Ashley: You take it from like, you have
a $50 bid for a CPA for a start trial,
um, and then you increase it to 70.
Then you think, oh, I'm
gonna get more volume.
It's almost the opposite with a ROAS
bid, if you take it from 50% roas
and you bump it up to 70%, you're
putting more strain on Google to try
and find those users, and there's
fewer users that are going to come in.
And, be able to perform that way.
Jacob: My name is Jacob Rushin,
CEO, and founder of Botsy.
Today on the Price Power Podcast,
we're talking with Ashley Black.
Ashley is a Google app
campaign's rockstar.
She was head of app ad sales at Google
where she shaped Google's go-to market iOS
roadmap and worked with thousands of apps.
She's now the founder of Candid
Consulting, where she's bringing all
this knowledge to the world to deliver
high performing marketing strategies.
I've worked together with
Ashley and a client, so I've
seen her expertise firsthand.
Today we're talking about how optimize
app campaigns for value-based bidding.
Ever think, why am I just sending
trial starts as conversion event
when those aren't real revenue?
Ashley teaches us about another
option called Target ROAS campaigns,
as well as other interesting
stuff from the Google app world.
If you want to get into the nitty
gritty of Google App campaigns,
this podcast is for you.
Okay, so when we're running different
types of, , campaigns with Google, uh,
we, we've got, you know, different app
campaigns and for our subscription app,
if we're going from an app campaign
to a target roas, target, return on
ad spend campaign, uh, what are common
mistakes you see when, when apps
are switching from these campaign?
Ashley: I would say the most common
mistake I see from people, just to
clarify, you have three different
campaign types that you can run.
Um, for, if you're promoting
your app, so you have just pure
install campaigns, so like a CPI
campaign, you have CPA campaigns.
Those are cost per actions.
This is if you're like bidding toward a
start trial or a signup or a registration
right there, there's a specific action
that somebody's taking, and then you're
setting a bid for that conversion event.
And then there's T Row as campaign
campaigns, which is telling Google.
Hey, we wanna see, you know, X
percentage of our ad spent back within.
This certain amount of time.
Um, so I would say the majority of
people, the majority of non-gaming apps
are still spending on the TCPA range.
TCPI is sort of like just for pure
volume and if you're, you know, uh,
uh, I don't see it as often in this
subscription space, let's put it that way.
Um, but I think the big, the big
challenge that I see in people trying
to make that shift over is one.
They are setting their
bid, their ROAS bid.
Way too high.
Like, of course you want a hundred percent
return on ad spend by like day one.
Like, yeah, that would be great.
But the likelihood of the, the algorithm
being able to find users at scale,
um, with a, a bid that aggressive
is just like, it's not gonna happen.
So I see people just coming in it with way
too high of expectations, thinking that
it's some sort of like miracle product
and it, that's not what it's gonna be.
Um, and then the other thing that
I see people doing is having really
long conversion windows and having
a really long conversion window.
So saying like, I want my, my, you
know, let's just say 50% of my ad
spend back within, let's call it like.
30 day, 30 days, 90 days.
That's just too long of a
window for Google to really be
able to optimize effectively.
And so I usually say, Hey, start with a
shorter window, seven days, for example,
and then decrease your ROAS target.
You know, so like, yeah, maybe you
want 150% back by day 90, that's fine.
That that's probably fair.
But you're gonna need to sort of like
do the backend math to figure out,
okay, well what does somebody need to,
what sort of ROAS do we want at day
seven that's going to get us that day
91 50% return, if that makes sense.
Jacob: it.
Yeah.
So less, you know, we have these,
you know, uh, high ROAS targets, but
more aligning kind of the ROAS at
the point of the conversion event.
So Google can effectively optimize that.
That makes a lot of sense.
That,
Ashley: Yeah, exactly.
Jacob: and, uh, do you think
like this value-based bidding is.
Like should, should everybody, should all
subscription apps just adopt this today?
Is it a good fit for all apps?
Is there a certain
level of scale you need?
Like what are signals that, uh, you should
be thinking about this type of bidding?
Ashley: So it's not for everybody.
Um, I think we, you know, we've talked
about this a a little bit or in the, um,
the article that I think we'll sort of
follow with this goes into detail about
it, which is if you are risk averse.
Probably not for you.
Like there's a steep, um, learning window.
I would say, you know, two weeks at a
bare minimum, usually a little bit longer.
So that's one.
Um, if you're not, if, if you don't
financially feel comfortable losing
some money to figure this out,
then I probably wouldn't recommend
it for you right out the gate.
Um, the other thing that
you need to have is.
Just like a high enough volume of
conversion events coming through.
So if you're only generating, let's
use start trials as an example.
If you're only generating like five start
trials per day, that's just not enough
data for Google to really be able to say,
okay, well then from that start trial,
how many of these are going to convert?
Let's say it's only like 50%, so
two to three, you know, users, and
it, it, it's just not enough data.
So if you have limited volume, um.
Then again, it's probably
not a good fit for you.
Jacob: And,
Ashley: Um,
Jacob: the,
Ashley: yeah.
Jacob: volume thresholds might even be
higher than a normal campaign where, you
know, maybe you need maybe five or 10
conversion events is enough, but, 'cause
we actually need the revenue values coming
Ashley: Right.
Jacob: and so it's even lower.
