Jun 16, 2020
Show Resources:
LinkedIn Learning course about LinkedIn Ads by AJ Wilcox: LinkedIn Advertising Course
Ep 287 of LinkedInformed Podcast
Contact us at Podcast@B2Linked.com with ideas for what you'd like AJ to cover.
Show Transcript:
How big should your LinkedIn Ads audiences be? Hearing conflicting advice? Let's clear that up right now.
0:15
Welcome to the LinkedIn Ads Show. here's your host, AJ Wilcox.
0:23
Hey there LinkedIn Ads fanatics. There's a lot of conflicting
advice out there about audience sizes for your LinkedIn ads. And I
get questions about it all the time. So I figured we'd go ahead and
clear that up. We've tested about every which way you can think and
we've come up with our own strategy for what types of audience
sizes work best. So let's definitely get into it. First off the
news. This week, Microsoft had their quarterly earnings release.
And this was super interesting to me. I actually heard about this
from the LinkedInformed podcast by the great Mark Williams if
you're not already subscribed to that podcast, it goes over the
organic side of LinkedIn. And it's absolutely brilliant. In Episode
287, he goes over this in his new section. And even though LinkedIn
isn't actually publishing these stats, Microsoft is. So Microsoft
says that LinkedIn's revenue growth is up by 21%, which is
fantastic. And they also mentioned something really cool and really
key. Sessions are up 26%. Oh, I love this. These aren't stats that
LinkedIn is sharing too often. But if Microsoft will share them,
I'll take them. He also mentioned in this episode that an unnamed
source, let this news outlet know that conversations on LinkedIn
are up 55% in March, and comments in the newsfeed are up 272%. So
similar to what we've talked about in previous weeks with cool
COVID news happening. I know nothing's really cool about COVID, but
there have been some cool things coming from it. I wanted to
highlight a few of the reviews that you guys have left. Betsy
Hindman in Nashville, who's a good friend Betsy, thanks so much for
listening. She says "AJ is the real deal. Great podcast. Super guy.
Very knowledgeable." Thanks so much for leaving that. Then Robinhfr
from France says "really helpful. I love this content, so much
value in this podcast". Robin thanks so much for leaving that it
means the world to me. Maria Cole mentions "the most knowledgeable
LinkedIn Ads experts out there. AJ knows more about LinkedIn than
anybody I know. His podcast is full of amazing insights and
knowledge. Often I find myself agreeing with him out loud while
listening". Oh, I love this. I hope all of you are kind of nodding
along if if we're striking a chord and touching on things that you
found through experimentation, Oh, I love to be geeking out with
you. Go to whatever podcast player you're on. And leave us a
review. I would love to help feature you. Okay, with that being
said, let's hit it. When I was very first getting started running
my very first LinkedIn Ads back in 2011. I read the recommendation
that LinkedIn recommended 300,000 in an audience. And this was back
when the only ad format we had was text ads. And text ads, as you
know, average somewhere around a .025%. click through rate, which
is incredibly low. Then in 2014, LinkedIn came out with the
sponsored content ad format. And I was absolutely certain that
these recommendations were going to change because text ads had a
.025%, click through rate and sponsored content had 12 times the
engagement. But sure enough, for years and years afterward, I still
see LinkedIn recommending 300,000 for an audience size. I
definitely think that there was a good reason to have 300,000 as
the audience size recommendation, but now that the ads can
literally spend 12 times or more, I think there's a lot more
opportunity to tighten up our audiences. So we'll go into how I use
smaller audiences and why. But it's important to understand the
history here. And then adding to that the minimum audience size on
LinkedIn used to be 1000. But then back in 2017, in April 2017,
LinkedIn came out with matched audiences the retargeting product.
