Advanced Google Shopping By Andreas Reiffen

#SMX #11B @AndreasReiffen

Data Insights from Behind the Scenes

Advanced Google Shopping


About…

•  Data-driven online advertising strategist

•  Online retail expert

•  Entrepreneur

•  €3 billion in customer revenues this year

•  SaaS product for Google Shopping & Search

•  130 true experts in their field

•  Offices in Germany & UK, new office in NYC

… me … crealytics & camato


Using Shopping to manage inventory

Topics we‘ll cover today …

How does price influence performance? Will pricing and bidding become one in the future?

Can you use Shopping to push slow sellers? How to avoid common pitfalls?

The role of price in Google Shopping


Google Shopping spending already surpasses text ads in in the US and UK

DE spending share

US spending share

UK spending share

72% 67% 57%

H116

43%

90%

H214

23%

77%

H215

28%

H115H114

33% 10%

ShoppingTextAds

36%

H214

59%

H114

64%

41%

56%

H215H115

44% 46%

54%

H116

44%

56% 34%

H116

49%

56%

H215

51% 46%

H214

44%

44%

H115

54%56% 66%

H114

AnalysisbasedoncrealyticsdatafromretailcampaignsinFashion,Luxury,OutdoorandSports;20Mclicksintotal


Using Shopping to manage inventory


You want to push slow sellers while saving budget for products which are almost sold out

Product Stock level over time Action

Week 1 Mark down!

Week 1 Sold Out!


40% of budget is allocated to products that will sell on their own within 3 weeks

Comment

PPC budget should be allocated to inventory that is high in stock or sells slowly At the moment 40% of all budget is spent on quick sellers while only 21% is spent on products that will be in stock after 3 months time

PPC budget allocation by stock projection for top 100 products

* top 100 products, at least 1500 clicks per week / product

40% of budget 21% of budget

10

21 16 14 13 12 11 10

15

24

20

0

20 17 18 23

5

22 19 15 2 1 6 8 4 7 9 5 3

Spend


Most people do not buy what they are looking for

Only 34% of conversions match the product that was clicked


Same Designer Different Designer

purchase click

34%

16%

14%

15%

21%

64%

36%

Diff

ere

nt

Ca

teg

ory

Sa

me

Ca

teg

ory

We analysed what people actually bought when they clicked on a product ad


If people don’t buy what they are looking for, how can you manage inventory via Shopping?


With yield we refer to the real value contribution of selling a product via paid advertising

How-to Example

$80

Product

$80

$0

Stock level over time Yield Action

Week 1 Mark down!

Week 1 Sold Out!

$90


Track all products which were sold after a click on a product ad

How-to Example

1,000 clicks

11 Items sold

x 4

x 4

x 2

x 1

Conversions Product clicked


Normally you take your margin values to inform the bidding about the value of an advertisement

How-to Example

Products purchased

4

4

2

1

$120

$80

$8

$22

Total Margin = $230

Product Clicked Quantity Margin totals

$150

Margin

$30

$20

$4

$22


Based on this data the bidding calculates a value per click and suggests a bid

Clicks = 1000

Value per click: $230 / 1000

Existing value per click

= $0.23

How-to Example


The blue sneaker is a slow seller which we want to push

How-to Example

Products purchased

4

4

2

1

$120

$320

$8

$22

Total Margin/ Yield = $470

Product Clicked Quantity Margin/Yield Totals Margin/Yield

$30

$80

$4

$22 $150


New bid takes yield into account and will be higher

Clicks = 1000

Value per click: $470 / 1000

Updated value per click

= $0.47

Value per click increases $0.47 which will be reflected in a higher bid and more sales of the black sneaker

How-to Example


What people buy after a click on a product ad is often very random. Does our approach work anyhow?

1) Brands and Categories have specific stock level profiles 2) Stock level profiles per brands and categories stay consistent over time

Our approach will work if …

64% same designer

65% same category

Clicked to bought


Designers (and categories) have specific stock level profiles which are stable over time

QUAY AUSTRALIA Inventory average NEW LOOK

41% 45% 38%

30% 31% 35%

29% 24% 26%

week 3 week 2 week 1

39%

22%

avg

38%

21% 26% 31%

76% 70% 67%

2%

week 2

5%

week 1

2%

week 3

normal low stock high stock

Low stock = sold out within 2 weeks

High stock = will last 3 months


When we simulate the effects on CPC, clicks and conversions, we see a more effective acquisition