Like normally our campaigns are optimized
off trial start, but we actually need
the trial conversion data to get the
real revenue data to pass back to
Ashley: That's right.
Jacob: lower.
So
Ashley: I.
Jacob: need higher volumes
than uh uh, normally.
Ashley: exactly.
So like, I think Google's recommendation
is for, um, post-install events
to be generating about 10 per day.
And so let, let's say
that's 10 start trials.
I, I would hope.
Prefer it to be much higher than that,
but that's sort of their guideline.
So 10, um, start trials per day, but
if only, like we said, 50% of those
trials are converting into actual paid
subscriptions, then you're probably
gonna need at a minimum 20 start trials.
And so that we're getting the, the
system's, getting enough data back on
those revenue conversions to be able to
make optimizations and say, okay, this
was the right pocket, um, of inventory for
us to go after, or the type of user that
we need to go after, just more signal.
So low volume, this
probably isn't a great.
Jacob: Gotcha.
Gotcha, gotcha.
So, and, and.
I don't know if this, this makes sense,
but if you have a higher priced app or
higher price subscription, is that better?
Or, or does that not really matter?
As long as it's, it's more the, the
quantity of revenue events versus
like the, the amount of revenue.
I,
Ashley: Yeah, I would say it.
I'm trying to think through a
little bit like I think you, you.
Want to have just more
volume of those transactions.
Yeah.
Um, so the higher ones, obviously,
like if you're running $130, you know,
subscription, there's only so many
people that can afford that too, or
want to pay for something like that.
And so already your pool of users
is, you know, dwindled as opposed to
like something that's 6 99 a month.
Right now you've opened it up to
a lot more households that can,
you know, afford your product.
Jacob: Got it.
Got it.
So uh, thinking about this, so
normal campaigns we recommend like 10
events, 10 conversion events per day.
Uh, we still want those same 10 conversion
events, but now they just need to be.
events.
So we think of them as trial conversion
or other type of in-app purchases
that we can send back to Google.
Okay.
Ashley: Yes, yes.
Jacob: like if we recommend, um, sending
those back, those events back as soon
as possible, uh, and let's say, okay, I
have a seven day trial, that makes sense.
But what if I have a 30 day trial?
Is that not a good, is this
campaign just not a good fit for me?
Ashley: Probably not it.
That's just, that's, and that's
kind of where I go into like the,
how risk averse are you, because.
What we're talking about is like Google
is not, has no idea it could spend,
let's just say it's spending a thousand
dollars a day and after it has spent,
let's give your 30 day trial example.
After it spent $30,000, it
doesn't get its first signal.
That like a purchase
actually occurred, right?
So like that's your call if you
wanna spend that much money without
knowing, um, without the system
knowing, hey, how are these users
ultimately going to convert for us?
And so, um, yeah, opt, you know, again,
if you have a really long trial window,
probably not the best product for you.
Jacob: Got it.
Got it.
Okay.
Ashley: Yeah.
Jacob: let's say,
So we, we we're, we're probably want to
be at least a medium sized advertiser,
you know, have figured out probably
performance marketing a little bit.
We understand what
conversion rates look like.
We're, we're comfortable, uh, and have
a decent sized budget we can pass back.
Um, conversion events or, or
revenue events, necessary time.
Uh, like, so let, let's figure
out how, how do we actually.
you mentioned kind of having
a, a ROAS target below what
you actually wanna achieve.
Ashley: Right.
Jacob: I remember seeing in your article
that maybe like a good guideline is
starting 20% below your, your target roas.
Is, is that
Ashley: Yeah, sometimes
you gotta go lower.
Jacob: okay.
Ashley: Um, so you kind of have
to figure out, I mean, you'll
figure this out pretty quickly.
You're gonna put a target in, and so when
you're, you're in Google Ads, you're gonna
select the, um, conversion event that
you wanna optimize for, which for most
subscriptions, uh, subscription apps,
that's gonna be, um, it's a fire base app.
By the way, you have to be doing
this on Firebase in order to use
the T Row as product, but um, you're
going to put in the app store.
Um, subscription convert event.
That's, that's what they
get directly back from play.
And, um, and by the way, we're
mostly talking about Android here.
They do have an iOS offering.
Um, we can get into that later.
But, um, so that's directly back from
play on when that trial ultimately
becomes a, a, a paying user.
And again, in Google ads, you set up
the conversion window for that event.
So you can go in and set it
to seven days or 30 or 90.
The recommendation here would be, you
know, seven or less, um, if you can.
And, um, and so now I'm forgetting
where we were going with this one.
Jacob: Uh, yeah, just how do you
Ashley: Uh
Jacob: that ROAS target, like, so you
Ashley: oh.
Right, right, right.
Jacob: uh,
Ashley: So.
Jacob: yeah.
Ashley: Exactly.
So you, once you put in the
actual event, then you're gonna
have to put in your ROAS bid.
And, um, this is, I always tell people
it's like the reverse of a CPA bid, right?
So like with your CPA bid.
You go high and you would
expect more volume, right?
You take it from like, you have a
$50 bid for a CPA for a start trial,
um, and then you increase it to 70.
Then you think, oh, I'm
gonna get more volume.
It's almost the opposite with a ROAS
bid, if you take it from 50% roas
and you bump it up to 70%, you're
putting more strain on Google to try
and find those users, and there's
fewer users that are going to come in.
