Jun 9, 2022
Show Resources
Here were the resources we covered in the episode:
NEW LinkedInLearning course about LinkedIn Ads by AJ Wilcox
Contact us at Podcast@B2Linked.com withideas for what you'd like AJ to cover.
Show Transcript
Micro segmentation of your LinkedIn Ads. Your LinkedIn Rep saysit's a bad idea. I say it's great. Let's discuss why this strategyis so divisive. On this week's episode of the LinkedIn AdsShow.
Welcome to the LinkedIn Ads Show. Here's your host, AJWilcox.
Hey there LinkedIn Ads fanatics, your LinkedIn Rep may have warnedyou against something called micro segmentation, which is what theycall it when you create a campaign that has too few members andyour target audience. Now, I highly recommend micro segmentation.But our definitions are very different. On this episode, we'll diveinto how you can use segmentation to get better ad performance,easier reporting, and definitely to learn more about who yourtarget audience is, and what they engage with. Let's jump right inand hit it. And definitely the next time I decided to coin a term,I'm going to try to make sure that it doesn't match what a LinkedInRep warns you against.
So everyone knows that LinkedIn ads is great for lead generation.That is, of course, why we come to the platform is access to anultra premium audience. But a little known fact about LinkedIn Adsis that it's also spectacular for market research. And by marketresearch, I'm not saying run ads, asking people about what theythink about your company or anything like that. But you probablyalready have a good idea of who your ideal customer persona is,right? That's why you're using LinkedIn Ads. So you can targetexactly those people. So for example, let's say that your idealtarget audience is decision makers in IT. It would be really easyto create one campaign targeting the IT job function, and thenlayer in all seniorites that or manager and above. And you can runthat and that's totally fine. But then after you've run thatcampaign, you don't get any sort of understanding further aboutyour audience, you might get good ad performance, you might get badad performance somewhere in between. But ultimately, because youdon't have anything to compare this against, you are left with nolevers to pull. So instead, imagine that instead of that onecampaign, you've created four separate campaigns. One is IT with amanager seniority. Another one might be IT with a directorseniority, another might be IT with a VP seniority. And thenfinally, your fourth one would be IT with a CXO seniority. So let'ssay you are running an AB test to this audience, well, you put thesame AB test in each of those four campaigns and now you'retargeting the same size of audience, you didn't have to create manymore ads, you just created the same to over and over again, but nowyou have four campaigns that you can compare against each other, tolearn more about what different segments of your audience aredoing, what they like, what they engage with. By doing this, youcan get some insights like for instance, maybe CIOs convert higherto the lead. But maybe they're harder for the sales team to get ahold of for a demo. So their cost per SQL ends up being higher. Andmaybe IT managers have a higher cost per click, but they're easierfor sales to get an appointment with. So maybe they have initialhigher costs, but when you let it all bake out in your salesfunnel, it all of a sudden makes a lot more sense. I like to dothis basically, anytime that I'm targeting an audience withmultiple singularities. I'll break up that audience by seniority,so I can learn who likes what content. You might see that certainoffers are interesting to those who are VP and above. And otheroffers are more interesting to those who are like manager anddirector level. And this is really easy to do, you could do thesame thing by segmenting by industry. So for instance, let's sayyou're going after tech companies, what you could do is takeLinkedIn's industry of internet, which is obviously very broad. Itincludes things like Google and Facebook, and Snapchat are allclassified as internet. And then you might do a separate campaignfor anyone who says that they're in computer software. And then ifyou compare these results against each other, you might find thatone industry actually produces higher quality leads than another.And if you can figure that out early, great. Now you can pare downyour budget for the one that isn't working, and feed the one who isand get more of those high quality leads that your sales teamloves. A lot of times I'll see campaigns that either clients havebeen running before they've come to us, or accounts that I'mauditing. And I'll see that they'll have one campaign with a wholebunch of different job functions. So let's say marketing, andaccounting, and business development are all in the same campaign.And when you ask them about it, they'll say, well, our messagereally is irrelevant to all of those people. And I'm notdisagreeing with that fact. Your message could be really good forall those people. But the fact that they are all lumped into thesame broad campaign means that you'll never learn. Whereas if youbroke it out, you might find that, oh, wow, accountants don'tinteract with this at all. So hey, maybe we spend less on them ormaybe we create a custom Add Variation just for the accountants.And this is data that you would only know if you've broken themout. To learn more, you've got to have them in different audiences.We've done the same thing with geographies. So of course, if you'readvertising to a whole bunch of