Google Ads Strategy: Removing Google’s Algorithm From Standard Shopping

Jon: 00:00:11

Hey everyone, John Moran here.

2

: 00:00:12

And today I'm about to share

with you a quick update to my

3

: 00:00:14

low TROAS bidding strategy.

4

: 00:00:15

There's actually been a

issue that I've been able to

5

: 00:00:17

isolate, identify, and correct.

6

: 00:00:19

And it's something that I think you'd be

interested in seeing, because a lot of you

7

: 00:00:21

have been reaching out to me on LinkedIn

through direct messages, through comments

8

: 00:00:24

about how you've been actually even

doing some posts mentioning me about how

9

: 00:00:27

you're using a low TROAS bidding strategy

and your standard shopping campaigns

10

: 00:00:30

and have been seeing some good success.

11

: 00:00:32

And I have been as well.

12

: 00:00:33

There's the inefficiency though.

13

: 00:00:34

It sometimes will go

really high with our CPCs.

14

: 00:00:37

And that's simply because the

bidding strategy doesn't know

15

: 00:00:40

too much about these people.

16

: 00:00:41

It's the first click that they've

ever potentially have searched

17

: 00:00:43

or first click that, have ever

interacted with Google ads.

18

: 00:00:46

So the predictive bidding strategy

is not necessarily as efficient.

19

: 00:00:48

So low TROAS trying to get a low TROAS is

going to bid very high to make sure that

20

: 00:00:52

they can maximize that conversion value

within a half attributed bidding strategy.

21

: 00:00:57

When you're talking about attribution,

so with low TRO as the byproduct of

22

: 00:00:59

that is, is going to bid high and

because it's going to be a bit high,

23

: 00:01:02

it only really needs to average out to

at least 40 percent or 50 percent of

24

: 00:01:05

what your actual, return needs to be.

25

: 00:01:07

So when you're talking about the

campaign here, for example, you'll see

26

: 00:01:11

that I have a headband campaign that

was getting a dollar 91 average CPC.

27

: 00:01:15

On its own, but this is all July.

28

: 00:01:18

I got 12 clicks, 12, 11, 10,

9, 8, 7, 6, 5, 5, 5, 5, 4.

29

: 00:01:23

And we barely broke the bottom

of the 5 CPC in the first 50.

30

: 00:01:27

Now there's 30, 000 search terms here.

31

: 00:01:29

And the first page is above five ish, but

I still have an average of a dollar 91.

32

: 00:01:32

Why?

33

: 00:01:33

With the low TRO as bidding,

it says, okay, I can spend 12.

34

: 00:01:35

And if I make a sale, that's coming

in around 40 percent TRO as if

35

: 00:01:38

it's a six or seven or 8 sale.

36

: 00:01:40

It's going to be inefficient.

37

: 00:01:42

Now, typically a person would say

why don't I just set a bid cap to

38

: 00:01:45

do a portfolio bidding strategy?

39

: 00:01:46

And I thought instead of doing that

why are we leaning into an algorithm

40

: 00:01:49

that knows a little bit about

the people when they first click?

41

: 00:01:52

Again.

42

: 00:01:52

This is the first time they've

ever actually searched with Google.

43

: 00:01:55

Maybe.

44

: 00:01:55

Maybe they've been looking at some

things online, and maybe you can

45

: 00:01:58

find the people that have been,

identified by Google's algorithm.

46

: 00:02:01

As this is the searches that they've

been doing on Google, or this is certain,

47

: 00:02:04

not search, this is a search that they

may have done on YouTube, or they've

48

: 00:02:06

may been looking at some blogs, or

they may been reading some articles,

49

: 00:02:09

or maybe looking at competitive sites.

50

: 00:02:11

But if it's a first click

inside of Google, Google

51

: 00:02:13

knows very little about them.

52

: 00:02:15

And so it doesn't know where

to actually set the bid.

53

: 00:02:18

So what's interesting about

this is if I said, let's remove

54

: 00:02:20

the algorithm altogether.

55

: 00:02:21

Let's just say, rather than using a half

smart algorithm because of data driven

56

: 00:02:25

attribution, giving too much to brand and

also a half smart algorithm, because it's

57

: 00:02:28

also the first click, what if I just focus

on search terms and focus on new volume?

58

: 00:02:33

So we're going, we're turning

the clocks back 10 years.

59

: 00:02:35

We're going back to max clicks with a bid

cap and why max clicks with a bid cap.

60

: 00:02:39

I'll tell you is because when

you're looking at the average

61

: 00:02:41

cost per click of a dollar 91, but

you're getting 12 and 10 clicks.

62

: 00:02:45

Every one click I get at 12 would

basically be six clicks at 2.

63

: 00:02:50

I take my average CPC and I increase it by

about 25 to 50%, or what I would do here.

64

: 00:02:54

For example.

65

: 00:02:55

Has taken my search terms and

let's say adding in a CPC that

66

: 00:02:59

is above, let's say a four.