And at that point, the minimum audience shrunk from 1000 down to
300. And I'm so grateful, thank goodness this happened, because the
tighter we can make our audiences the better. Now I checked out
LinkedIn's Help section on this and link to that is in the show
notes. They say for sponsored content and sponsored messaging, we
suggest a minimum of 300,000 audience size. For text ads, we
suggest that you target between 60,000 and 400,000. Okay, so that's
LinkedIn's recommendations. For some reason text ads are
recommended to be a smaller audience than sponsored content,
something there I'm not understanding. My recommendations are if
you're using sponsored content, try to stick between about 20,000
to 80,000. If you're using text ads, try 20,000 to 80,000. Hey, how
about dynamic ads? 20,000 to 80,000. Oh, AJ, I understand where
you're coming from. How about sponsored messaging? Oh, AJ, I
understand where you're coming from, you're gonna tell us 20,000 to
80,000? No, I'm gonna change the script on you a little bit. For
sponsored messaging. I love to start with ultra tight audiences
because they are able to spend so much more. But when in doubt,
stick between 20,000 to 80,000. And you'll be pretty safe. Okay, so
why are our recommendations so different? Why would you listen to
LinkedIn versus our advice? Are you getting such differing
opinions? It's important to understand where LinkedIn's motivation
is, because to LinkedIn, it is a significantly larger risk for a
new advertiser or a less sophisticated one to come and test out the
platform come and try things out. And then not spend any money and
then leave and go tell their friends that the LinkedIn Ads Form
didn't work, they didn't get any traffic. So they would much rather
that you had a large audience size. So you were pretty much
guaranteed to spend your whole budget. To me though I have a very
different motivation here. To me, it's all about getting the best
performance, the highest efficiency, and the best data out of an
account. And so I use these tightly themed audiences. They act like
little silent focus groups, which I'll tell you all about here in
just a little bit. So certainly, if you don't have the expertise on
the platform, and if you're listening to this, I'm guessing that's
not you. But then you'll want to follow LinkedIn's advice for
audience sizes. It's pretty much foolproof at getting you to spend
your budget without expertise. You may not spend very efficiently,
but certainly all of those dollars got allocated. If you do have
the expertise and attention to manage smaller campaigns, then you
will get so much more control and insights out of your account, and
you'll get a tighter ability to optimize and find efficiencies.
that you shouldn't, quote unquote be able to do realize that as
we're talking about audience sizes here, audience sizes are totally
subjective. I can't just tell you that an audience size of 30,000
is a good thing or a bad thing, because that's totally dependent on
number one, how active that audience is on LinkedIn. And number
two, how often people are clicking on your ads. Because if you had
a minimum sized audience of 1000 people, but you have 100% click
through rate, you could probably still spend a healthy budget. Of
course, that's probably not going to happen, but you get my drift.
Even a small audience that's very active, can spend more than a
large audience that's largely inactive. There are certainly risks
with both styles of campaigns here, you can have campaigns that are
too small and there are definite risks there. And then you can have
campaigns that are too large, and that has its own set of risks. So
first of all, if your campaigns are too small, there's a really
good chance that you won't spend enough, you don't have enough
people in those audiences, and all else held equal, small audiences
won't spend as much as large audiences. Maybe if you are dedicating
the time to build these audiences, but they don't spend very much,
maybe that wasn't worth your time, maybe you see that as a waste.
And then maybe you've got these small audience sizes, which means
you're not going to spend very much, which means you're not going
to generate a large quantity of leads. So maybe you have internal
pressure in the company that not only do you have to spend this
budget, but you also have to hit a lead goal. And then finally, if
data is accumulating slowly, it means that we can't make
optimization decisions because we're dealing with these small data
sample sizes. Well, hang on, because I've got a cool solution for
you that I'll cover here in just a few minutes about rolling data.
And all of these risks are totally mitigated by yes, we build small
audiences, but we build lots of them. So together they act like
large audiences. Separately, we still get data. So hang on, I'll
break that one down. I know that gets a little bit ethereal. But
what about having campaigns that are too large? If you're taking
LinkedIn's recommendations, which we don't recommend, I think the
risks are even higher. So number one, you can have too many
different people in there. So I'll give you an example. Let's say
you had a campaign where you had both finance and operations
professionals in the same campaign. Well, let's say finance
professionals do amazingly well. But operations folks are terrible.
If you're looking at the results from that campaign, it might just
look average, it might be okay. And you missed out on the fact that
if you would have split those out into two separate audiences, you
would have found that wow, finance professionals are killing it for
us. So you're missing learnings about your audience, you're losing
that ability to understand more about what they care about, and
really who your ideal target audience really should be. Also,
audience that are too large, they spend budget too fast. So if you
have a small budget, let's say you're running something at the
minimum of $10 per day per campaign, if you have a large audience,
chances are you're going to blow all of your money in the middle of
the night before your audience even gets into work. So really, it
shouldn't be about what your budget is, on a per campaign basis, it
should be finding the right budget per audience. And of course,
listen back on episode six, where we go in deep on bidding and
budgeting to find out more about that. Okay, I'm going to give you
an example. And this is a real world example of the type of
targeting that we would build. A client comes to us and they say,
our target audience, our marketing decision makers, at companies
with more than 500 employees. So I go and build this campaign in
LinkedIn. And I see a target audience size of 360,000. And this is
in the US. According to LinkedIn recommendations, yeah, an audience
size of 360,000 is fantastic. But I looked at that and went, man,
having everyone from manager all the way up to cmo represented in a
single campaign. That doesn't tell me anything. So what I did is I
broke that out into four separate campaigns. One that is marketing
job function with manager seniority, at companies with more than
500 people, of course, I did the same thing, but instead of
managers, it's directors. And then the same thing instead of
directors, it's VPS. And then finally, I have one just for the the
chief marketing officers. There are 160,000 managers, 160,000,
directors, 58,000 VPS, and then only 26,000 CMOS. So breaking these
into their own campaign, your LinkedIn rep will definitely tell
you, ah, these audiences are too small. You should really make them
larger. But what I've done is taken the same two ads and put them
in all four of these different campaigns. And notice, I'm still
targeting the same size of audience. It's still 360,000 total
senior marketers, no one made it in that audience that wasn't
supposed to be there that wouldn't have been there from the
beginning, I'm still able to spend all of my budget. But now I
actually have information, I have data about which levels of
seniority, engage with my ads, and how and even how they convert.
