Stop Wasting Money: How to Analyze Keywords for Effective Ad Campaigns

Speaker: 00:00:11

Hey everyone, Regina

here with starter PPC.

2

: 00:00:13

Join me.

3

: 00:00:14

We're going to look through

an inbound search campaign.

4

: 00:00:16

We're going to see if there's anywhere

we can cut costs, make the campaign

5

: 00:00:19

a little bit more efficient, get it

to stop wasting money and, hopefully

6

: 00:00:23

reallocate some of the budget to

the better performing keywords.

7

: 00:00:27

So let's begin.

8

: 00:00:28

this is one of our clients

and they are actually a farm.

9

: 00:00:33

It's unique, but they sell a

subscription to exotic fruit.

10

: 00:00:38

They also have exotic fruits.

11

: 00:00:40

Gift boxes, and they also

have a nursery right now.

12

: 00:00:43

We're just going to be looking

at the fruit keywords in

13

: 00:00:46

the inbound search campaign.

14

: 00:00:47

since they are an e commerce company,

most of the budget is spent on the

15

: 00:00:51

shopping network in other campaigns.

16

: 00:00:54

But the.

17

: 00:00:55

Inbound search campaigns are usually

the ones that require the most manual,

18

: 00:00:59

heavy lifting to get them to, become

optimized and working appropriately.

19

: 00:01:04

so right now I just have the last 30 days

pulled up and I'm looking at all keywords.

20

: 00:01:09

So keyword status all,

and then I put a cost.

21

: 00:01:12

So I only want to see the

keywords that have spent at

22

: 00:01:15

least 1 within my date range.

23

: 00:01:17

Cause I just want to filter through any.

24

: 00:01:18

irrelevant data, it's not spending

enough to really even matter.

25

: 00:01:22

so you can see here that we're

using exact match keywords.

26

: 00:01:26

fun fact, when we first started,

we were using broad match.

27

: 00:01:29

Then we ended up moving to phrase match.

28

: 00:01:31

And then we ended up

moving to exact match.

29

: 00:01:33

We didn't really know what

to expect with a farm type.

30

: 00:01:36

business model.

31

: 00:01:37

And it turns out that there are a lot

of people out there that just want

32

: 00:01:41

to look up pictures of fruit and have

no intention of actually buying them.

33

: 00:01:45

The problem with keyword

campaigns is the algorithm will,

34

: 00:01:49

it's going to waste your money.

35

: 00:01:50

It's going to go out and it's going

to see that, there's a huge amount

36

: 00:01:55

of people searching for fruit.

37

: 00:01:57

and it's going to try to, bid on all of

those placements, And even if you tell it,

38

: 00:02:02

Hey, I want you to maximize conversions.

39

: 00:02:04

You can't really trust Google care, right?

40

: 00:02:07

You can not really rely on those

automated bidding strategies.

41

: 00:02:10

That much because Google has a more

incentive to, fill those placements

42

: 00:02:15

for those, keywords like exotic papaya

fruit that person probably has very

43

: 00:02:21

little intention to actually buy fruit.

44

: 00:02:23

But Google loves to fill those placements

because if it can get the click, it

45

: 00:02:27

gets to, collect its commission, right?

46

: 00:02:29

So you just have to make sure that it's

not wasting money, especially if you're

47

: 00:02:33

using those automated bidding strategies.

48

: 00:02:35

so we ended up switching to exact

match after a while of trying the

49

: 00:02:39

broad match and then the phrase match

because we wanted to rein it in, right?

50

: 00:02:44

Rein the algorithm in.

51

: 00:02:45

We don't want it to bid on things

that are just people looking up

52

: 00:02:49

fruit information or fruit pictures.

53

: 00:02:51

We want people that are

like wallet in hand.

54

: 00:02:54

I want to buy some fruit

on the internet, right?

55

: 00:02:56

that's what we're trying to identify.

56

: 00:02:58

And that's why we are restricting it.

57

: 00:03:00

And we're saying, Hey, exact match.

58

: 00:03:02

This is the most strict type of

keyword we want you to go after.

