Google Ads Data Demystified: Measure the Success of Your Campaigns Accurately

And that's what's really interesting about this, whole measurement thing, is

anytime that you know as a good marketer that you're not gonna track all the

conversions because of this campaign.

You have to look elsewhere.

You have to look at the overview of everything.

But the next thing I'm gonna show you is actually our account.

And if you look at solutions eight, I look like the worst as manager in the world.

So people that clicked on the ad to watch more that have navigated

back to my website are five.

There's probably about 25, so I'm, missing five x.

So talking about diagnosing supplementary campaigns and how they

actually have an effect the overall account structure and performance.

, not even account, just really , the company, , , the business.

Google does a good job at tracking its own conversions when they're based on clicks.

They do a poor job when it's based on influencing a market

to take an action that's not clicking on that ad that they saw.

And you've seen this everywhere.

You'll see this in YouTube, you'll see it in G S P, which

is now part of Performance Max.

But discover campaigns, display campaigns, and it's kind of uncanny

when you're looking at the differences between, , just the overall account.

, like let's say an analytics was tracking all of the conversions where it's tracking

all of the, conversions from direct and email and, , social and referral.

And when you look at all of that, , it's an m e r.

So as long as the m e r, which is a media efficiency ratio, which all

cash in, all cash out, which means how much did we spend on all advertising

and how much did we get back, that's truly the best way to look at an.

It's, something that is, paramount to the success of any business where it's okay

to have campaigns that are running at low row as, or only getting a one x return.

There's a company that I actually spoke to yesterday that I'm speaking

to next week, that developed an AI tool to give you a predictive LTB on

a specific type of client because.

Usually you're getting like a 0.8 or one x return, but they know that the

LTV of that specific type of client is gonna turn into a 2.2, and that's

what , they're safe with, but they don't wanna spend four months to identify

if that's actually gonna come true.

We have a bit of . . that.

. that in our, just baseline knowledge in our mind that yeah, if you run heavy

YouTube, YouTube's gonna look like it's terrible conversions, but actually

it's working really, really well.

And it all has to do with how you're measuring the campaign, , and how

Google is measuring the campaign.

Cuz those are gonna be two different things.

And that's what we're gonna be discussing today.

, this is a part, , that we'll have to actually.

, because it's, if this goes on YouTube, but it's a pretty important thing to identify.

There was a client that we launched a, let me just make sure, , but

this is actually, , one that Regina and I work on, , Regina mostly.

And then I've just been roped in recently for a Pmax test.

Now, this is the last 30 days of our Performance Max campaign and performance

Max did a poor job at, necessarily, , getting good amount of conversions.

But what we're not gonna see is the.

Amount of ancillary.

Marketing that's not attributed to clicks that can move a needle.

So , this campaign ran for, , it was about three weeks, spent two

grand, had 13 conversions, tracked and had $1,700 in conversion value.

But there was 25,000 people that saw.

And during the time that we turned it on compared to the previous period,

there was a 23% increase in the conversions, and direct channel went

up by 14% in, in terms of, , revenue.

The referral channel went up by 8%.

it did double the amount of sales.

Organic search state about the same, affiliate state.

About the same.

Now this campaign was shut off on the, , 14th or 13th.

Really?

So 14th, 13th.

So if we wanna look at, let's say this next week compared to the previous period,

the 15th versus the last week, we see that, I think I might have a day off.

Cause I think there was a conversion that might have been tracked.

So now we see after it's been shut off, traffic is down across the.

and the revenue is down across the board.

Now, the transactions actually did increase, but it's a different

potentially type of user.

I'm not sure.

I'm looking at only revenue in and out.

So you look at the direct that went up.

Now it's down by 13%, referrals down by 32%.

The organic search is down by 50%, and our paid search is down 82% Now, I'm not

a, a mathematician, but this is probably, if I'm tracking revenue, a bad idea to,

to shut it off and then look recently and saying, oh, this really isn't working.

It's technically shut off, I think on the 14th, , where that revenue here is down

31% and we see, fairly large decreases.

Now, the referral channel here, that went down by 38.

in terms of revenue and I'm just looking at before and after.

. That's all we can do with M mer.

All cash in.

All cash out.

The, , revenue still was on their own website, so it's not like people

googling things and coming in from other areas, which will sometimes happen.

, previous companies that theirs had pairing companies of theirs or even

people that are Googling things like free e-commerce, things like, this

company coupon code, which you'll sometimes see in referral, the more out.

YouTube you do.

A lot of times you'll see inbound searches for scam coupon code reviews.

You get people that are in their purchase decision and they're

starting to look around to say, am I actually gonna buy this?

I've never heard of this company.

It's the first time I'm seeing them.

Am I gonna immediately see a YouTube bag, click on it and buy something

and be like, my life is complete.

No, people don't operate that way.