different countries, you know thatthe culture in each of those countries is going to be verydifferent. And so if you haven't broken them out, you might not beable to tell. The big one I remember from my past, I wasadvertising for franchises in all of North America. And I foundthat Canada was not converting very well at all. And I had no ideawhy it eventually dawned on me that in my form, I was asking for azip code. And in Canada, they don't call it a zip code, they callit a postal code. And so it was very, very apparent that I was anAmerican, trying to advertise to Americans, and just added Canadaon. So as soon as I gave them their own form, conversion startflowing in. So you can do this, if you are going after all NorthAmerica. Or maybe you're advertising to all English speakingcountries, like the US, Canada, the UK, Australia, New Zealand, aslong as your audience sizes are large enough, in each of thosecountries by themselves. If you give them their own campaign, itthen allows you to suss out and find what are those culturaldifferences that are making people convert higher, interact higher,convert lower, interact lower, we've even done this in the US withtime zones. So we'll add all of the states that are in the PacificTime Zone, to their own campaign, all states that are mountain timeinto their own campaign, Central Time, Eastern Time Zone, etc. Andby doing that, it allows us to do some really cool day parting.Because if you start showing ads to people on the East Coast, at6am, their time, you'll all of a sudden be advertising to people onthe west coast at 3am, which isn't a great time to be advertising.So if you've put them in different time zones, and then you'returning your campaigns on and off at those times, it's gonna be alot more exact, you won't waste your money nearly as much. If youwant to dive more into day parting and ad scheduling, definitelycheck out episode 63.
7:31
We also really like to segment by company size. So what you can dois say, hey, if we go after all companies that are size 200. andabove, well, let's put 200 to 500 in its own campaign, and then 500to 1000 in its own campaign, and then maybe 1000 to 5000 in itsown, and then 5000 and above in its own. And now all of a suddenyou get to judge what is your cost per lead by company size. And alot of sales teams will look at a company size to tell you whatquality of lead that is. And they care more about the largercompanies, because they're gonna have a lot more budget to do this.And that's super easy to do exactly like we talked about earlier.And there's a whole lot of other segments that you can also breakup campaigns by, for instance, you could do field of study. So whatdid someone study in college, you can do growth rate of a company,those that are growing slowly versus those that are rapidlygrowing, and message them differently, you can really segment byany sort of targeting that is broad. And what I mean by broad, itmeans that there are a few options that broadly define the entiremarket, it doesn't usually make sense to do this to segment by jobtitle, or by group. Because when you put one set of people into agroup, or specify one job title, and then you specify another,you're still leaving out a huge chunk of the population. This isn'ta great way of splitting the audience up to learn more about it,you're going to end up with really small target audiences. So thinkabout how you can break up by some of those elements like seniorityor job function, industry, geography, company size, those types ofthings. Plus, it really didn't cost you anything extra to break outall of these micro segments of your audience. But now you can testall of your offers and motivations to find out what engages anddrives each micro segment of your target audience. Plus, now youhave all these extra levers that you can pull when you want toimprove performance. So for instance, if you're pacing ahead ofbudget one month, it's really easy to just go down and bid down thelowest performing micro segments. And that's going to naturallypush your ad performance higher while restricting the budget. Ormaybe your sales team tells you that certain types of leads areworth more to them. They're a higher quality lead, so you can biddown or shut off the micro segments that are pulling yourperformance down and feed the ones that are more like what saleswants. And since you already have these campaigns, you can take ita step further by adjusting your ad copy to talk directly to thepeople. For instance, in the past, when I was going after businessowners or CEOs, I would put in the ad copy, "Hey CEOs" to try toget their attention. Because you've already defined who youraudience is in these little micro segments, if you can think of acall out to them that they would recognize, this becomes a testthat you can easily do to call out to people. One thing that Inoticed when I was doing this, calling out to people by some sortof an identity, it certainly makes sense sometimes when theyidentify themselves by that entity. But sometimes you'll call outto people, and they don't define themselves by that. So forinstance, if I were going after marketing directors, if I shout outto them, "Hey Directors", they may not actually identifythemselves, strongly by the fact that they're a director, theymight identify themselves more powerfully as marketers, so I mightdo better to say, "Hey Marketers", or "Hey Marketing Leaders".Okay, here's a quick sponsor break. And then we'll dive into howyou can actually use these micro segments to optimize and controlyour account performance.