67

: 00:03:01

What is my, what does that

look like here for example?

68

: 00:03:04

And do I have a lot of conversions

compared to the ones that are below?

69

: 00:03:07

Yeah, you can see that you see the

differences, 45 conversions in July.

70

: 00:03:12

And if my CPC though is over three.

71

: 00:03:15

Okay, so that's 2000 clicks.

72

: 00:03:16

There's 222 clicks.

73

: 00:03:17

It cost me a thousand dollars made 10.

74

: 00:03:19

All right.

75

: 00:03:19

What about, let's say 3 and 50 cents?

76

: 00:03:21

I'm just using a really.

77

: 00:03:22

bad example here because it's

just the one that was come up,

78

: 00:03:24

but now I have above three 50.

79

: 00:03:26

It's still 4.

80

: 00:03:27

2 ROAS.

81

: 00:03:28

Okay.

82

: 00:03:28

So conversion rate's decent, but what's

interesting is it's not necessarily

83

: 00:03:32

producing the majority of the results.

84

: 00:03:33

Now multiply that with a lot of

different shopping campaigns.

85

: 00:03:38

And then you start to see some

potentially, some opportunity.

86

: 00:03:42

If I have, if you're following

along with my 19 different

87

: 00:03:45

standard shopping campaigns.

88

: 00:03:47

And if I said, what if I just took.

89

: 00:03:49

All of the TROAS, change it into max

clicks, set the bid cap 25 to 50 percent

90

: 00:03:54

higher than my average CPC, cut out

the really expensive fat at the top.

91

: 00:03:58

What would that do?

92

: 00:03:59

So I started to look through each one of

my campaigns and I looked and said, man,

93

: 00:04:02

anything that's above two on that one,

everything's above a three on that one.

94

: 00:04:05

They actually start to have higher

conversion rates, but way higher

95

: 00:04:08

CPCs, much lower amount of clicks,

and then much less conversions.

96

: 00:04:13

So there's inefficiencies there.

97

: 00:04:14

The ROAS inside the campaign, even though

it's only, Half attributed is lower than

98

: 00:04:20

the clicks that are coming in at sub 2.

99

: 00:04:22

what I did is I took the

campaigns and I changed them to

100

: 00:04:27

a maximum click bidding strategy.

101

: 00:04:29

So now I'm on max clicks across the board.

102

: 00:04:31

I set a bid cap at each one

of these to 2 and 15 cents and

103

: 00:04:33

said, let's see what happens.

104

: 00:04:34

We did that on the 8th.

105

: 00:04:35

So the 8th through the 11th, what

you'll see here, for example.

106

: 00:04:39

Is my cost is down 16%.

107

: 00:04:41

My clicks are up 78 percent and

my 62 percent more conversions and

108

: 00:04:45

56 percent more conversion value.

109

: 00:04:47

So I was able to cut my CPA in half

and almost double my row as in the

110

: 00:04:50

first four days, I'm like that's cool.

111

: 00:04:51

How is my positioning still good?

112

: 00:04:52

How my search terms are still good.

113

: 00:04:54

Excellent.

114

: 00:04:55

Can I scale this?

115

: 00:04:56

If I look at the last kind of 14

days, for an example, we can see

116

: 00:05:00

a really nice, healthy scale.

117

: 00:05:01

The clicks are going from 1100

and 28 half conversions or quarter

118

: 00:05:05

conversions to 6, 000 clicks, 178.

119

: 00:05:07

What about my cost and my

conversions costing conversions

120

: 00:05:10

are ramping up with it very well.

121

: 00:05:12

I can scale this to a one to one.

122

: 00:05:13

Let's take a little bit of a deeper look.

123

: 00:05:14

Let's take a look at all

of the campaign types.

124

: 00:05:18

Actually, I have that already here.

125

: 00:05:19

Let's look at all of the campaign

types that are search and shopping.

126

: 00:05:23

And then let's look at everything

here that's not this one

127

: 00:05:27

outside of the country search.

128

: 00:05:28

Let's filter by that campaigns.

129

: 00:05:30

What does that arch look like?

130

: 00:05:31

That looks pretty good here as well.

131

: 00:05:32

It's starting to ramp up in 3, 000,

cost in 400, all the way up to 11, 679.

132

: 00:05:38

That's interesting.

133

: 00:05:38

So now it's scaling.

134

: 00:05:39

If my predictions are correct though, and

I should see a majority of these people

135

: 00:05:43

coming back through my brand campaign,

my brand should look really good.

136

: 00:05:47

Let's check the last seven days compared

to the previous seven days here now.

137

: 00:05:50

And if I'm checking the last seven

days compared to previous seven

138

: 00:05:52

days, my brand campaign only spent 4

percent more, but Oh, I broke Google.

139

: 00:05:56

That's okay.

140

: 00:05:57

Spend 500 more.

141

: 00:05:58

So nothing too crazy, 12, 1 to 12, 6,

but I made 32, 000 more, 146 to 180.