We need to have trust.

We earned first, and then I will come back, however, which way

I may please and then convert.

So we're looking at the all cash in all cash out model.

The MER is truly the only thing that we can really tie.

, really time performance to.

Now, this is not necessarily just the first model, but the next thing I'm

gonna show you is actually our account.

And if you look at Solutions eight, I look like the worst as manager in the world.

, I would look really, really terrible if I, if I was me and I was looking at my

accountant, I didn't know any better.

I'd fire me as an agency first, solutions eight.

I wouldn't know Submarket anymore.

I'll share with you the reason why, , let me just pull this up here in analytics,

, and we'll be able to see a real.

effect.

Now, this is going to be much harder when you're talking

about omnichannel performance.

When I see by omnichannel performance, what I mean by that is

when you're looking at, everything in a silo, it's easy to identify.

We did one thing, we turned it on, we turned it off.

What we're going to have to get really good at is a good client

communication, but also we turn this on.

What happened to Facebook?

What happened to.

Instagram, what happened to TikTok Snapchat?

What did you email a bunch?

Did you run a sale?

Is it seasonality?

How is it compared to last year?

You'll need to rule out all the variables and get yourself into a

really good control group knowing that you can tie that back to the

the changes that you are making.

NorAm does this fairly well, , on a click basis, not so well on a view

basis just yet because it's having a little bit of trouble identifying

Pmax Lift and even YouTube, Lyft.

But for the most part it's fairly accurate when you're talking about the clicks.

And then you can look at the conversion path and verify that.

So I'll talk into about how it tracks and then we'll talk into some,

some real life scenarios about what we did and what we were watching.

So, , I'm gonna hop into the solutions eight, uh, analytics,

and just gimme one moment.

I'll share.

This is our YouTube, , campaign.

Now, in our YouTube campaign, it looks like I have a cost

per conversion of 3,739.

If this was anybody talking to a client, that would be a , very difficult

conversation to say, yeah, let's continue to spend $3,000 a day on this campaign.

Very difficult.

Now, we noticed that on July 18th we started our campaign.

We see the conversions.

They're fairly consistent.

, , there's some highs and lows and everything like that.

This is, , , good evidence, flow of normal, consistent results

as we test, add, copy, and open, creative, and all the good stuff.

Now, a $3,739 cost per conversion, again, I would fire, you're paying $3,800 a.

For a company that does Google ads on and we're running a

YouTube campaign, that's bad.

Now, when you look at the backend of analytics since the time that we

started, you can see that these are the goals that analytics are are showing.

We're not running a brand campaign.

The only thing we're running is . YouTube remarketing, display, remarketing.

That's it.

Our driver is simply just YouTube that's always spend our money on, and I

don't even send YouTube traffic to our.

I send YouTube traffic to our YouTube channel.

They come to our site on their own accord.

So when you're looking at action plans and when you're looking at contactless

pages going from 92 up to 183, that doesn't just necessarily happen.

Closed deals went from six to 11.

The action plans went from 132 to 174.

We're looking at some, the estimates requested 26 to 31.

This is how I'm marketing it alongside of our Slack channel.

Our Slack channel has a sales channel in it.

We know what's nice about having the sales channel in this is we can

actually identify the way that people are, , finding us based on what they're

filling out on the, Slack channel here.

Here's a lead here.

How'd you hear about YouTube?

This one's Google.

it's spoke here.

How'd you hear about?

, we got a couple more leads that I can't tell where they came from.

, so let's just see what it says.

How did you hear about us?

, so I'm sorry, we're getting some leads here.

How did you hear about us YouTube?

, this one, they heard about us on Google, but it could be organic.

, when you look at the sales channel and say, how did you hear about us YouTube

here, 10, 50,000, , how to hear of us?

Five, 10,000.

YouTube.

Obviously there's a theme here.

that person heard of us from a podcast.

Next one.

How did you hear about us?

We even see that the source is YouTube organic and they went to

the performance Max just stumbled across your YouTube video where

John talks about GM C suspensions.

Okay.

How'd you hear about YouTube?

Again, this is just in the last five days.

Just last five days, we see what like seven or eight

leads, , in the past five days.

Now, if we look instead of Google in the last, oops, let's

see, , last five days I've tracked.

Two conversions, and I haven't even gone back to the last five days.

I haven't even finished.

There's probably a couple more leads here.

So what's interesting is I'm seeing a lot less conversions inside of Google Ads than

I am inside of my analytics or even the Slack channel with proof of then on the

leads own accord, saying, sorry, YouTube channel, how to hear about YouTube.

YouTube isn't the first selection on our dropdown.

The first selection is Google, then it goes YouTube.

So they have to physically click the button, go down to

rows, and then click YouTube.

So I know this is.

But Google is not tracking well at all like this.