And, you know, uh, be
able to perform that way.
And so you'll see your scale decrease
the higher your ROAS target, right?
You drop it down to, let's say.
You know, 30%.
Now the system's like, oh, well
we're not held to as high of a
standard to bring in these users.
We can go out and try
and find more of them.
Um, you know, and then how do you
even come up with that ROAS target?
I mean, that's on each individual team.
You guys gotta figure out what,
what ultimately makes you money
and, um, how long you're willing to
wait to get back full returns or how
quickly you want 'em back, et cetera.
But, um.
You know, Google's recommendation
is, yeah, if you know you want a
50%, uh, ROAS by day seven, then
you should actually go in 20% lower.
So starting at 40%, in my experience,
I've had to, in some cases go even lower,
lower, lower, lower to even get volume.
Um, and then once I do get volume, try and
scale it back up again to the true target.
But again, that that's just
a function of how aggressive
your target is from the start.
Jacob: And I imagine like how, uh, how
big your budget is as well, or if you
have a larger budget and get, yeah.
Okay.
And, and, and so what's
that ramp up period time?
Is that like a month?
Is that a week?
How, how should, how
would I think about that?
Ashley: Again, like, I'll
give you two examples.
I have one client who I've been
running this with and their, um, seven
day roas target was like 17, 18%.
It took us no time at all.
To get going, like the campaign was
starting to spend pretty quickly.
What's interesting is on that client,
we actually ran an um, A CPA campaign
and a T ROAS campaign side by side.
We did a geo split test on that.
So like to make things simple, it
wasn't like this easy, but like a 2025
states, US states in the CPA campaign,
the other 25 states in the ROAS
campaign, the CPA campaign, definitely.
Spent a lot more, um, versus
the T ROAS campaign, but the,
the performance was pretty good.
On the t OAS campaign, we didn't run
the, the test for that long, but the
17% target we got going right away.
Okay.
Switch it over to another client.
Their target was 65% by day seven.
Much, much more aggressive, right?
Like a lot higher.
It took us six weeks.
To of me just being like,
Jacob: wow.
Ashley: it a little
bit here, drop it here.
Like, you know, keep going,
keep going, keep going.
Add in more assets until we
finally figured out, okay,
now we have it at scale.
Once we did scale it, what was frustrating
is, uh, ultimately the, the bid at that
point was like 30%, and so it did not
perform very well until we were able
to kind of like slowly and gradually.
Build it up again.
Which again is why I'm saying
for those that are risk
averse, like it can be tricky.
Jacob: Yeah.
Yeah.
Got it.
But, but if it works out, you can
see higher, better performance
where there's, there's a
Ashley: Yeah.
Jacob: we're not just doing
this to, to uh, uh, you know,
inflict pain on ourselves.
We're doing it
Ashley: Right.
Jacob: we can see, see, uh, a
Ashley: I.
Jacob: performance in the long run.
Ashley: It does tend to perform better.
And what I find, you know, truthfully
is like somebody who like sets these
campaigns up and optimizes them.
I just find it so much.
Um.
Easier but also more accurate to be able
to opt, optimize the campaign for the end.
The true end goal, I always
found it really challenging.
I mean, even though I still do it,
it is challenging to be like, okay,
this is what our CPA target is.
But then you're having to look at
this like other metric after and
say, okay, well like, yeah, we hit
a $20 uh, CPA on a START trial.
That was our target, but how
did the ROAS perform after?
And so it was sort of like.
Sometimes could be misleading.
Whereas with the ROAS campaign, you
know, I think you have a bit more
direction as to like what the future's
going to hold, um, longer term.
Jacob: Got it.
Yeah, because, because, you know, you
can be bidding on trial starts, but
all of a sudden if your conversion
rate drops, um, you could still be
spending the same, but all of a sudden
you're knocking the same returns.
And so that
Ashley: Exactly.
Jacob: feedback is going back to Google
what we're, can more effectively optimize
and, and find the right audiences for you.
Um, that makes sense.
So start, uh, it sounds like.
Advice is start with a lower
target and, and work your way up.
That's probably a, a better route.
And, and uh, from, you mentioned doing
that geo test, it's not, can you run
both campaign bid strategies in parallel
or is that not like recommended?
Ashley: Can, um, again, Google's
gonna tell you no, that's not
the best way to go about it.
And there's always the conversation about
like, cannibalization of like, if you have
two campaigns that are targeting the same
audience and potentially getting at the
same inventory, are you hurting yourself
by like competing in the same auction?
Um.
I always find that that's
like really hard to prove.
So yeah, generally they,
um, they recommend some sort
of test to prove it out.
But I think the challenge with ROAS
campaigns, and I think anybody will
say this, you know, gaming, non-gaming
subscriptions is the scale aspect.
And so.
You know, it, it can be kind of a
challenge to be like, we're gonna go all
in on T roas because you are going to see
the overall volume in most cases decline.
Like yeah.
You know, the, the revenue might
be, I shouldn't say the revenue
might be higher, but like the
profitability might be better, but
overall, maybe revenue decreases.
So that, that's sort of something
I think case by case, you'll have
to figure out of like, what's the
best strategy for us if we run them.
You know, we just run 'em at
the same time simultaneously.
Or if you are able to generate
enough volume to just do roas.
Jacob: Interesting.
Got it, got it.
That makes sense.
