Enhanced Ecommerce - AllWeb

0 downloads 226 Views 1MB Size Report
Nov 17, 2017 - Study about people behavior. • Research real (not artificial!) intelligence. • know statistics/analyt
OPTIMIZING CAMPAIGNS AND INVENTORY:

A WEBSHOP CASE STUDY (from the psychological perspective)

Robert Petković Analytics Lead AllWeb MK, Skopje November 17th 2017

Bruketa&Žinić&Grey

Google Analytics ROI

Analytics over time

ROI

Conversion optimization

Investment

Critical point

Instalacija

Installation

Izvještaji

Reporting

Edukacija

Education

Analize

Insights

* Brian Clifton: Successful Analytics

HIRE AN ANALYST 

Google Analytics ROI

As an analyst you have to sell yourself to the team: •

Designers (browsers, resoutions,…)



Developers (page load, browsers, …)



Content marketing (content grouping, pageviews, page value,…)



Project Managers (



CEO (Revenue, Users/Sessions, comparison,…)







Develop the culture of education

What is Google Analytics?

If only there were scientists who:



Study about people behavior



Research real (not artificial!) intelligence



know statistics/analytics/p-value •

Know how to explain data and reports



Understand personality scales



know how to talk to people



Know when and how to ask WHY





HIRE A PSYCHOLOGIST 

Simple yet useful reports

UX analysis - Current Time – Session Start Time



Current time – session start •

Time to put first item in the basket



Time from start to purchase



GA Events analysis



Specific analysis and optimization •

Internal Search



YouTube ads in app

Enhanced Ecommerce simple but useful reports

Internal Search Analysis •

Search terms lists that return 0 items •



Different QueryString for campaign traffic!

Paid traffic landing pages that returns 0 items

Simple yet useful reports

Campaign Analysis:



Calculated metrics: Cost per goal/conversion



Cost import data



Utm tags

Simple yet useful reports

Gender Analysis - example

Women use mobile devices to browse through inventory during the day, purchasing cheaper and smaller items.

Bigger and more expensive items are bought by men

Gender analysis - example Insight: • Women use mobile devices to browse through inventory during the day, purchasing cheaper and smaller items. • Men tend not to browse but to purchase, mostly on desktop. • Bigger and more expensive items are bought in the evening, using his (or mutual) desktop device Tactics: Send different (remarketing) messages to genders on different times of day using various devices: - mobile devices with informational messages - desktop devices with transactional messages

Behavioural analysis

What: Analysing behaviour, not just revenue. • Analysis of visitors’ behaviour before purchase • Analysis of product impressions, adding to cart, removing from cart,… Why: • Selecting visitors who are more likely to buy certaing products • Send remarketing messages (banners) only to them

Enhanced Ecommerce

Enhanced e-commerce – Reports

Enhanced Ecommerce

Enhanced e-commerce – Reports

Enhanced Ecommerce

Enhanced e-commerce – Calculated Metric: Buy to Cart 200%



Buy to Cart



Cart to Detail



r=0.5

+ CtD + BtC

150%

- CtD + BtC

Dječji bicikli

100%

Vrtne garniture

Role

Samostojeći hladnjaci

50% Balerinke 0% -200%

-150%

-100%

-50%

0%

50%

100%

Ženske torbice -50% Smartphones Ženske haljine Prijenosna PC

- CtD - BtC

Cestovne tenisice za trčanje

-150%

Buy to Cart ratio: {{Unique Purchases}} / ( {{Product Adds To Cart}} - {{Product Removes From Cart}} )

200%

Ljetne gume

računala -100%

-200%



150%

+ CtD - BtC

Detail to Cart

Optimizing inventory with EE - Product analysis based on users behaviour

Q1 Emotional products: Looking a lot Adding to cart Not buying

Q2 Looking a lot Adding to cart Buying

CURIOUS, Price cutters

FAST, Valuable

Q3 Less adding to cart Not buying

Q4 Rational products: Less adding to cart Buying

NOT INTERESTED

FAST, few items

Cart to Buy

Enhanced Ecommerce

Enhanced e-commerce – Category analysis •

Which products are easy to sell



Which products need some push

Enhanced Ecommerce

Ecommerce – Practical questions •

Analysis •

Which acquisition channel is the most successful one?



Which campaigns are more successful the others?



Does Facebook sell more shirts or shoes?



What’s the revenue of the last newsletter? The view rate?



Are we selling shoes size 45?



Which brand is the most successful one?



Which products people tend to add to cart but not to buy and why?



Which products or categories are easy to sell?



What does iPhone users vs. Android users buy?





Enhanced Ecommerce

Ecommerce – Practical implementation •

Remarketing •

We have men XXL shirts in stock. Let’s offer them to anyone who’s been looking at XXL products in the last xy days!



People are looking at Samsung LED TV but they’re not buying. Let’s

send a 5% discount to all the people who added it into cart but didn’t buy the TV! •

AMEX has an affiliate offer. Let’s send the newsletter to all visitors who purchased something using AMEX card first!





Who bought Nike shoes? Send them an ad for Nike shirts!



Show the dog food ad to all previous dog accessories buyers!

Offline integration •

Brick and mortar shoppers behavior



Loyalty cards integration

Future Ecommerce – digital psychology •

Behavioural analysis – discovering visitor’s intention/likelihood to purchase •

Tracking people’s mouse movements and/or page timings to find out how vulnerable they are right now to impulsive shopping



Remarketing and personalization based on personality types



IoT (Internet of Things) •

Using Heart Rate monitors (FitBit) show ad to people who were excited about the brand



Sneakers with steps counter – show new sneakers ad once these have over 1mil steps…



Artificial Intelligence - AI



Machine Learning •

Smart ad placement



Smart ad content



Automatic webpage adjustment

Enhanced Ecommerce

Analytics & Insights – Practical tips



KISS



Listen to your report recipients and their feedback



Educate, Teach



Do not over-analyse



Don’t be afraid of simple metrics / graphs / reports





BIG data Data is just like a good cold-war spy. Won’t tell you much, but if you

squeeze and torture him hard enough, may start telling some interesting stories.

THANK YOU FOR YOUR TIME! Robert Petković https://www.linkedin.com/in/ropetko/ [email protected] | [email protected]