59

: 00:03:06

search terms where users are searching

for something that is very similar,

60

: 00:03:12

almost synonymous to this phrase,

that's what exact match means.

61

: 00:03:16

I'm going to sort by cost and I actually

want to see more data than this because

62

: 00:03:20

this is only 340 in the last 30 days.

63

: 00:03:23

Again, this is not a big money spender

compared to our other campaigns.

64

: 00:03:27

the client is spending,

under 5, 000 overall.

65

: 00:03:31

Because, at starter PPC, we

manage, businesses that have

66

: 00:03:34

budgets of 5, 000 or less.

67

: 00:03:36

but even still within that small

budget of 5, 000 or less, this search

68

: 00:03:40

campaign is spending, Not as much

as the, shopping campaigns, but

69

: 00:03:44

it's still spending money, right?

70

: 00:03:46

So I'm just going to pull up all time

data so I can see as much as I can.

71

: 00:03:50

I can see that overall in this

campaign, 2, 000 has been spent.

72

: 00:03:54

all of it in, since summer of 2023.

73

: 00:03:57

So a few months ago, and

here's some of our old broad

74

: 00:04:00

match keywords that we paused.

75

: 00:04:02

so I'm just going to see what we can see.

76

: 00:04:03

So the one that's spending the

most is exotic fruits delivery.

77

: 00:04:07

second most fresh exotic fruits is

spending the second most, which is

78

: 00:04:11

little more than half the amount

that the main keyword spent.

79

: 00:04:15

this is very interesting.

80

: 00:04:16

So exotic fruit delivery indicates

that somebody has their wallet

81

: 00:04:21

in their hand and they're asking

for fruit to be purchased on the

82

: 00:04:26

internet and delivered to their door.

83

: 00:04:28

And the click through rate on this one

is 17%, And then fresh exotic fruit.

84

: 00:04:33

has some purchase intent, right?

85

: 00:04:35

Somebody is looking

for fruit that's fresh.

86

: 00:04:38

So they're probably not just looking

for images, and they're probably not

87

: 00:04:42

just looking for information, but maybe.

88

: 00:04:44

But because they're not saying I

want it delivered to my door, it has

89

: 00:04:48

a slightly lower purchase intent.

90

: 00:04:49

And so you can see that the click

through rate here is only 5%.

91

: 00:04:53

416 clicks on this one,

400 clicks on this one.

92

: 00:04:56

So we have plenty of data to get a good

average on that click through rate.

93

: 00:05:00

And this is what happens, you guys.

94

: 00:05:02

People search for something.

95

: 00:05:03

The sponsored search

results are at the top.

96

: 00:05:05

The organic is at the bottom.

97

: 00:05:06

Someone who has an intent to buy

is more willing to click on an ad.

98

: 00:05:10

They're going to be fine with dealing

with those sponsored search results.

99

: 00:05:13

And someone who is not interested

in buying, they're looking up

100

: 00:05:18

fresh, exotic fruit for some other

reason, those are going to be more

101

: 00:05:22

likely to skip over the sponsored

section and go right over to the.

102

: 00:05:26

organic search results because

they're looking for articles or maybe

103

: 00:05:29

they're even going to click over

to the images and look for images.

104

: 00:05:32

Let's just, to see what is the conversion

rate on these conversion rate, so we

105

: 00:05:37

have a slightly higher conversion rate

even on the exotic fruits delivery one.

106

: 00:05:40

So 0.

107

: 00:05:41

7.

108

: 00:05:43

2 percent on exotic fruits delivery

and fresh exotic fruits is 0.

109

: 00:05:47

5%.

110

: 00:05:48

Now, if you guys are going, Oh my gosh,

these conversion rates are very low.

111

: 00:05:52

Just keep in mind that this

is not the only ad that most

112

: 00:05:56

people interact with, right?

113

: 00:05:58

So probably what's happening

is the conversions.

114

: 00:06:02

If there's to add interactions, one

from the search campaign, one from the

115

: 00:06:05

shopping campaign, the algorithm is

attributing most of that conversion

116

: 00:06:10

to the shopping campaign and only a

small amount to the search campaign.