And so maybe if there's a case where,
you know, we have decent scale as an
advertiser, but we're struggling with, um,
getting kind of the conversion metrics or,
or kind of return on ad spend We want from
A-A-A-C-P-A based bidding strategy that.
might want to try, uh, kind of the, the
T row as, uh, to, to help control costs.
Um, so maybe not right for everybody
and you might need a decent volume,
but, um, and, and harder to scale,
but probably more profitable.
I,
Ashley: I would say generally,
once you get it to that point,
Jacob: Yeah.
Yeah.
Got it.
Interesting.
Interesting.
Ashley: Mm-hmm.
Jacob: and, and so.
You know, thinking about, uh, the
different bid types, we we're not
gonna reach, different placements based
on the different bid types, right?
It, it's still all the same eligible
placements, no matter the the bid type.
Ashley: I think eligible
is the key word there.
Like technically, you know, all
placement types are available and
just so everybody's clear, um, they're
sort of like five, uh, placement types
when we're talking about Android.
So you have the, um, the play store.
Which is a huge one.
Like that's why Google performs
so well on Android, in my opinion.
I mean there's, you know, it's kind of
like having Apple search ads, right?
Um, you have the Play
Store, you have google.com,
so that's just straight up search.
You have, um, AdMob inventory,
you have YouTube inventory,
and then you also have, um.
I guess it depends on every
advertiser, but you have,
uh, mobile display inventory.
They call it MGDN.
So this is like display ads
that serve in a web, uh, or a
mobile browser, I should say.
Um, so those are your three primary
sources of traffic with, um.
With Google, but you may see on
ROAS versus your CPA campaign.
You know, you getting into like, maybe you
have way more search traffic on your ROAS
campaign than you do on your CPA campaign.
Um.
You may be.
Also something I pay attention
to is like, what's the CPM?
I tend to find that the CCP
M is a lot higher on ROAS
campaigns than on CPA campaigns.
Again, we're talking usually with
a CPM going up, you're talking
about, um, higher quality traffic.
Jacob: And I think that's a, a good
thing to think about where don't really
just want the cheapest install, right?
We don't want, oh, our CPM is going down.
We're getting cheap installs.
And then I hear this a lot, right?
Where you people are, people maybe
come to you and say, my install,
like, you know, CPI is so low, but
no one is converting in my app.
You
Ashley: Yeah.
Jacob: that?
Well, you're, you're
getting crap users, right?
The quality of your
Ashley: Mm-hmm.
Jacob: is poor, and you actually
don't want the lowest CPM possible.
And so I think that's
a good, good reminder.
Ashley: Right, exactly.
So yeah, that's just a, another
metric to pay attention to.
Um, I think, and, and so to, to your
question earlier is like, oh, or your
assumption maybe are we just, we're,
you know, the inventory's the same.
Yes, theoretically the inventory
is the same, but you may notice
that you're serving distribution
across that inventory is different.
And by the way, I I, you know, obviously
my, um, I worked at Google for a long
time and so I talk about like, you know,
all the different placements that you
see within the, the Google's front end.
What advertisers see
is search, uh, display.
YouTube and then search partners,
which by the way, I don't see
that much traffic going to search
partners I never have, but um.
Within search that includes play search.
So people like searching for an
app in the Play store, but it also
includes searches on google.com.
Um, and display includes ad mod
inventory, so your ads serving in
other apps, and it includes your ads
serving on a, um, mobile browser.
And then there's a variety
of YouTube placements, right?
You have shorts, you have instream, you
have, you can serve on the homepage.
Um, so it, it's not as like, I
know you think you're getting some
information there, but like, it, it
could be cut so much more granularly,
Jacob: And, and so you can't target by
those different placements, but can you
see the, the end placements in like a,
a performance report or do you not even
have insight into that after running a
Ashley: so.
Not to the same degree.
Google's gonna lump it.
Um, so they'll do, if you pulled
like a placement report, what it will
show is Google owned and operated.
Which includes all play traffic and all
YouTube traffic and all search traffic.
Okay?
So like all of that traffic is
just lumped into like one line item
and then basically what you can
see is um, and only impressions.
Which is really annoying.
Um, you can't see cost, you can't
see the number of conversions.
Um, but you can, you can see impression
level information at the, like if
an ad serves in another app or on
another site, so not Google owned.
Jacob: Got it.
Yeah.
Yeah.
And we, we can probably a whole nother
uh, uh, conversation on why they do that.
So I, I
Ashley: Yeah, sure.
Jacob: a pretty straightforward
answer why they do that, but we,
Ashley: Mm-hmm.
Jacob: get into that as much as, well,
it's more, more favorable for Google.
Um, but,
Ashley: Yes.
Jacob: that's okay.
Um, so, and so like what, when
we think about targeting, are
there, is there any targeting?
We do.
Ashley: Not a whole lot to be honest.
Um, you know, you can add in other layers.
You can also add in other exclusions,
which I think people should, I think
people overlook a lot of times,
especially newer advertisers, I think
make assumptions about, oh, well
of course, why would you show an ad
to somebody who already has my app?
Uh, you have to tell Google that
you don't want your ads to serve to
people who already have your app.
So, um, there's exclusion
lists that you can apply to it.
For example, like if you don't want
ads to show to people who have, uh,
made a purchase in like the last, I
don't know, 365 days or something like
that, you create that audience and
then you upload it for your campaign.
