of the companies who've become a lead for your organization, but haven't yet closed. So this becomes kind of a lead acceleration or a sales acceleration type of campaign.

Alright, so why is this feature so sexy? Why is it my favorite part of LinkedIn Ads? Well, here's the reason. This is the one feature that no other platform can touch. Facebook, they allowed us to target for a long time people's organizations and their job titles, but guess what, so few people on Facebook ended up putting their professional information in. Facebook is just not the place where you have that kind of data. And so even if Facebook released the ability to target by company in bulk, it would just give you access to such a small percent of the population, it just wouldn't even be worth it. But hey, everyone on LinkedIn tends to list the company they work for. That's kind of the point. Add that to the fact that when you're doing account based marketing targeting, you're targeting a much smaller population. And what that means is, it's not going to allow you to spend as much money. This is a downside for the larger advertisers. But for the smaller advertisers, this is something really powerful you can do and it doesn't take much budget budget. I firmly believe that every B2B company on the planet should be doing this, targeting their absolutely ideal accounts that they want to go after. And it doesn't cost very much, just a few bucks here and there to reach your absolutely ideal audience so they know who you are. That's pretty awesome. Also in the LinkedIn Ads support group on LinkedIn, Jennifer Karos asked this question. She said, "Hey, guys quick question, we uploaded a list with roughly 12,000 contacts. The system recognized about 6,000 of them, but it says it matched 85%. Do you know how this inconsistency could be?" And then Joanna from LinkedIn came in and said, "Hi, Jennifer. I'm Joanna. from our Product Marketing Team at LinkedIn, we see this when we find a match to multiple inputs. For example, if a personal and business email was provided, and we found a single member match to both records, you would see a higher match rate versus the member count." So to put that into perspective, if you upload a list with let's say, first name, last name, company, and title, if it found a match with the company name, and the title on, let's say, half of everyone, it could still say 100% match, but you'd only end up with half of the contacts in your list being represented in your targeting. This was absolutely news to me. So thank you, Joanna, for explaining this to us. We'll come back to this for sure. The next is, it's really important to understand when you upload a list of any kind, it goes through a processing stage. If you've uploaded one of these, you'll see the words "your audience is currently building and may take up to 48 hours or on rare occasion longer to start delivery". One thing I really like about this is you can attach an audience to a campaign even before it's finished building. And what's going to happen is as soon as it's done building, it will immediately start serving the ads. So you don't have to be watching for the status to change, and then go and launch your own ads. Okay, so LinkedIn says right there in their documentation, "It may take up to 48 hours, or on rare occasion longer for a list of process." I call bs on this, I have never ever seen a list to finish processing in under 48 hours. Most of the time it goes to 72 hours and beyond. There's obviously not much I can do. But that's pretty disappointing. I would love to see LinkedIn do something about that. Okay, here's a quick sponsor break. And then we'll dive into some research about what I found that gets lists to match at a higher rate.

The LinkedIn Ads Show is proudly brought to you by B2Linked.com, the LinkedIn Ads experts.

If the performance of your LinkedIn Ads is important to you B2Linked is the agency you'll want to work with. We've spent over $140 million dollars on LinkedIn Ads, and no one outperforms us on getting you the lowest cost per lead at the most scale. We're official LinkedIn partners, and you'll only deal with LinkedIn Ads experts from day one, Fill out the contact form on any page of B2linked.com to chat about your campaigns. We'd absolutely love to work with you.

All right, let's jump into it. So I've been doing audience research now for years. And what I share with you this certainly isn't like a benchmark data by any means, but they should help you understand a little bit about how the matching algorithm works. So a couple years ago, when LinkedIn stopped including email addresses in our connections export, I exported all of my connections on LinkedIn without their email address, and then uploaded it to LinkedIn. This was so interesting, because it reported a 90 plus percent match rate, but then the resulting audience that it gave me was only a quarter of the size of my network. So that means there were a lot of those records matching on other things, but a whole bunch that matched on none, at least from what Joanna from LinkedIn shared. So then what I did is I uploaded a list of all of my followers, and this was just their first and last names. And I just wanted to see what it did. It matched it only 50%. But it did return about half the number of my followers. Because first and last names really aren't all that unique. I would not be surprised at all, if some of these were matching people who had the same name as my followers, but weren't actually the followers. I didn't actually advertise this audience. That would be, I don't know, I'm not very confident in that one. And then me and my team found out that you can get a much higher match rate by including in addition to first and last name, also job title and company name. By doing this, we ended up finding that we got a 90% plus match rate, which is awesome. And then just last week for fun, I uploaded my followers list again, my most updated one,, and I wanted to test two things. I included in both of them first name, last name, and company. But for the job title field, I wanted to see if LinkedIn pulled more from their actual job title in the experience section or if it was their headline. My hypothesis was actually that LinkedIn was going to pull more from their headline because more people use their headline I feel like, but boy was I wrong. When I use the headline as the job title, LinkedIn reported that it matched 85%. But it only matched just a little over 61% of the contacts that I uploaded. Okay, that's interesting 61%. Then I uploaded that same list, but with the job title field as the job title. LinkedIn reported this as matching at 90%, but it actually matched 75%. So that's 14 more percent of a match, just by using the job title field instead of the headline. That was really interesting to me. So the takeaway is here, the now what, when you go and upload individual lists, make sure you include first name, last name, job title, their actual job title would help, and their company name, that's going to help you get a much, much higher match rate. One thing you'll notice if you go in and export a list of connections from LinkedIn, if LinkedIn sees an @ in any other field, except for the email address field, it's going to fire off an error. So for instance, if someone has the @ in their headline, because they're saying, I'm the VP at this company, you're gonna have to get rid of that before the list will even validate to process. The way I do that is right inside of Excel, I just do a search and replace on every column except for email address, where I search for the app sign and replace it with nothing.