117

: 00:06:14

data is not really relevant in judging the

business model as a whole, or using that

118

: 00:06:19

conversion rate as a benchmark for this

business, we would have to go to the all

119

: 00:06:23

campaigns view and look at the account

holistically to try to determine that.

120

: 00:06:27

and actually even better is to look

at the actual conversion rates of

121

: 00:06:30

the website from Google analytics.

122

: 00:06:32

So that you're not relying

on Google's attribution and

123

: 00:06:35

their ability to track users.

124

: 00:06:37

but we can at least compare two

keywords to each other, right?

125

: 00:06:41

0.

126

: 00:06:42

72 up here, 0.

127

: 00:06:42

5.

128

: 00:06:43

The click through rate is insane.

129

: 00:06:44

And so that's going to result in, a ROAS.

130

: 00:06:47

Oh, okay.

131

: 00:06:48

This is interesting.

132

: 00:06:49

So the ROAS on exotic fruits delivery

is actually lower than the ROAS, return

133

: 00:06:55

on ad spend for fresh exotic fruits.

134

: 00:06:57

Even though these two

numbers are different.

135

: 00:06:59

So what could be the difference?

136

: 00:07:00

Okay, cost per click is higher on

this one and lower on that one.

137

: 00:07:04

That makes sense, right?

138

: 00:07:05

This is going to be

higher purchase intent.

139

: 00:07:07

So it's going to be higher competition.

140

: 00:07:09

how much data are we dealing with here?

141

: 00:07:11

Okay.

142

: 00:07:11

So we're dealing only with three

conversions here and two conversions

143

: 00:07:14

here that have been attributed to

those keywords into this campaign.

144

: 00:07:18

So we don't have a lot

of data to work with.

145

: 00:07:20

I think what we should do is dig

into this a little bit further.

146

: 00:07:24

Fresh fruit delivery, exotic fruit

box, passion fruit delivery, dragon

147

: 00:07:29

fruit delivery by passion fruit tree.

148

: 00:07:31

We can dig into all of this.

149

: 00:07:33

if I sort by conversion

value divided by cost.

150

: 00:07:36

Fresh Fruit Online Delivery has

one conversion at a very high ROAS.

151

: 00:07:41

Passion Fruit Delivery has two

conversions at a pretty high ROAS.

152

: 00:07:47

So we would like to spend

more money on this one.

153

: 00:07:49

It's only spent 86.

154

: 00:07:51

because it's more specific

passion fruit delivery.

155

: 00:07:54

Someone definitely wants to buy that.

156

: 00:07:55

Okay.

157

: 00:07:56

I think what we should do is keep

all this in mind, head over to the

158

: 00:07:58

search terms and see what we can see.

159

: 00:08:00

This type of keyword

research is very messy.

160

: 00:08:02

You guys have to sometimes.

161

: 00:08:04

You feel like you're going in circles.

162

: 00:08:05

You're going back and

forth between keywords.

163

: 00:08:07

You're going back and

forth between search terms.

164

: 00:08:09

You're pulling up all data.

165

: 00:08:10

You're pulling up last three months.

166

: 00:08:11

You're pulling up different columns,

you're sorting, you're unsorting.

167

: 00:08:15

and so it's messy and, there's no right

or wrong way to do it, but I'm going

168

: 00:08:19

to show you how I think through it.

169

: 00:08:22

so let's.

170

: 00:08:23

See what we can see here.

171

: 00:08:24

Okay.

172

: 00:08:25

So I'm going to sort by cost.

173

: 00:08:26

So the search terms, this is what the

user is actually searching for, right?

174

: 00:08:30

And it matches up with a keyword.

175

: 00:08:32

So we told it to go after

fresh exotic fruits and the

176

: 00:08:35

algorithm bid on exotic fruits.

177

: 00:08:37

And this is the search term that

has spent the most amount of money.

178

: 00:08:40

Wow, by far.

179

: 00:08:41

The next closest one has

spent 97 instead of 119.

180

: 00:08:45

and yet it's received zero

conversions that have been attributed.

181

: 00:08:49

162 clicks and not a single conversion.