Um, you can, uh, remove, like if you
don't wanna serve on your brand terms.
That's another, like, again, more
exclusions than inclusions, right?
So like you can say, don't
serve my ad on my brand terms.
Um, you can say exclude
these specific sites.
Usually people don't do that until
they see that there was, like, they
saw a drop in quality or something
like that, and then they could see
that there was a lot of impression
serving on a, in a certain app or site.
Um, and then they go in and add it.
But, um, a lot more exclusions
than creating audiences that
you think would be a fit.
You can tell Google like some signals,
like, Hey, target, um, people who are
interested in education, for example, but
it's not going to do that exclusively.
Again, it's just like a signal that
it's passing back to the campaign.
Jacob: Gotcha.
Gotcha.
And
Ashley: Yeah.
Jacob: it, it used to be,
maybe I'm making this up.
Can you actually add, uh, keyword
exclusions directly in the dashboard
yourself, or do you have to get like
an account manager to do that for you?
I, I remember there was somewhere
you had to like request it and you
couldn't actually do it manually.
Ashley: Yeah.
Now I think that you can
do it directly in the ui,
Jacob: For the, for
Ashley: so it kind of sits in a.
Yeah, it sits in a sort of a weird place.
It's like buried in like an admin section.
It's not just like a standard
setting in your campaign.
Um, but it, it now exists sort of
like in the front end to advertisers,
Jacob: Do you think that's like.
That's the biggest starting point
for exclusions of just making, of
excluding brand keywords where,
where that's a, an easy place to,
to cannibalize or, or just, uh,
Ashley: right.
Jacob: a, a false impression
of your, of your performance.
Ashley: That's, that's
what I would recommend.
You have to be careful.
I did it for one customer who didn't,
um, who didn't have it, and they just
saw a complete, you know, it was like
a 60% decline in their installs and.
Um, they sort of had to make the
call of like, well, do we want that?
Jacob: Yeah.
Ashley: you know, it's like, yeah,
it, it certainly resonates that like
we're we're paying for people who maybe
would've, um, installed organically, but
actually at the same time, they didn't
see their organics increase by 60%.
It was a much more modest
increase, and so they sort of had
to make the call to keep it on.
But I would say most people, um.
Sort of our, our want to limit
the amount of brand exposure
they get because you know, it's
taking away from their organics.
Jacob: Yep.
Yep, that makes sense.
Ashley: Yeah.
Jacob: kind of gets into the, you
know, world incrementality and, um,
Ashley: Yeah.
Jacob: there's lots of potential
ways to try to measure the
incrementality of branded keywords.
And same with like going back to, you
know, app store search and uh, uh, app
store ads and Apple search ads of, you
know, bidding on your brand and keyword.
Is it making a difference?
Okay.
Is it.
know, okay, you're getting a
X percent incremental lift.
Is that worth the cost?
And, and, and so it's, there's, yeah,
there's not really a right answer.
Right.
It kind of, it varies by product.
Ashley: Exactly.
Jacob: Yeah.
Gotcha.
Gotcha.
And, and any other, uh, exclusion rules
that you go, Hey, let's just start with
these, or, or you kind of default to,
Ashley: It's primarily the
users who already have your app.
So, um, that's the sort
of the biggest one that.
I just feel like, yeah, it's kind
of frustrating that like it doesn't,
it's not done automatically.
Um, but, and then the other one, you
know, to each his own, you, you have
to figure out what's best for you
is people who, if you wanna exclude
people who have reins or reinstalling
Jacob: Hmm.
Ashley: like if, um, you see value
in paying for those users or if
you don't, um, and again, like
that's something that you would
need to tell Google, like, do not.
Target users who are now coming
back and you know, they uninstalled
and now they're reinstalling.
Um, so like I said, some, some people see
value in those users and paying for them.
Um, but I think a lot of people
are like, would they have
just done this on their own?
Jacob: Right.
And, and that's
Ashley: Yeah,
Jacob: possible on Android
specifically, where on iOS you wouldn't
have that same level of insight.
Is that right?
Ashley: yeah, yeah,
Jacob: So,
Ashley: exactly.
Jacob: talk about that a little bit like.
I, does it, is this a strategy that
makes sense for iOS, uh, campaigns?
How do I, how do you think about
Android, uh, app campaigns versus
iOS, iOS app campaigns generally?
Ashley: Oh man, that's like
such a, a big meaty topic.
Um, I spent a lot of time and I, I have
talked about it a little bit, sort of on
my own platform, but, um, you know, not
talking about the ROAS campaign type,
just, uh, let's sort of leave that out.
That's relatively new for Google to
be, have a ROAS campaign type on iOS.
Um, Google already has some limitations,
which I'll get into on iOS, so.
Uh, you know, I know that they're, they're
pushing it a little bit, but I, I, for
me, right now, I sort of am waiting to
hear more about like, how it's performing.
I don't necessarily wanna be one of
the Guinea pigs because they had a
lot of improvements to do on their,
um, like earlier products, even on
their CPI and CPA campaigns on iOS.
Um.
So, as I mentioned, like the big,
big advantage that Google has on
Android is that it has the Play
store and that it owns Android and
it gets a significant amount of data.
So, um, you know, if you are running,
let's just put it this way, like if
you have an Android app and you're not
running it with Google, like you're doing
yourself a disservice, I would say that.