There's, and there's more conversions that are coming in that we had to skip like

probably six or seven conversions because there were different type of conversions

they filled out our contact us for.

I don't believe that people are just, you know, finding us randomly and, converting.

, and Mike, you're and Gabby, you're here right now.

Yes.

Yes.

I'm here . Okay.

Of the last week.

, sorry.

Yeah.

Off the top.

Yep, no worries.

Off the last week, just off the top of your head, how many leads came in and

how many could we identified YouTube.

Just, you know, and not holding your feet to the fire for an exact

number, but it's like ten two.

this week was kind of a low week, so not much of our clients, but almost all of

our live generally come through YouTube, which probably could be around like.

Even in a week, like 26, we can get 26 clients that come from YouTube.

But this week was particularly low.

. Yeah.

This week was definitely a lower week for sure.

, but we confirm verbally with them too, John.

And they're all YouTube and a lot of 'em referenced specific ads nowadays too, so.

Right.

So in the last 14 days, , was there 4.8 leads that came from YouTube?

About five leads in the last two weeks, much.

That's, , exactly what this topic's about.

We know that those campaigns, the people that clicked on the ad to

watch more that navigated back to my website are five leads.

There's probably about 25.

So I'm, missing five x now.

This is also Performance Max.

Performance Max has YouTube built in.

It has Discover built in.

It has G S P built in, it has display built in.

It has remarketing on those channels, also built in.

But then we measure performance Max on, well, did they click on the shopping

or does they click on the search?

And it's an inaccurate way to measure it.

, we already know that , the campaign, you can look in a

segment by, , Adam and typing.

You can see that those have other view with truda conversions, but

they're not captured all the way.

There's still a lot missing.

So when you're looking at supplementary campaigns, in order to

help change over the, performance and the overall me r there's a

few things that we could look at.

There's a few apps that we could.

and I, I wish I had my one app, but the one app that I had

that I was gonna present today, expired yesterday from a client.

, they kept on a free trial version and they haven't updated it yet,

so I can't go back the 90 days like I really, really wanted to.

But long story short, , and you're just gonna have to trust me on this one.

I ran a test for a client that was a feed only upgraded from Smart Shopping p m

Max campaign at $500 a day and a YouTube campaign that was running at $500 a day.

I tracked probably about one quarter of the conversions that actually had a lift.

, what this, and I'll, , started this test on the 27th and went to the 17th,

16th, so from the 27th through the 16th.

27th, , half on the 17th, so let's just call it 17.

, there was a 66% increase in total sales.

New customers went down, returning customer rate went down by 19% they

were at 83 when we first started this before even performance

bags before this test as well.

They're at 83.

We got that down to 60 with an increase of their online store

by 66% total order set up 40%.

Sales attributed to marketing.

38%.

Facebook's still at 3,800, but that didn't change.

There was a 1% change in sessions from Facebook.

That's what's really interesting about this is Facebook had a 1% change and

YouTube is a thing that started, we, we started with YouTube and pax, and there's

a large movement of this Neil here.

I don't have the buy the numbers app.

I wish I did.

, we had it expired.

I don't think that the gentleman, , has updated it yet, unfortunately,

so I can't go back to 90 days.

I go on back seven, so it's not really great.

, it's not really great, , test.

But he didn't change anything in, Facebook.

Right now, we're still in the last three days, same exact 0.5, like

, we're really running really well.

I stopped YouTube though on Monday.

So I stopped YouTube on Monday.

I let feed only P max go.

So we have the same amount of total sales, but that's cuz our

recharge subscriptions were good.

Facebook went up 783%.

The online store actually went down 11%.

My returning customer rate popped up 36%.

My sessions went down 20%.

My total orders went down 14%.

, the sales attributed marketing, cuz they also run email went up to 220.

But you can see YouTube had a 45% reduction this last week, and I

lost 40, 11%, , of my online store.

Everything else has been the same.

I shut off YouTube three days ago and we're starting to see this, this

downturn starting to happen when I was up 63% during this time period.

Google says, I'm, horrible.

Google says, this is terrible.

Like nothing's working in here.

Your YouTube conversions are costing you $800.

your PAX feed only campaign is nice said a hundred dollars cpa, so

that's running really, really well.

But your YouTube campaign's terrible.

YouTube is not working at all.

And that's what's really interesting about this, whole measurement thing,

is anytime that you know, as a good marketer that you're not gonna track

all the conversions because of this campaign, you have to look elsewhere.

You have to look at the overview of everything.

That's very, very difficult to do, especially when Pmax becomes

larger and larger and larger.

PM M Max doesn't get view attribution data in nor Beam.

It doesn't.

So if you look at Pmax and do click views in Nor Beam, it doesn't give

you any additional conversions, even though Google shows.

Yep.

We had 27 people last week that saw the ad monitor 10 seconds and

they converted without touching it.