182

: 00:08:52

So this is a prime example

of the algorithm seeing.

183

: 00:08:56

Tons of people looking for exotic fruit,

deciding it matches up with our keyword.

184

: 00:09:00

Even though we've chosen

exact match, right?

185

: 00:09:03

It's just ignoring the

word fresh and fresh.

186

: 00:09:05

In my opinion is the only part of that

keyword that indicates purchase intent.

187

: 00:09:11

Because if someone's searching for exotic

fruit, they probably just want to read

188

: 00:09:15

about or see pictures of exotic fruit.

189

: 00:09:18

If they're searching for fresh exotic

fruit, that means they're actually

190

: 00:09:21

looking for fruit that's fresh, right?

191

: 00:09:23

So they're probably looking to

buy it, or at least they're much

192

: 00:09:26

more likely to be looking to buy.

193

: 00:09:28

and yet, because the amount of people

searching for exotic fruit is so

194

: 00:09:33

vast compared to fresh exotic fruit,

the algorithm just takes liberties.

195

: 00:09:37

It goes, I know they want the exact

match, but it's close enough, especially

196

: 00:09:42

considering how big this market is.

197

: 00:09:44

I swear, that is how the algorithm thinks.

198

: 00:09:47

Because it's close enough, especially

considering how big our opportunity is

199

: 00:09:51

over here, how big the market size is.

200

: 00:09:53

it's continuing to waste money

here without a single conversion.

201

: 00:09:56

even though we're using an

automated bidding strategy.

202

: 00:09:59

Okay.

203

: 00:10:00

Exotic fruit box.

204

: 00:10:01

Second most spend.

205

: 00:10:02

It has 1.

206

: 00:10:03

52 conversions attributed to it.

207

: 00:10:06

the cost per click is higher, which

makes sense because someone looking for

208

: 00:10:09

a box is a much higher purchase intent.

209

: 00:10:11

And, This is.

210

: 00:10:13

1.

211

: 00:10:13

53 ROAS, Hi there.

212

: 00:10:16

Quick interruption.

213

: 00:10:17

Do you know the main thing that

prevents small business owners from

214

: 00:10:20

getting their Google Ads account

into a position to grow and scale?

215

: 00:10:23

Budget.

216

: 00:10:24

A lot of businesses, especially

those that are just starting

217

: 00:10:26

out, have limited budgets.

218

: 00:10:28

And so because of this, they're turned

away by most ad agencies because most ad

219

: 00:10:31

agencies have minimum budget thresholds

that they're willing to work with.

220

: 00:10:35

So what happens is the business owners

end up learning Google ads themselves.

221

: 00:10:38

And the problem with that is

that most of the advice online is

222

: 00:10:41

geared towards larger accounts.

223

: 00:10:43

And the advice doesn't have any of those

strategies or tricks that can kickstart

224

: 00:10:47

the algorithm into giving a small

account a leg up over larger competitors.

225

: 00:10:52

So it often just doesn't work.

226

: 00:10:53

And the business just ends up

losing money month over month.

227

: 00:10:56

If this sounds familiar

starter PPC can help.

228

: 00:10:58

We offer Google ads management services

that are designed for accounts that

229

: 00:11:02

have between 1000 and 5, 000 budgets.

230

: 00:11:05

Because all of our clients are just

starting out, we've come up with ways to

231

: 00:11:08

keep our management fees significantly

lower than most agencies, because

232

: 00:11:12

we know that every dollar saved on

management fees just goes towards

233

: 00:11:15

the ad budget, which is going to help

the algorithm gather speed and power.

234

: 00:11:19

So if you're serious about growing your

business and you'd like a team of Google

235

: 00:11:22

ads experts to help you without breaking

the bank, check us out at starter PPC.

236

: 00:11:26

com.

237

: 00:11:26

Okay.

238

: 00:11:27

Back to the video.

239

: 00:11:28

Okay.

240

: 00:11:28

So we're looking at two conversions.

241

: 00:11:29

At least it's not one Double the data.

242

: 00:11:32

Woo.

243

: 00:11:33

153 is not bad for ROAS.