So you should do that.
If you have an iOS app and
you're not running with Google.
I get it.
The product has not
performed all that well.
Um.
And that's because of
a, a variety of reasons.
I think one of the big ones that
people don't realize is that, you
know, a lot of Google's inventory
still on iOS still comes from web.
Um, and so, you know, like if you
think about it, just like google.com
and so, um, there's, there's a little
bit of like a, a disconnect in being
able to do tracking there, the web
to app stuff, but also another huge.
Factor.
Is that the only place where
Google prompts for a TT is YouTube?
And they only started
doing that like a year ago.
Maybe even less.
Like I didn't actually see the prompt
until I think December of last year.
So.
There, there, they weren't
getting idfa for a very long time.
This is all to say they're
making a lot of improvements.
Um, from what I hear, there's supposed
to be some substantial improvements
to their, um, to the, the product
itself, the iOS product itself.
And I think mostly this is just
that they're gonna start using IP
addresses to, uh, inform their models.
And that should be coming in.
Um.
Next quarter, so like hopefully
October or November time.
So I'm really excited to test it out
for a, a few, I have a few, um, like
clients who are gonna start with me
in October to start testing out iOS
Pro, their iOS Google campaigns.
But up until this point, I've seen, I've
had to turn it off for more of my clients
because the performance just wasn't there.
Jacob: Gotcha.
Gotcha.
And, and, yeah.
So, uh.
Google was Yeah.
Late to adopt scan, uh, and still what
all conversions are like modeled, right?
They're, they're, uh,
Ashley: Mm-hmm.
Jacob: on kind of the, the,
the limited data they do have.
And so now I guess on YouTube.
that's worth trying again for some
people now that there might be better,
uh, uh, conversion tracking there.
But yeah.
So it sounds like probably, uh,
t ROAS doesn't make sense on iOS
and then quite yet, but, but maybe
you try out, uh, general Google
app campaigns and, and see how it
Ashley: Yeah.
Jacob: is, is that probably a, a later
term strategy or, or kind of a scaling
strategy versus starting with that on iOS?
Ashley: Well, again, I think I, I
hate using the term, it depends.
Um.
I, but I mean, what I do observe most of
the time is that people are, you know,
running pretty heavily with meta for iOS
and, and obviously they have, um, apple
ads too, and then they're like, Hey, are
we missing out on an audience on Google?
And then that's when they come
over and, um, I think that's fair.
In some respect, just based on my
experience running the campaigns and
seeing people run the iOS campaigns
for a long time, I think that's fair.
So, um, but I do think that if you
feel like you're reading, reaching
a point of saturation on meta yes.
Like Google, Google technically does
have the largest network on iOS.
Like YouTube is still.
Like first or second most
downloaded app on iOS devices.
And so you can't really overlook it.
Um, but you, you might have
to be sort of patient with it.
And I will say on iOS, you do have
to have a pretty strong, um, creative
strategy because most of the time,
again, from my experience, you get a
lot of traffic flowing through YouTube,
um, YouTube shorts in particular.
So, um, wanna make sure that you have
those creatives like really dialed in.
If you are gonna launch
with Google and iOS.
Jacob: Got it.
And that's like good
videos essentially, right?
You, you have
Ashley: Yeah.
Jacob: yeah.
And, and so if
Ashley: Yeah.
Jacob: the, um, production capabilities
to kind of create those videos
or confident ability to do that,
that's probably a starting point.
And to be honest, it's so much easier
than it used to be to create amazing
Ashley: Right.
Jacob: So may maybe it's less
of a barrier than it used to be.
Ashley: And we're talking mostly like UGC
stuff tends to work really well there.
So again, like it's, it's mostly
shorts, inventory, and, you know, so
you can think about that as sort of
like reels or the TikTok comparison.
And so those sort of like authentic
user type videos I've seen perform
pretty well on, on YouTube.
Jacob: Got it.
So maybe if you're, if you're
running those on another channel and
those work well for you, it could
Ashley: Mm-hmm.
Jacob: that.
These could, could make sense to try.
Ashley: Yeah, absolutely.
Jacob: Interesting, interesting.
Um, something, something else I, I
saw you writing about, uh, recently
was that if you're, if you're not
getting, good enough signal with maybe
gist, uh, trial conversion events,
Ashley: Yeah.
Jacob: you should look deeper at,
uh, retention events to further
optimize those campaigns and, and
getting those stronger signals.
I thought that was like, made,
made so much sense and I think, um.
Um, you know, I, I work a lot in kinda
the retention monetization world,
so, you know, I, I do similar kind of
analysis and correlation analysis to
see like, you know, what, uh, early,
um, early engagement events correlate
with long-term conversion retention.
And so, you know, when is that, or I
guess, talk me through like how you
think about doing that and, and then
when that's a good strategy for apps.
Ashley: I think honestly, you should
be thinking about it all the time.
Um, I think too often people are like,
sort of have this mindset on like,
okay, well what's that one event that
we're going to track and measure?
And, um, you have to kind of think
about it a little bit more dynamically.
And so, you know, the example that
you're referencing or what I had
talked about is like, I had, um.
Had one client who was optimizing
for just registration complete.
So they just wanted to see who was going
through their whole registration flow.
And if they completed that whole
flow, then like that's what they
wanted to optimize for, but.
