#wpewebinar
James Jory & Igor Krtolica, Amazon Web Services
Anthony Burchell, WP Engine
Use Amazon.com personalization on
your WooCommerce store.
New Amazon for WordPress plugin features,
co-authored with WP Engine.
#wpewebinar
What You’ll Learn
● Why personalization is important
● What AWS Personalize for WooCommerce is
● How AWS Personalize works - demo and next steps
#wpewebinar
Ask questions as we go.
We’ll answer as many questions as we can after
the presentation
Slides and recording will be
made available shortly after
the webinar
Use the “Questions” pane
throughout the webinar
#wpewebinar
Igor Krtolica
Solutions Architect, Applied AI
Amazon Web Services
James Jory
● Dabbled in winemaking and
growing grapes
● Simulation racer in spare
time
● Wannabe BBQ pitmaster
● Worked with Personalize
during beta
● Loves good coffee
● Misses in-person
presentations
Partner Solutions Architect
Amazon Web Services
Anthony Burchell
● WordPress core
committer
● VR/AR Enthusiast
● Makes music on a
Gameboy
WordPress Developer
WP Engine Labs
#wpewebinar
Why personalization is
important
6
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
https://www.business2community.com/
marketing/30-amazing-personalization-statistics-02289
044
63%
of customers see PERSONALIZATION
AS THE STANDARD LEVEL OF
SERVICE
Market leaders are investing
in personalization to meet
customer expectations
THE CURRENT LANDSCAPE
7
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
IMPROVING BUSINESS OUTCOMES
The power of personalization
Understanding, measuring, and improving
user experiences across digital channels
Increasing time spent
engaging with products and content
ENGAGEMENT
Attracting new customers
Retaining customers in a crowded
digital environment
ACQUISITION AND RETENTION
Improving digital marketing efficiencies
Increasing average revenue per user
EFFICIENCIES AND REVENUE
Helping customers easily and quickly
discover products and content they want
Highlight new products,
content, and promotion offerings
DISCOVERABILITY
8
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved | 8
Delivering sophisticated,
unique experiences
to customers across
channels and devices
using machine learning
NOW
First feature launched for recommendations
in 1998
THEN
PIONEERING PERSONALIZATION AT AMAZON
The evolution over 20+ years
9
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
Deliver high quality recommendations
Personalize every customer touchpoint
Easily implement an ML solution at scale
Data privacy and security
Leveraging ML to
improve business
metrics
The benefits of
Amazon Personalize
10
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
Personalization in Retail
Deliver unique
homepage experience
Personalize users’
homepage with product
recommendations based on
their shopping history
Refine product
recommendations
Recommend similar items on
product detail pages
to help users’ easily find what
they are looking for
Improve
discoverability
Help users’ quickly find
relevant new products, deals,
and promotions
Relevant
product rankings
Easily re-rank relevant
product recommendations to
drive tangible
business outcomes
Enhance marketing
communication
Personalize push
notifications and marketing
emails with individualized
product recommendations
Boost upsell
and cross-sell
Combine Amazon
Personalize with business
logic to create high
quality cart upsell and
cross-sell recommendations
50%
increase in customer
engagement on
”recommended for you”
product row
Common Use
Cases
11
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
HOW IT WORKS
Amazon Personalize
Customized
personalization
API
Item metadata
(details of articles,
products, videos, etc.)
User metadata
(age, location, etc.)
User events / interactions
(views, signups,
conversions, etc.)
Inspect
data
Identify
features
Select
hyper-parame
ters
Train
models
Optimize
models
Host
models
Real-time
feature
store
Amazon Personalize
Fully managed by Amazon Personalize
12
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
User impressions
Modeling impression data, or items that are seen but not clicked
Item exploration
Balance between exploring for new items and items a user is likely to find relevant
Filtering for events
Exclude or include items to recommend based on event criteria
Filtering based on metadata
Exclude or include items to recommend based on item or user criteria
Cold start
Include recommendations for new users and new items
Key features of Amazon Personalize
13
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
Personalized online
shopping experiences
MECCA uses Amazon Personalize to extend its very high touch
and personal in store shopping experience to its online channels
such as email and web. Amazon Personalize has helped to
increase customer engagement by 65% in e-mail click-through
rates and a corresponding increase in email revenue from
personalized product recommendations.
Since integrating Personalize, we are seeing our customers respond positively to the new
recommendations with a 65% increase in e-mail click-through rates and a corresponding
increase in email revenue relating to the products recommended by Personalize. To
personalize our customer experience further, we are now extending the use of Personalize to
additional areas, including our website.