244

: 00:11:35

what we're seeing in this account is that

if we're getting a ROAS around 180, that's

245

: 00:11:40

a MER of I think that's a MER of 400.

246

: 00:11:43

ROAS.

247

: 00:11:44

being 152 is really not bad because

that's going to equate a really high Mer.

248

: 00:11:49

and this company gets a lot of

returning customers because they have

249

: 00:11:52

a subscription box so that Rose really

doesn't have to be that high in order

250

: 00:11:55

for the business to be profitable.

251

: 00:11:57

It's a good business model.

252

: 00:11:58

Y'all should start a farm.

253

: 00:12:00

okay.

254

: 00:12:00

So here again, exotic fruit.

255

: 00:12:03

Oh my goodness.

256

: 00:12:03

Another 90 towards the same

keyword as the top one.

257

: 00:12:07

Only this one's plural and this one's not.

258

: 00:12:09

So we have to add both of these together,

an extra 98 clicks to the wasted spend.

259

: 00:12:14

Oh, and here we have one conversion

at 167%, which sounds great until

260

: 00:12:20

you factor in the fact that this

top keyword is the exact same.

261

: 00:12:24

keyword and that brings

that row as way down.

262

: 00:12:27

yeah, I'm very skeptical about this.

263

: 00:12:30

I think that what we need to do is dig

into this keyword, fresh exotic fruit.

264

: 00:12:33

So I'm going to add a filter and

say keyword text contains or I guess

265

: 00:12:40

let's do equals fresh exotic fruits.

266

: 00:12:43

We'll just look at this data.

267

: 00:12:46

Alright, so overall this

keyword has spent 295.

268

: 00:12:51

That we can see, right?

269

: 00:12:53

Google doesn't share with us all search

term data, which is very frustrating.

270

: 00:12:56

and out of that almost 300, we

have one conversion at a row as

271

: 00:13:01

conversion value divided by cost of 0.

272

: 00:13:03

51 again, I'm looking at all time

data here, there's 32 versions of the

273

: 00:13:07

same search term, rare, exotic fruit.

274

: 00:13:10

It's just completely

ignoring the word fresh.

275

: 00:13:12

And at the bottom you can

compare, the, Okay.

276

: 00:13:15

Cosmic it's actually getting a ROAS.

277

: 00:13:18

That's exactly the same as

the total campaign ROAS.

278

: 00:13:23

So I was expecting it to

drag the total campaign down.

279

: 00:13:27

But it's actually not dragging

the total campaign down.

280

: 00:13:32

so I'm not sure whether we

should pause this or not.

281

: 00:13:35

My suspicions are if the other

keywords can Spend more, they

282

: 00:13:41

will be able to optimize.

283

: 00:13:42

And since they have a higher purchase

intent, we could probably make a

284

: 00:13:46

higher row as on them over time.

285

: 00:13:48

Let's look at some different

date ranges just to double check.

286

: 00:13:51

Whenever you're deep diving

like this, it's always best

287

: 00:13:54

to look at a few date ranges.

288

: 00:13:55

I always do all time.

289

: 00:13:57

Most recent three months, that type

of thing again, total campaign 0.

290

: 00:14:01

51, 12 conversions in

total that we can see,

291

: 00:14:04

So now that I've removed the filter for

cost is greater than 1, I can see that

292

: 00:14:09

there is a little bit more wasted spend.

293

: 00:14:11

So this is the real

ROAS for this one is 0.

294

: 00:14:14

49, whereas the campaign is 0.

295

: 00:14:15

51.

296

: 00:14:16

So it is indicating, that it's

dragging the average down.

297

: 00:14:20

And if we stop going after it, there's a

chance that the average will drag upwards.

298

: 00:14:24

if the other campaigns are getting

a slightly higher row as I think

299

: 00:14:28

this is my biggest indicator.

300

: 00:14:30

What I'm usually looking for is it

dragging the overall row is down.

301

: 00:14:33

If so, let's try turning it off

for two weeks and see how it does.

302

: 00:14:37

let me show you an interesting column

that I think will help in your keyword

303

: 00:14:40

research is this click share column.