Ultimately, like those users weren't
sticking around for that long.
Like the retention on those users wasn't
great, and retention was a big play for
them because for this particular app, they
really cared about monthly active users.
That was sort of their, their main,
um, KPI that they were looking at.
So we did a bunch of analysis
to try and find, well, what was
another event that was happening
with a great frequency, right?
Because like that's one thing like,
yeah, sure everybody would love to
optimize for like something super
deep funnel, but then you can't
optimize based off of five users.
So like we need something that has volume
and for them they're a messaging app.
What they ended up, what we ended
up finding is that, oh, actually
it's not so much the registration
complete, but if they actually start
a conversation with somebody, so
they start messaging somebody, that
is when we see the stickiness, the
retention improve a lot, and again.
It at volume, right?
There were other things that
showed greater stickiness, but
like the number of users was too
low, so we couldn't use those.
So that's how we ultimately ended up
deciding to pivot for them and say,
we're gonna shift away from just this
registration complete filling out a bunch
of, you know, questionnaire questions,
um, and, and move toward them actually
starting to take some action in the app.
Um, and so I think if you can think
through, you know, what are those early.
Early additional, um,
actions that a user takes.
I think, you know, you and I are
familiar with one particular app,
right, that we've, um, talked about
and, and worked on, and one thing that
we're looking at is like, okay, what
if we stop just optimizing for a start
trial and instead like the, the action
that they take immediately after.
So did they watch, like,
let's call it the tutorial?
Did they complete the tutorial after the
start trial or did they, you know, um,
I don't wanna give it away, but like,
you know, did they do something else
immediately following something that
probably happens within like the next
five minutes after starting the trial?
And is there enough volume in
that for us to be able to then.
Optimize for that action
instead of just start trial.
So, you know, it's gonna vary
depending on your, um, your
app and like how it functions.
But thinking beyond just, you know,
what is sort of required of the user.
Jacob: Yeah, this reminds me of
like activation analysis to figure
out, you know, general, like what
is our activation moment in the
app and we're trying to figure out
what leads to long-term retention.
And so, you know, I, I do that.
By, you know, we we're trying to
figure out the right path for a user.
How, what is this activation
moment that we can kind of guide
users to through product changes
or growth changes or experiments.
And it sounds, you know, pretty similar
to kind of that process of looking
at a number of events and trying
to figure out what correlates with
long-term retention or, or conversion.
Um, and so is that like.
generally how you'd recommend
someone trying to go find this event?
Just looking at kind
Ashley: Yeah,
Jacob: between a bunch of events
and seeing what, uh, uh, ha
has, you know, potential for
like a predictive uh, uh, value.
Ashley: yeah, for sure.
Um, hopefully, you know, you have a
team or you're capable of doing it
yourself to be able to understand.
And I, and I think honestly it.
A lot of times it's, um, it's
probably a little bit obvious if,
if there's like a clear flow, right?
Like, let's use the messaging app.
It's like, okay, did they, if they started
a conversation, it means that they had
somebody else to talk to on the app.
And so that right there shows that
there's like more stickiness, right?
Like they're, they're, they're joining
or they're on the app so that they
can communicate with somebody else
as opposed to, let's just say like.
Watching a video in it, like that's not
as meaningful probably to you because
it's, that's not the intent of the app.
Um, they're not in that particular app.
They're not like serving
ads in videos or anything.
And so there's not as much
value in a user like that.
Um, so I don't know.
To me, I think that they, it kind of
stands out a little bit, but yeah,
you gotta do some of the analytics
to figure it out in this case.
Um, you know, like it did take us a
while to sort of like, understand.
You know what?
There were so many other events that,
like I said, showed high retention, but
we needed something that happened within
24 hours, really, really quick, so that
we could still send that signal back
to Google in a short period of time.
And like I said, there was still
a high volume of users so that
we were getting enough volume.
Jacob: Yeah.
So in reality
Ashley: Yeah.
Jacob: may think, oh, the what,
what event do I choose there?
There's so many, but in reality, the
actual set of events that, happen
that quickly, and you probably have
some intuition already around what
is, if you haven't done the analysis.
know, maybe there's probably only a
few that could fall in that category.
And then maybe just do some basic
correlation analysis to, to see which
one is, is potentially more predictive.
Um, okay, that makes sense.
Yeah.
Ashley: Yep, exactly.
Jacob: that,
Ashley: Okay.
Jacob: that's probably not something
you, that strategy is probably not
something you lean on immediately.
You may go to that strategy
if you don't see performance
through optimizing towards a
Ashley: Right.
Jacob: or the, the normal event.
Ashley: Right.
I think, you know, in the examples
that we used, um, the other one that
I'll call out, it's not as, you know,
relevant maybe for this audience
because it was, it's an, um, ad funded
app, but they, they had previously
been just optimizing for installs.
They were like, just
whatever, like cheap installs.
That's what we're looking for.
And we had to pivot a little bit
and we're like, well, if people are
sticking around, like they're not
uninstalling the app after, um, day one.
Like if they're still a user by day one.
One, it's still high volume
'cause we're just going after,
they're like utility apps.
So it's like basically anybody.
Um, and you know, we can, we can afford
to pay a little bit more for these users
and be more competitive in the auction
if we're optimizing for users who we
think are gonna stick around for longer.