Account result change

Product acquisition change

100%

Yield

100%100%

171%147%

190%

Cost Revenue

MarginacquisitionYieldacquisition

Comment

Our simulation shows that by incorporating yield values more high stock level products and less low stock level products will be sold Yield grows faster than revenue as yield is the KPI we are optimizing towards

lowstock

highstock

normal

+87%

-32%

+20%


Pro-actively managing inventory is

possible

People usually don’t buy what they click on

Incorporate yield values to

manage inventory

$.47

Your key takeaways


The role of price in Google Shopping


Impressions and clicks in Google Shopping are often very sensitive to price changes

Clicks drop off after product price increase Chart Info

5% price increase coincides with a 60% decrease in clicks

0

20

40

60

80

100

120

0

10

20

30

40

50

60

Pricein£

+5%

clicksownprice Google product category: Apparel & Accessories > Shoes > Sneakers Brand:

Days

Clicks

Price


Google will take traffic away from you if your product prices exceed the average market price

Product price change

to avg market price Sum of Imps Sum of clicks Result

+43%

After

106%

Before

74%

-70%

30,002

Before After

100,239

683

-79%

After

3,222

Before

Google takes away 70% of traffic Result: to maintain traffic, you will need to bid much higher *based on 700 products

Product pricing is key to success in Google shopping


We compared cheap vs. expensive products and used a test setup to guarantee meaningful results

We compared

cheap products (below avg. price)

with expensive products (above avg. price)

The Idea

•  At least six competitors •  Available at all times •  Similar products (all

sneakers)

The Criteria The Test

•  Products were excluded from normal shopping activity

•  Results were not influenced by regular bidding activities


For all products we often see S-shaped curves, but cheap products generate traffic much earlier

There is a direct relationship between product price, maxCPC bid and impressions Key insights

Impression volume for expensive products significantly lower Cheap products hit Impression limit after 5 days

1.2

1,000 0.6

1.0

0.4

0.2

0 0.0

500

0.81,500

2,000

5-13-165-11-16 5-12-16 5-15-16 5-17-165-16-165-14-16

MaxCPCImpressionsperproduct

MaxCPC

expensiveproductscheapproducts


Average CPC is higher for products with prices above average market price

Max CPC bid, Avg CPC while bidding up Key insights

Avg. CPCs of expensive products is about 15% higher despite same bids

5-16-16 5-17-16

0.25

5-11-16

0.43

0.750.65

0.710.65

5-15-16

0.580.54

5-14-16

0.480.35

5-13-16

0.33

5-12-16

0.180.14

cheapproducts

MaxCPCexpensiveproducts


Cheap products generate the lion‘s share of total traffic

Traffic of similar products within one shop Key insights

Cheap products generated much more traffic despite similar number of products in both groups

Imps

134%

4.282.611

1.828.412

103.803

135%

Clicks

44.122

2.047

9%

#ofproducts

1.876

expensiveproducts cheapproducts


Despite getting more traffic, the performance of the cheaper products is much more efficient

Performance of similar products within one shop Key insights

Number of orders through cheap products almost 3 times higher Higher CR resulting in CPO being ~30% lower Cheap products generate 280% more conversion

40

28

-29%

CPO Conversions

1,042

+280%

274

+61%

0,6%

CR

1,0%

expensiveproducts cheapproducts


180

55

0

5

45

15

50

40

10

There is not really a long tail: A small number of products drive the majority of the sales

Conversions per product multi brander UK Key insights

More significantly than in search, shopping conversions are driven by just a few products Concentrate on those products for account optimisation.

Top 10% products = 58% conversions

*Chart displays all products with conversions (448)

ExpensiveCheap


Cheap products are converting significantly better than expensive ones

Key insights

Only a few products are responsible for more than half of all conversions

Selling only a few products at a cheap price can make a big difference

Total conversions of similar products within one shop

Conversion Distribution

3% 38%

58%

Top 10% Middle 80% Bottom 10%

20%

Share of top 10%

converting products

80%

Cheap Expensive


Your key takeaways

Don’t overbid on expensive

products, rather consider price

changes

Price and bid management

will be merged one day

Discounting only a few select

products could be a killer strategy

$.47 $.47 -50%


SEE YOU AT THE NEXT #SMX!

Thank you!

Subscribetoourblogblog.crealytics.com

Advanced Google Shopping By Andreas Reiffen

Marketing