11:26
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$150 million on LinkedIn Ads, and no one outperforms us on gettingyou the lowest cost per lead, and the most scale. We're officialLinkedIn partners and you'll deal only with LinkedIn Ads, expertsfrom day one. Apply to work with us on the contact page ofB2Linked.com, we'd absolutely love to get to work with you.
Alright, let's go ahead and jump into how you can actuallycontrol and optimize your account when you have these microsegments. So I use the example before how if you're pacing ahead ofbudget, you can turn some things off that aren't super helpful toyour account. Every so often, a client will come to us mid monthand say, for some reason, my boss just cut our budget in half forthis month, we have to slow things way down. Well, if the accounthas an average cost per lead of like $80, for instance, it's reallyquick to go and segment all of my campaigns by their cost per lead,and sort from high to low. And all of a sudden, here's this list ofthe 20 campaigns that are bringing leads in for significantly morethan that. So I have some that are bringing in leads for $100 andsome $150 and $250, for instance. So we're reacting to a prettyurgent call from the client to cut down on budget and spend less.But by doing it this way, we go in and pause all of the campaign'sthat are performing worse than average, we've stopped spending asrapidly. And now we don't have spend going to those campaigns thatare having a high cost per lead. So as we're cutting down on ourspend, we're also watching our average cost per lead drop. So itwas an ad. And after cutting out the higher CPL campaigns, we mightbe averaging 60 or 50. And it's a great way to finish that month,especially if we know that we had less budget to work with. We hada client who was in the business financing space, and they hadthree different industries they were going after. So naturally, wehad the campaigns all segmented out by industry and we found thatthe retail industry for whatever reason, was underperformingcompared to the others. And so it was really easy to go through asa blanket change and pause or bid down all of the retail industrysegments. And we put that budget into the to other industries thatwere performing well. And it's important to know that if you hadall three of these industries lumped into the same campaign, youwould have totally missed that fact that that was an optimizationyou could do. If you haven't listened to episode seven, where wetalk about account organization, you'll definitely want to go andlisten to that. Because I recommend a campaign naming structurethat identifies within the campaign name exactly who that audienceis. And what's so cool about this is when we micro segment andwe're describing who the audience is right within the campaignname, it makes reporting super easy, and gives you capabilitiesthat you wouldn't have otherwise. So for instance, if all of yourcampaigns that are targeting directors have something like DIR fordirector in them, when you go to start your campaigns and look atthe results by campaign, you can very quickly find all of thosethat are targeting directors. And if you find that directors areperforming really well, great, you now know all of the campaigns togo and increase your bids on or give some expanded budget. Also inour campaigns, we identify the type of targeting we're using. So wemight put JF in a campaign name, when it's targeting by jobfunction, we might put JT, if it's targeting by job title. Thenwhen we get all of the results, let's say from the whole salesfunnel, we can very quickly combine in a pivot table all of thoseaudience names, and combine all of the performance from job titletargeting versus job function versus skills targeting versus groupstargeting. And we'll find some things like oh, wow, job functionover here is more expensive, or leads to more volume, or groupstargeting is really, really limited on audience size. But boy, it'sefficient when it works. So definitely go back and listen toepisode seven to get more information about how we do that so youcan do the same thing.