So ultimately with that advertiser, we
totally shifted and, and now everything
bids on like a day one retention event.
Um.
So, yeah, I, I think you can start
out with just like your start
trial, if that's what you wanna do.
But if you're like, eh, these users
like aren't, you know, they're not
doing exactly what I want them to
do, we're seeing a big drop off, or,
you know, um, a lot of people are
canceling, then I think you have to
start to look at some, um, additional
parameters that you might wanna add on.
Jacob: That's interesting where it's
a bit counterintuitive where you're
actually, um, you're able to bid more
on this event, uh, and so potentially
reach larger scale by going further
down funnel because that event is, is
more predictive of long-term value.
Ashley: Yeah.
Yeah, exactly.
Jacob: Yeah.
Interesting.
Um,
Ashley: Mm-hmm.
Jacob: I, I wanna, I wanna
end on this question.
Uh, this has been awesome so far.
And so you, you mentioned the past having,
uh, hot takes, you, you weren't able
to share when you're working at Google.
Any, any, uh, uh, con contrarian views
about Google ads or app campaigns
that maybe most marketers or mobile
marketers get wrong, or, or you just
want to have to get off your chest?
Ashley: Okay.
I don't wanna, you know, I still
have, I'm not trying to burn any
bridges, let's put it that way.
You know, they were a very good
employer to me for almost a decade.
So, um, I'll have to be a little
bit reserved in some sense.
Um, but.
I think when you talk about the, the
like advertisers and sort of the way
that I see them approaching things like
Google certainly positions this product
in a lot of its products, by the way.
Keep in mind it's like UAC was the
first machine learning ads product, now
they've got Demand Gen and P max, right?
And like you're seeing all of these
other Google products start to
move into the same realm as UUAC.
Um.
But I see people being
like, well, it's automated.
Like it, why isn't it doing this?
Like if it's, if it knows that
it's not hitting my CPA, then why
is it still spending my money?
It's like Google's still a business.
Like I always joke, please, I hope
nobody takes me seriously here because
I don't know if this is the case, but
I'm like, you know, if you looked at the
algorithm like lineup, it would be like.
Hey Google first get customers good
performance, you know, and then
like all of these other things.
So like Google I think still
looks out for number one.
And, um, so it's all to say that you
still have to do work on these campaigns.
Yes, they're automated, but like you can't
trust that the system's going to always.
Be doing, you know exactly what
you think it should be doing.
It still requires you to go in, make
optimizations, make some tweaks.
And so please don't overlook that just
because it is an automated product.
Um, and yeah, I'll, I'll leave it
there that I'll leave my, I'll, I'll
sort of table the rest of my hot takes
maybe until I have a glass of wine.
Jacob: Okay.
Yeah.
Yeah.
We will get you on another
podcast after the glass of
Ashley: There we go.
Jacob: um,
Ashley: There we go.
Jacob: I think that's a good warning
for like all, all ad networks
do that you have to realize,
yes, they, they are a business.
Uh, they're out to make money.
They, it's this dual goal of
maximizing performance for you and
balancing performance for them.
And so, uh, I think that's.
You know, going back to what we
said earlier, some of the fine
tuned controls have been removed
where sometimes performance can be
hi hidden in, in blended, uh, uh,
Ashley: Yeah.
Jacob: and blended networks.
And so yeah, don't, uh, um, stay diligent,
uh, when running your, your campaigns.
Ashley: Yeah.
And there are, there are
things that you can do.
Like I, I think a lot of people are
like, well, there's only so much, right?
Like, and people tend to
think of it as like targeting.
I hear a lot of people comparing Google
to meta and the meta, just having so
many more parameters in which they allow
you to make tweaks to the campaigns.
Those, yes, it doesn't.
Exist to the same degree in, uh,
app campaigns with Google, but
there are things that you can do.
Um, and so like, yeah, don't, don't
assume that there's nothing there
to, you know, left to tinker with.
There are things that you can do and, um,
and if you're not sure then like follow
along for more stuff for me and maybe,
you know, uh, what working with Jacob too.
Um, hope to give some more tips.
Jacob: Yep.
Yep.
If you're struggling, to Ashley.
Um, yeah.
Do you want, do you wanna do
a quick shout out of, of, uh,
uh, your consulting business?
Anything else you want people
Ashley: Sure.
Yeah.
So the company, my company is Candid
Consulting, mostly work, um, as an
app agency, specifically supporting
clients on their Google strategy.
Um, and then do you know, general
consulting for apps in their space.
And, um, and so yeah, sort of.
All encompassing, but obviously with
a lot of expertise on the Google side.
Jacob: Yeah.
Awesome.
Um, yeah, we'll, we'll, we'll link out,
uh, uh, the blog post on on T Row as
campaigns and, uh, the podcast notes.
And if you ever have any
questions on, on Google Ads, just.
Just Google, Ashley Black, uh, Google Ads
for my subscription app, and there's so
many research, uh, uh, research out there
from you and blog posts and other content.
Uh, so, so, and then if you,
and if you still, you're still
stuck, reach out, uh, go find
Ashley: Yes, please.
Jacob: in touch.
Um, cool.
Well, this was
Ashley: Sounds good.
Jacob: I really appreciate you joining us.
Uh, so many great insights.
Ashley: Yes.
Great to be here.
Thanks so much.
Jacob: All right, thanks.
Talk to you later.
Bye.