Okay, I know the LinkedIn products team listens to the show. So I'm going to give you and everyone else here, my wish list for how LinkedIn can make their matched audiences product even better. The first is, like I mentioned before, faster processing. It really shouldn't take 72 hours to process lines of text, I would expect it to take that long, if there was a human in the background, like manually doing things in Excel before the list could be processed. That would make a lot more sense. If the list has 300,000 rows, which is the maximum, I could understand it going longer. But like I said, I've literally never seen any list finished processing under 72 hours. And most of the lists that we upload are in the 10,000 to 30,000 rows range so we're not maxing it out by any means. Something else I would absolutely love in the contact list. I want to be able to target by LinkedIn URL. Forget matching by first name, last name, company name, email, job title, if you have the LinkedIn URL for the person that you're going after, even better, I would much rather use that, it would match it 100%. Please LinkedIn give us that. Then back in 2017, when we got the matched audiences feature, I asked why LinkedIn just shows 90% or higher as a match rate for company names. Why can't they show us the actual if it's really 100%. And LinkedIn replied that it was because they were worried about privacy. Now I get it being a privacy issue, if you are showing an exact match for individuals for a contact list. But there is absolutely no privacy issue with company names and targeting. A company name is public. I just think that's totally a non issue here. So I would absolutely love it. If we got a real match rate for company match. Show us the exact percentage matched. Now recently, if you go into your your list uploads, there will be a tab for companies matched and companies not matched. And that's really cool. If I targeted let's say I was targeting IBM, and I typed in just the letters IBM, but LinkedIn was only going to match it if it was i dot b dot m dot, then if you go into that list, it will tell you that IBM did not match on your list. And that's pretty cool, you can fix it, but I just don't see why the general percentage match shouldn't reflect the exact match for company. And finally here, we've talked about this a little bit, but the match rate really should be based on the number of matches that occurred out of the total number of rows uploaded, I definitely shouldn't have LinkedIn reporting to me 85% match rate to do my own calculation and find out it's actually 61%. That one definitely seems like a product that was built by engineers, not advertisers, I would love to see that one fixed, or heck give us both metrics. Maybe one shows the number of rows that resulted in a match and another one that shows the strength of each match. That could be cool, but I'd want to see both. I don't want to just have the strength of the match shown to me instead of the raw number. Alright, I've got the episode resources coming right up for you. So stick around

25:01
Thank you for listening to the LinkedIn Ads Show. Hungry for more? AJ Wilcox, take it away.

All right, first of all, there's that article that LinkedIn published about the LinkedIn Audience Network, and its measurement glitch. So I'll link to that it's on Adweek.com. Just be prepared, the article says absolutely nothing, but you're welcome to take a look at it. Also, if you're trying to learn LinkedIn Ads better or have a colleague or a coworker who is make sure to point them towards the LinkedIn Learning course that I'm the the author of. It just got a refresh early this year and we added about 25 extra minutes of content plus updating everything. It's really good. If I don't say so myself. If this is your first time listening to the show, thanks for tuning in. please do hit that subscribe button on whatever podcast player you're listening with. Please do rate the podcast and like I said before, it really helps the show if you review. That's exactly how you can repay me for dropping all this knowledge, those are knowledge bombs. Okay, so if you want to reach out to us, give us any ideas or any feedback for the show hit us up at podcast@B2Linked.com. And with that being said, we'll see you back here next week. I sure hope. We're working on more consistency here. I'm cheering you on in your LinkedIn Ads initiatives.