- MECCA e-Commerce & CRM Director
“
”
14
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
Providing Personalization
at scale
Pomelo Fashion used Amazon Personalize to build a
recommendation engine to personalize the shopping experience
to each customer. It now reflects user preferences on product
pages in minutes, driving sales. As a result, click-through rates
from category to product pages have increased by up to 18%,
gross revenue from category pages have gone up by 15%, and
Pomelo increased their return on investment by 400% within 1
month.
“It’s so fast that you can even see your recommendations changing during the same session.
”
15
© 2020 Amazon Web Services, Inc. or its affiliates. All rights reserved |
Amazon Personalize for Magento
Many small retailers use Magento, an open-source
ecommerce platform, to create websites or mobile
applications to sell their products online. Personalization is
key to creating high-quality ecommerce experiences, but
small businesses often lack access to resources required to
implement a scalable, sophisticated personalization
solution—especially one that is powered by machine learning
(ML).
Hoopologie, a small hula hoop supply company, has seen an
increase of 40% in sales and AOV increase of $50 per order
since integrating Amazon Personalize.
“Creating personalized recommendations for every type of user on the site would take us
hours and hours every month. It’s not affordable, and I’d rather have our limited staff help our
customers in other ways.
- Melina Rider, Founder, Hoopologie
”
#wpewebinar
How Amazon Personalize
works with WooCommerce
#wpewebinar
Amazon Personalize Architecture
Amazon
Personalize
Real-time Data
Historical Data
Web App Server-Side
App
Mobile
App
Historical
Activity
User
Attributes
Item
Catalog
S3
Bucket
#wpewebinar
Amazon Personalize Components
User / Item
Interactions
Catalog
Data
Interactions
Dataset
Solution
Version(s)
Recipe
Event
Tracker
Campaign
Personalized
Recommendation
s
Dataset
Group
Solution
User
Data
Users Dataset
Items Dataset
Filters
User
Behavior
#wpewebinar
WooCommerce and Personalize Architecture
AWS Cloud
AWS for Wordpress
Plugin
Amazon Personalize
Datasets
Train
Models
Recommendation
Function
User Events
Event Tracker
WooCommerce
User Interactions
Product
Recommendations
#wpewebinar
• 25 unique users
• 1,000 interactions
• 100 products
• Users must be known / logged in
Things to consider before getting started
Recommended Minimum Requirements
Deploying Amazon Personalize
#wpewebinar
$0.05 / Gigabyte
uploaded to Amazon
Personalize. This
includes data sent in real
time via trackers.
Data Ingestion
$0.24 / training hour for
the training hours
consumed to train a
custom model with your
data.
// First 20K TPS-hour per month
$0.20 per TPS-hour for real-time
recommendations
// Next 180K TPS-hour per month
$0.10 per TPS-hour for real-time
recommendations
// Over 200K TPS-hour per month
$0.05 per TPS-hour for real-time
recommendations
Training Time Real-Time
Recommendations
Amazon Personalize Pricing
See https://aws.amazon.com/personalize/pricing/ for
more details.
#wpewebinar
AWS for WordPress Personalize
How to use Amazon Personalize
with your WooCommerce Shop
#wpewebinar
Don’t...if you are already using
WooCommerce, you can add
machine-learning powered
personalization using the AWS
WordPress plugin.
Setting up recommended
products the hard way?
Install the AWS WordPress Plugin
Enter your AWS account ID (this is required)
Enable the Amazon Personalize integration
Train an Amazon Personalize solution
Deploy recommendations to your WooCommerce site
Set up AWS services directly from the Admin console
Personalization
AWS Wordpress Plugin Setup
#wpewebinar
AWS for WordPress
Plugin
• Set your AWS credentials and
you’re ready!
• Details on setting up AWS for
WordPress can be found here.
#wpewebinar
Setup
• “New Products in Recommendations”
▪ for when new products are
added after training is
complete.
• Manual cycle for training is
recommended for lower traffic
shops.
#wpewebinar
New Products in
Recommendations
Start Training
createPersonalizeFullAccessPolicy
createPersonalizeS3BucketAccessPolicy
createPersonalizeS3BucketAccessPolicyBatch
createS3BucketAccessPolicy
createS3BucketAccessPolicyBatch
createRole
createDatasetGroup
createInteractionsSchema
createItemsSchema
createInteractionsDataset
createItemsDataset
uploadCustomersCSV
uploadProductsCSV
uploadInteractionsCSV
createDatasetImportJob
getDatasetImportJobStatus
createSolution
getSolutionStatus
createSolutionVersion
getSolutionVersionStatus
createCampaign
#wpewebinar
Setup process
Start Training
Now
Collect interaction data
And upload when
enough is gathered
#wpewebinar
When training completes, recommendations will start
AWS Cloud
AWS for Wordpress
Plugin
Amazon Personalize
Datasets
Train
Models
Recommendation
Function
User Events
Event Tracker
WooCommerce
User Interactions
Product
Recommendations
#wpewebinar
Block Demo
#wpewebinar
Recommendations Results
• A simple way to track the success and details of
your campaigns.