15:57
I'm sure many of you already know my feelings on using audienceexpansion on LinkedIn. Let me tell you why I hate it. Because it isexactly the opposite of micro segmentation. We're talking abouttaking specific audiences and breaking them out so that we cancompare them against each other and learn from the differences.Well, when you use audience expansion, it takes an existing largeaudience, and it adds people to it, that you don't know who theyare. You don't get to gauge the quality of that audience untillater when your sales team comes to you and says, hey, the qualityof leads have been really crappy recently. So not only are yougiving LinkedIn the opportunity to insert low quality audiencesinto your campaign, but it mashes those audiences together in areally opaque or non transparent way, and makes it impossible toactually learn from. So if you want to expand on your audiences, Iwould highly recommend go and create a look alike audience instead,because at least a look alike audience will allow you to run thenative audience and the look alike that was created from itseparately. And now you can compare and find out like, oh, lookalikes are performing well, or no, they're not, we should stopdoing that. So please, please, please, if you have audienceexpansion turned on anywhere in your account, please go turn itoff. You can thank me later. So I can already hear a couple of yousaying like, well, if I break out all of my audiences into microsegments, that's a lot of campaigns to manage. And yeah, that's thetruth, you may be managing more campaigns, but it only takes a fewminutes to create a duplicate of a campaign. And so the rewards ofdoing this are much, much higher than I feel like what the costincrease is for you in managing the account in your time. Plus, wemanage accounts with 1000s of campaigns in them. So I can tell you,it's definitely scalable, it's possible, you're not going to make ahuge mistake by breaking things out to learn more about youraudience. It's also really likely that you're not going to berunning all of these campaigns forever. Because once you find thosemicro audience segments that are underperforming in comparison toothers, you can always cut them out by pausing. I mean, obviously,pausing is the best way, but you could always bid them down if youjust wanted to get their costs lower. And when you take thosecampaigns out of the mix, you're giving yourself fewer campaigns tomanage. I can also hear some of you saying, Well, my LinkedIn Repsays that I should only run campaigns that have an audience size ofat least 300,000. Well, by breaking up your existing audiences intosmaller segments, you're still targeting the same audience size intotal. So you can tell your LinkedIn Rep, we're still targeting thesame people, they're just in more campaigns. Plus, now you'll seebetter performance with your ability to turn off or bid down anysegments of the non performers, which you wouldn't even be able totell were being targeted before when it was just one big lumpcampaign. And my opinion is, you're already paying a premium forusing LinkedIn, which means you're paying more for LinkedIn leadsthan you are under their channels. So why not have that budget alsogive you insights into the behavior of your customer persona. Plus,everyone who's running LinkedIn ads is always looking for ways ofdecreasing costs, because costs are already so high. This is one ofthe ways that you can do that really effectively. All right, I'vegot the episode resources for you coming right up. So stickaround
19:29
Thank you for listening to the LinkedIn Ads show. Hungry for more?AJ Wilcox, take it away.
All right, here are a couple of great resources for you. Numberone, go check out Episode 63 on ad scheduling. To find out how weschedule our ads at different times of day and different days ofthe week. Check out Episode 07 about account organization for howyou can make it easier to report and easier to find insights. Ifyou or someone on your team are looking to learn more aboutLinkedIn Ads, check out the course that I did on LinkedIn Learningwith LinkedIn. The link is right there in the show notes below. Andthis is by far the least expensive, and the best quality course outthere. If you're a new listener to the show, hit that subscribebutton. So you can hear more of this when we come out with newepisodes weekly. And please do consider rating. And definitelyleave us a review on the podcast. We would absolutely love to hearwhat you think. And when you leave a review, we'll shout you outlive on the show. With feedback or suggestions on the podcast,reach out to us at Podcast@B2Linked.com. And with that being said,we'll see you back here next week. Cheering you on in your LinkedInAds initiatives.