142

: 00:06:03

Excellent.

143

: 00:06:04

So for pretty much no cost, I gained

32, 000 in new branded conversions.

144

: 00:06:10

Where are they coming from?

145

: 00:06:11

Everyone that's engaging in my search

campaigns that are getting more ROAS,

146

: 00:06:14

sorry, not ROAS, more convergent

value, more convergent, excellent.

147

: 00:06:17

So what we found out.

148

: 00:06:18

Is that the T row as bidding

strategy was over clicking?

149

: 00:06:21

Why?

150

: 00:06:21

It was a too small of a goal.

151

: 00:06:23

I had to do that though, because it had

too small of attribution as a proven here.

152

: 00:06:27

It's going to come back

through the brand campaign.

153

: 00:06:28

And then I thought, okay well,

let's let's actually verify this.

154

: 00:06:31

Google ads is a horrible reporting model.

155

: 00:06:33

So let's look at a couple

other things inside of the.

156

: 00:06:36

Lifetimely.

157

: 00:06:37

How was my week doing this last week?

158

: 00:06:38

I gained 25 percent more net

profits and I'm up to a quarter

159

: 00:06:41

million in net profit last week.

160

: 00:06:43

That's good.

161

: 00:06:43

That's scaling up well.

162

: 00:06:44

What about if that was Facebook though?

163

: 00:06:46

Facebook must has to be pushing, right?

164

: 00:06:47

Let's verify that inside of the

dashboard of Norpe, you can see Google

165

: 00:06:50

ads is 41 percent more spend 53 port.

166

: 00:06:53

53 percent more new visits to the

site, my true cost of acquiring a

167

: 00:06:58

first time customer globally dropped

and inside of Google went up 0.

168

: 00:07:01

3 percent in the last seven days.

169

: 00:07:03

So I made 32 percent more conversion

value off of 31 percent more transactions.

170

: 00:07:08

Again, this is multi click attributed.

171

: 00:07:10

I'm taking all of the clicks

that already came from Facebook.

172

: 00:07:13

Out including brand and then

seeing what this is working with.

173

: 00:07:16

So I was able to scale this up in

the last week, 41 percent get 32%.

174

: 00:07:19

I didn't change my CAC.

175

: 00:07:21

My CAC stayed about the

same, which is fantastic.

176

: 00:07:24

That's what I want.

177

: 00:07:25

And what's nice is we're on a really nice,

good ramp up here in the last 14 days.

178

: 00:07:29

And we're going to continue to ramp up.

179

: 00:07:30

So this is a really nice scale.

180

: 00:07:32

Net profits holding CAC is holding.

181

: 00:07:34

I'm scaling up.

182

: 00:07:35

My campaigns are becoming much, much, much

more efficient all because we wiped out.

183

: 00:07:39

The above where I needed to be clicks and

Google's algorithms don't know enough.

184

: 00:07:44

The single click attribution is dead.

185

: 00:07:46

Data driven is over attributing

to my brand campaign.

186

: 00:07:49

Fine.

187

: 00:07:49

But then I get to see my in app standard

shopping CACs, my in app standard

188

: 00:07:52

shopping ROAS, and then measure the

brand return search, ensuring that

189

: 00:07:56

no one else is actually scaling.

190

: 00:07:57

Facebook spent about the same, made 6.

191

: 00:08:00

6 percent more.

192

: 00:08:00

Why?

193

: 00:08:01

It's because they're running the advantage

plus shopping campaigns that are also

194

: 00:08:04

capturing more new visits with less.

195

: 00:08:05

With less ad spend.

196

: 00:08:06

So we're competing with each other

there a bit, but the ratio of new

197

: 00:08:09

visits to visits is actually 13.

198

: 00:08:11

5 or 13 percent higher than visits.

199

: 00:08:14

more new cold traffic, better better

efficiencies, cutting out the fat at

200

: 00:08:17

the top, multiply that by 19 campaigns.

201

: 00:08:20

And the best part is now I'm going to be

able to still continue to scale this up.

202

: 00:08:24

And if you look at the, the

cost factor since the 10th.

203

: 00:08:27

If I just look here, for example, I was

able to take my campaigns and increase

204

: 00:08:32

them 130 percent more in one week

without hitting a point of diminishing

205

: 00:08:37

returns in either my MER or my CAC.

206

: 00:08:41

Top line actually got better.

207

: 00:08:42

Why?

208

: 00:08:42

Because Google ads sucks at attribution.

209

: 00:08:44

And this is all that matters is your

clients about, bottom line or top

210

: 00:08:48

line, whatever you want to call it.

211

: 00:08:50

Low TRO is good.

212

: 00:08:51

There isn't, there is an inefficiency

in there, but you still can scale.

213

: 00:08:56

By taking a deeper approach,

removing the algorithm of what Google

214

: 00:09:00

doesn't know and leaning into good

old fashioned quality marketing.

215

: 00:09:04

Thanks so much.