• On submit of a purchase the total is added to the
campaign total.
• More to come as we refine tracking results!
#wpewebinar
Extending Personalize
in your own plugin
#wpewebinar
Use Case: ML Assisted Shopping Experience
#wpewebinar
Use Case:
Use Case: ML Assisted Shopping Experience
CRM
Chat/Support
Product
Recommendations
From userID
Or hide ID!
AWS Cloud
AWS for
Wordpress
Plugin
Amazon
Personalize
Datasets
Train
Models
Recommendation
Function
User Events
Event Tracker
WooCommerce
User Interactions
Product
Recommendations
#wpewebinar
ML assisted customer experience
• Easy to pass data from your WooCommerce shop to
other apps.
• In this example, a chat app is sending personalized
data about the shopper to the agent while hiding
the identity of the shopper.
• Personalization can be a tool to closer engage with
your customers. ML assisting Humans!
Chat Agent Dashboard
#wpewebinar
Use Case: Chat Embed Code
#wpewebinar
Using
getProducts();
#wpewebinar
Passing $recommended_products to Javascript
PHP / Plugin Root PHP File
./js/chat-attributes.js
wp_localize_script() to pass the response as `personalize_info`
#wpewebinar
Personalized Chat Demo
#wpewebinar
#wpewebinar
Timing and next steps!
Target beta mid February.
Want early access?
Tell us about your shop at
wptech@wpengine.com
#wpewebinar
Slides and recording will be made available shortly after the webinar
QUESTIONS AND ANSWERS
#wpewebinar
Resources.
Want to get involved? Join the beta!
AWS for WordPress Plugin Download
Setting up the AWS for WordPress Plugin
#wpewebinar
Join us!
https://wpeng.in/decode21/
#wpewebinar
HELP US IMPROVE
#wpewebinar
Thank You.

Use Amazon.com personalization on your WooCommerce store.

  • 1.
    #wpewebinar James Jory &Igor Krtolica, Amazon Web Services Anthony Burchell, WP Engine Use Amazon.com personalization on your WooCommerce store. New Amazon for WordPress plugin features, co-authored with WP Engine.
  • 2.
    #wpewebinar What You’ll Learn ●Why personalization is important ● What AWS Personalize for WooCommerce is ● How AWS Personalize works - demo and next steps
  • 3.
    #wpewebinar Ask questions aswe go. We’ll answer as many questions as we can after the presentation Slides and recording will be made available shortly after the webinar Use the “Questions” pane throughout the webinar
  • 4.
    #wpewebinar Igor Krtolica Solutions Architect,Applied AI Amazon Web Services James Jory ● Dabbled in winemaking and growing grapes ● Simulation racer in spare time ● Wannabe BBQ pitmaster ● Worked with Personalize during beta ● Loves good coffee ● Misses in-person presentations Partner Solutions Architect Amazon Web Services Anthony Burchell ● WordPress core committer ● VR/AR Enthusiast ● Makes music on a Gameboy WordPress Developer WP Engine Labs
  • 5.
  • 6.
    6 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | https://www.business2community.com/ marketing/30-amazing-personalization-statistics-02289 044 63% of customers see PERSONALIZATION AS THE STANDARD LEVEL OF SERVICE Market leaders are investing in personalization to meet customer expectations THE CURRENT LANDSCAPE
  • 7.
    7 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | IMPROVING BUSINESS OUTCOMES The power of personalization Understanding, measuring, and improving user experiences across digital channels Increasing time spent engaging with products and content ENGAGEMENT Attracting new customers Retaining customers in a crowded digital environment ACQUISITION AND RETENTION Improving digital marketing efficiencies Increasing average revenue per user EFFICIENCIES AND REVENUE Helping customers easily and quickly discover products and content they want Highlight new products, content, and promotion offerings DISCOVERABILITY
  • 8.
    8 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | 8 Delivering sophisticated, unique experiences to customers across channels and devices using machine learning NOW First feature launched for recommendations in 1998 THEN PIONEERING PERSONALIZATION AT AMAZON The evolution over 20+ years
  • 9.
    9 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | Deliver high quality recommendations Personalize every customer touchpoint Easily implement an ML solution at scale Data privacy and security Leveraging ML to improve business metrics The benefits of Amazon Personalize
  • 10.