304

: 00:14:43

So when you're considering whether you

want to pause a keyword, the question

305

: 00:14:47

is, can all the other keywords do better?

306

: 00:14:49

And part of that question is how.

307

: 00:14:52

Much of the total volume for that keyword.

308

: 00:14:54

The total search volume is

currently being taken up.

309

: 00:14:58

Can you easily shove more money into

the other search terms, the other

310

: 00:15:02

keywords and, or are you maxing out all

those other keywords and it's going to

311

: 00:15:06

start getting expensive to use them.

312

: 00:15:09

so the click share column shows

you, click share is the clicks you

313

: 00:15:12

received on the search network divided

by the estimated maximum number of

314

: 00:15:16

clicks that you could have received.

315

: 00:15:17

For that keyword, how often your

product is clicked depends on your

316

: 00:15:20

ads target, targeting settings,

budget approvals, statuses, bids,

317

: 00:15:24

and the quality of your product data.

318

: 00:15:26

So the fresh, exotic fruits.

319

: 00:15:29

it has a vast market.

320

: 00:15:30

So we're going after less than 10%.

321

: 00:15:32

That's as low as it'll show you,

of the total market for that one.

322

: 00:15:36

Whereas passion fruit delivery,

we're already going after 12%,

323

: 00:15:40

but that's still pretty low.

324

: 00:15:41

if it was more like 80%,

I would be hesitant.

325

: 00:15:44

I would be like, I don't know if

we can shove more money into that.

326

: 00:15:47

because we're already at 80 percent of the

total amount of searchers every month, and

327

: 00:15:51

it's just going to start getting really

expensive to get that remaining 20%.

328

: 00:15:55

Exotic fruit box.

329

: 00:15:57

based on this click share, huge

market and a lower purchase intent,

330

: 00:16:00

especially considering Google

is just ignoring the word fresh.

331

: 00:16:03

And, see if the budget for this campaign

can, the keyword that's spending the

332

: 00:16:08

second most amount of money reallocate all

this budget essentially to all the other

333

: 00:16:12

keywords and see if it's going to, make

the ROAS trend upward over time, which I

334

: 00:16:17

suspect it will, since we just determined

that the ROAS for this keyword is slightly

335

: 00:16:22

lower than the average for the campaign.

336

: 00:16:25

that's my conclusion here.

337

: 00:16:26

I think we can try pausing that keyword.

338

: 00:16:28

I love the word delivery.

339

: 00:16:30

I'm not sure how I feel

about the word box.

340

: 00:16:32

It does indicate someone's looking

for a product though, so I'll leave it

341

: 00:16:35

for now, but it's not very specific.

342

: 00:16:37

if someone's looking for a specific type

of fruit, that's going to be always going

343

: 00:16:40

to be a higher conversion rate, but it's

also going to be probably more expensive.

344

: 00:16:44

So We'll have to see how it

does after pausing this keyword.

345

: 00:16:46

I want to do just 1 thing at a time.

346

: 00:16:48

So we'll do this for a week or 2.

347

: 00:16:49

See if the row starts trending

upward in this campaign and

348

: 00:16:52

then we'll do this again.

349

: 00:16:54

So all right.

350

: 00:16:55

Thank you for watching.

351

: 00:16:55

I hope that this was informative.

352

: 00:16:58

I know sometimes managing keywords

can be a little bit anxiety producing

353

: 00:17:02

because you have to take some risks,

especially when you have limited data.

354

: 00:17:06

You have to think logically about

whether the keyword has purchased

355

: 00:17:09

intent or how the algorithm is.

356

: 00:17:12

Thinking about this keyword based

on the market size, or like we just

357

: 00:17:16

discovered how the algorithm is just

ignoring parts of the keyword that

358

: 00:17:19

are the most important part, like

the word fresh, because oftentimes

359

: 00:17:22

the data isn't super conclusive, and

so you have to take some chances.

360

: 00:17:27

but I think that with the data we

found, we are using some education,

361

: 00:17:32

backed, guesswork here and I'm going

to pause it and we'll see how it does.

362

: 00:17:35

Okay.

363

: 00:17:35

Thanks for watching.