And I can keep following them all the way through the sales
process, and learn more about what it takes to get a CMO to a sales
qualified lead status, versus someone who's a, let's say, a
marketing manager. Okay, we're gonna take a quick sponsor break and
then we'll dive into the Goldilocks zone for audience sizes, and
actually how to create them.
12:44
The LinkedIn Ads Show is proudly brought to you by B2Linked.com,
the LinkedIn Ads experts.
12:53
If the performance of your LinkedIn Ads is important to you
B2Linked is the agency you'll want to work with. We manage LinkedIn
largest accounts, and are the only media buying agency to become
official LinkedIn partners. And of course, performance to your
goals is our only priority. Fill out the contact form on any page
of B2linked.com to get in touch, and we'd absolutely love to help
you demolish your goals.
13:17
Alright, let's jump into where the rubber actually hits the road
with these tighter audiences. So we've talked about audiences that
are too large, and some that are too small. Let's talk about the
Goldilocks zone of those that are just right. And if you remember,
last episode, we talked about the Goldilocks zone with spending
enough on ad creative to get statistically significant data. The
same principle applies here with the audiences as well. We're
constantly looking for the right mix between learnings and spend.
We want audiences that are small enough that you can actually still
learn something valuable about what makes up that audience. And we
also want them large enough that you can actually spend enough
budget to tell and get any of these learnings. So let's say you
have these large audiences and you want to break them up so you can
learn more. So how should you break them up? Well, I want you to
break them up by whatever you want to learn about my go to is to
break them up by seniority, like that example I use with marketers,
I broke them up manager, director, VP, and C level. And this is
really good because it helps with content creation. If we're
performing really poorly with the C suite, like with CMOS in this
case, we can then go to the client and say, Hey, your content is
not resonating with CMOS, you should change the content. Here's
what they seem to like. That's data we can pass. You can also break
up by geography, because we know that different cultures are going
to respond differently. So for instance, I wouldn't want the US and
Europe in the same campaign if I could help it. Of course, if
campaigns are, if these two audiences are so small, and I have to
combine them in order for them to run then sure, I'll do it. But I
know that the cultures are different. And so I would much rather
keep them separate and learn about those cultures. I really like to
break them up by company sizes as well. Even if you know that any
one company size 500 and above is a good fit for your product, even
breaking them up by let's say 500 to 1000, and then 1000 to 5000.