Clicks and clicks and views on NorAm does not change.

So we can't rely on Norby for this data.

This has to come from good marketers and you have to know if you run.

Find where it's converting.

Are we getting more email signups, checking the analytics

goals regardless of source?

Are we getting more direct and organic and referral lifts like we were with

the other company that I shared with you at the first example, we saw everything

kind of moving up to our, spend and our revenue instead of the paid as

channel and analytics only went up by like 20%, but we, 50 and 60% elsewhere.

Client also ran a direct mail campaign at the same time.

but you don't get direct business from direct mail.

No one types that website.

You URL directly.

They usually do the Google search for a brand name.

That's just kind of a consistency.

So when we're looking at this here, it's very hard to sell a client on,

Hey, trust me that that high CPA that you're seeing in YouTube, that's fine.

You're, you're okay.

Don't worry about it.

, very difficult to do.

It's very good.

Difficult to tell the client that they had a.

$562 cost per conversion, even though they, this account's never

been under 200 in the last three months out of the feed only.

I'm feeding it my own data.

Even the brand's at 193, I'm mean cheaper conversions on my remarketing

for my YouTube channel than I am.

I'm getting a branded conversion, which doesn't happen.

So that's what's really interesting is when we're looking at the

different campaigns, we have to measure it accordingly.

so far is there any, questions?

Cuz this is like, it's, nebulous, but it's more of like the tools box

in order to identify where to start to look at the overall effects.

okay.

Oh, I guess he, Katie, which app would you use in Shopify for this?

Caden, , I, that was like 10 minutes ago.

So what question was that about?

I'm sorry.

That was mainly just for the, , like people like the survey at the end,

you know, where people can say, oh, I found you on Google, or Oh, I

found you on YouTube, or whatever.

Gotcha.

Yeah.

The, uh, post-purchase survey, , that's actually one that I can pull up here.

That is something that I think that we should probably.

, look at doing more often for clients.

It's gonna be helpful because we actually had a client, , come to us and say,

Hey, did you know this is a thing?

? We ki we had a feeling, but they confirmed it.

With this, this is the, week, I guess I would do this here.

Let's say, okay, we're on the 12th, so there was 31 responses

in a post-purchase survey.

That's an email or an app?

can't remember exactly.

Either one is fine.

I don't really, I think there's is an email, but there's an

app for post purchase of like, how did you hear about us?

Same thing that Solutions eight does where it's like, how Hear about

us YouTube, out of 31 responses.

This is not all.

I think there's only like a 25% response rate.

24 of them said, I came from YouTube.

Right now YouTube has 19.56 and we looked the week before and 24 of

them that already converted out of, I don't know how many conversions

said, oh no, I saw it on YouTube.

That's a very interesting statistic.

YouTube tracked less conversions.

Then we had survey responses, which is less than the total conversion.

So one more time.

YouTube tracked less conversions than the survey responses, which is

only a fraction of all conversions.

That's big.

. That's one of the reasons why , , when the client first came to us and

said, Hey, did you know that YouTube's actually crashing it?

based on what people are saying in, in the survey, I think that happened around here

and we just started scaling up YouTube.

, it's pretty fantastic.

And 19 hundreds marketing for the win cadence.

He's pretty much right.

Trust me.

Now bell board's gonna work.

people love bell boards, . but that's what's really interesting.

So we started to scale up more and, what we did is when we looked at , , the m

e r, we actually saw that the as more and more that we put into YouTube,

YouTube showed incremental increases in MER when we looked at the whole

scenario minus . The lift that we're getting on Amazon, we had like a 2.4 me

R like, all right, just keep scaling up.

YouTube is obviously not tracking all conversions.

Maybe tracking a fifth maybe.

But when you look at the ME R of everything is they don't do any

of the marketing besides YouTube, which is why all their conversions

post-purchase surveys, say YouTube.

But when we started to scale this up, I could care less what

the conversions are on Google.

, I'm not even looking here.

As long as this doesn't fall off the map, I shouldn't see a fall off the map

on the back end of the client's website.

But when we're running YouTube, this is gonna be saved.

Display.

It's gonna be the same for Discover.

It's gonna be the same for Performance Max.

This is something that we have to look at.

If I'm spending this, what is the overall lift?

, it's becoming more and more and more the norm of how we're gonna

have to measure these campaigns.

Google, especially when the third party attribution kicks off, where

we're no longer gonna be able to see the third party cookies.

This may be the only thing that we could really mark, we just turned into like

iOS or post iOS 14.5 Facebook marketers now and be like, well, we lost 90% of

our tracking, but here's what's going on.

Like, that's gonna be a potentially scary time.

And that's why we often to get really good at ignoring clients information,

ignoring Google's information and finding out for ourselves as

good marketers, why do we run?

This is what we saw, depending upon what else was in that, media mix of, spend.