And then 5000 and above can be a great way of learning "do we work
very well in mid market in enterprise in giant companies, even
fortune 500" can be a cool way to do this. I also like to break up
audiences by job function. Like that example I shared before with
finance and operations in the same campaign, I would always want to
keep them separate because they're going to care about vastly
different things. In short, cut your audiences by whatever
meaningful segments that you want to learn about. And these tend to
be very broad things. I wouldn't cut my audience by let's say
something like job titles or skills or other things that might be
too narrow. I want these to be very broad categories. And then if
you have a large audience size, don't just cut it for the sake of
cutting it. Like, for instance, saying, hey, this audience is twice
as large as it should be, I'm going to create two different
segments, one targeting just male and one targeting just female. If
that insight about how men and how women interact with your content
and offers wouldn't actually be a meaningful insight to you, then
don't do it. I've seen people break things up by interest, or by a
college degree, or something else that they didn't really care
about. They were just going for an arbitrary audience size. Yeah,
don't do that. It's just not worth the effort. Okay, so here's the
gold here, how do you make use of audiences that are micro
segmented? Well, this is the B2Linked strategy. This is where I'm
about to drop some real value bombs on you. If I take an audience
of senior marketers. And I break them up into their levels of
seniority and timezone. So we're talking some altra tiny audiences
here. If I do that each single campaign may be getting results
really slowly, it might be spending $5, $10, $15, $20 a day or
something like that. If I'm looking at these individual campaigns,
I'm probably going to get impatient. Or I might not be able to make
optimizations because these data sample sizes are too small. Ah,
but here's what I can do. I can start rolling these campaigns,
their performance together with other like campaigns and start to
learn things about it. With a single pivot table in Excel. I now
have enough data for understanding performance by timezone and I
also have an understanding of performance by level of seniority. So
each individual audience Yeah, it's too small to care about, but
when I roll them up to all other campaigns with a similar trait,
and I compare These things in aggregate, now I actually get some
really valuable information and insights. We recently did this with
a client where we've got about 200 campaigns in this account. And
they are defined by targeting type, seniority, geography, and
company size. Now, granted, they're a large spender, but I was able
to build a model that shows what our cost per sales qualified lead
is by each type of targeting by each level of seniority, by their
geography, by company size. Do you know how valuable that is? As
the account scales up, we know exactly which campaigns to bid up
and get aggressive with, to maintain extreme account efficiency as
we go up. I'm sure many of you know that you really get a choice
between scale and efficiency. If you're going to scale up, you're
going to lose efficiency along the way. Well, we've got a great way
and how we don't actually lose efficiency as we scale and that is
because As of this, this microsegmentation, giving us tight
controls over every little aspect of an account. And conversely, if
the client came to us and said, we have to cut budget, we know
exactly which campaigns to cut out first. Maybe they're okay.
They're not total losers. But if I have to cut budget, I'm going to
cut the worst performers first, for sure. And that means as we
scale back, we actually get also more efficient. So many of you may
be listening and going, Oh, man, it feels like a lot of work to
manage an account with 200 campaigns. Is it worth it? Absolutely it
is. If I managing an account with 200 campaigns, versus what it
might be at maybe 20, otherwise. The level of control that I have
over efficiency, as well as the learnings and insights I can get
about every variation of their target audience. I absolutely
guarantee that that is worth the extra time you have to spend
creating ads, and I also guarantee your competitors aren't going
through this much trouble. for managing the account, you will get a
leg up on them. Okay, so should you force audience sizes? No. If
the audience is larger than my 20,000 to 80,000 recommendation, but
there just isn't an intelligent way to slice it. Leave it alone.
Don't just aim for arbitrary audience sizes. There are some
benefits to larger audience sizes, like you can actually get lower
cost per click sometimes, because you have a higher probability of
being able to bid the absolute minimum or lower and still spend
your budget. And conversely, if your audience is smaller than our
recommendation, should you add irrelevant people to the audience to
make it larger just to fit this mold? No, absolutely not. LinkedIn
is way too expensive of a channel to ever spend a dime on someone
who isn't your ideal, perfect audience. What about if it's too
small? Should you just not run the campaign? Is it not worth your
time? Well, ultimately, I'm going to leave that up to you. I still
think it's valuable. If you have an audience size of 300 people who
are Perfect for what you do. Yeah, sure that campaign may never
really spend a lot of money. But boy, any money that that does
spend, or any lead that it generates will be well well worth your
time. So ultimately your audience sizes will have to be what makes
sense for you. Don't let someone else tell you what size your
campaigns need to be, especially if what you're doing is working.
But hopefully that at least provides a guide for you. Okay, I've
got episode resources coming up for you right after the break. So
stick around.
21:35
Thank you for listening to the LinkedIn Ads Show. Hungry for more?
AJ Wilcox, take it away.
21:43
Okay, here in the show notes, check out the LinkedIn recommended
audience sizes in their help section I've linked right to it. You
can at least see what they recommend and why. And in the new
section I told you about Microsoft's quarterly update report. The
links there, go ahead and check that out. See what Microsoft says.
about LinkedIn performance. Also, I mentioned Mark Williams, the
LinkedInformed podcast, there's a link there to his Episode 287,
where you can hear his commentary on it as well. If you're just
getting started in LinkedIn Ads, check out the course that I did
with LinkedIn Learning. It's only $25 if you're not a premium
subscriber to LinkedIn, and free if you are, and it covers the same
stuff that in a one on one training, I would be charging $500 an
hour for. This could be also really good. If you're trying to train
coworkers or new recruits in on LinkedIn Ads to help support you
then check out this course it's fantastic. Whatever podcast player
you're on, look down right now and make sure that you are
subscribed. And then please review the podcast I would absolutely
love to share your review with everyone and give you a shout out.
Any ideas for the show any thing you'd like us to cover, reach out
at Podcast@B2Linked.com. And with that being said, we'll see you
back here next week. Cheering you on in your LinkedIn Ads
initiatives.