    10 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | Personalization in Retail Deliver unique homepage experience Personalize users’ homepage with product recommendations based on their shopping history Refine product recommendations Recommend similar items on product detail pages to help users’ easily find what they are looking for Improve discoverability Help users’ quickly find relevant new products, deals, and promotions Relevant product rankings Easily re-rank relevant product recommendations to drive tangible business outcomes Enhance marketing communication Personalize push notifications and marketing emails with individualized product recommendations Boost upsell and cross-sell Combine Amazon Personalize with business logic to create high quality cart upsell and cross-sell recommendations 50% increase in customer engagement on ”recommended for you” product row Common Use Cases
  • 11.
    11 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | HOW IT WORKS Amazon Personalize Customized personalization API Item metadata (details of articles, products, videos, etc.) User metadata (age, location, etc.) User events / interactions (views, signups, conversions, etc.) Inspect data Identify features Select hyper-parame ters Train models Optimize models Host models Real-time feature store Amazon Personalize Fully managed by Amazon Personalize
  • 12.
    12 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | User impressions Modeling impression data, or items that are seen but not clicked Item exploration Balance between exploring for new items and items a user is likely to find relevant Filtering for events Exclude or include items to recommend based on event criteria Filtering based on metadata Exclude or include items to recommend based on item or user criteria Cold start Include recommendations for new users and new items Key features of Amazon Personalize
  • 13.
    13 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | Personalized online shopping experiences MECCA uses Amazon Personalize to extend its very high touch and personal in store shopping experience to its online channels such as email and web. Amazon Personalize has helped to increase customer engagement by 65% in e-mail click-through rates and a corresponding increase in email revenue from personalized product recommendations. Since integrating Personalize, we are seeing our customers respond positively to the new recommendations with a 65% increase in e-mail click-through rates and a corresponding increase in email revenue relating to the products recommended by Personalize. To personalize our customer experience further, we are now extending the use of Personalize to additional areas, including our website. - MECCA e-Commerce & CRM Director “ ”
  • 14.
    14 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | Providing Personalization at scale Pomelo Fashion used Amazon Personalize to build a recommendation engine to personalize the shopping experience to each customer. It now reflects user preferences on product pages in minutes, driving sales. As a result, click-through rates from category to product pages have increased by up to 18%, gross revenue from category pages have gone up by 15%, and Pomelo increased their return on investment by 400% within 1 month. “It’s so fast that you can even see your recommendations changing during the same session. ”
  • 15.
    15 © 2020 AmazonWeb Services, Inc. or its affiliates. All rights reserved | Amazon Personalize for Magento Many small retailers use Magento, an open-source ecommerce platform, to create websites or mobile applications to sell their products online. Personalization is key to creating high-quality ecommerce experiences, but small businesses often lack access to resources required to implement a scalable, sophisticated personalization solution—especially one that is powered by machine learning (ML). Hoopologie, a small hula hoop supply company, has seen an increase of 40% in sales and AOV increase of $50 per order since integrating Amazon Personalize. “Creating personalized recommendations for every type of user on the site would take us hours and hours every month. It’s not affordable, and I’d rather have our limited staff help our customers in other ways. - Melina Rider, Founder, Hoopologie ”
  • 16.
  • 17.
    #wpewebinar Amazon Personalize Architecture Amazon Personalize Real-timeData Historical Data Web App Server-Side App Mobile App Historical Activity User Attributes Item Catalog S3 Bucket
  • 18.
    #wpewebinar Amazon Personalize Components User/ Item Interactions Catalog Data Interactions Dataset Solution Version(s) Recipe Event Tracker Campaign Personalized Recommendation s Dataset Group Solution User Data Users Dataset Items Dataset Filters User Behavior
  • 19.
    #wpewebinar WooCommerce and PersonalizeArchitecture AWS Cloud AWS for Wordpress Plugin Amazon Personalize Datasets Train Models Recommendation Function User Events Event Tracker WooCommerce User Interactions Product Recommendations
  • 20.
    #wpewebinar • 25 uniqueusers • 1,000 interactions • 100 products • Users must be known / logged in Things to consider before getting started Recommended Minimum Requirements Deploying Amazon Personalize
  • 21.
    #wpewebinar $0.05 / Gigabyte uploadedto Amazon Personalize. This includes data sent in real time via trackers. Data Ingestion $0.24 / training hour for the training hours consumed to train a custom model with your data. // First 20K TPS-hour per month $0.20 per TPS-hour for real-time recommendations // Next 180K TPS-hour per month $0.10 per TPS-hour for real-time recommendations // Over 200K TPS-hour per month $0.05 per TPS-hour for real-time recommendations Training Time Real-Time Recommendations Amazon Personalize Pricing See https://aws.amazon.com/personalize/pricing/ for more details.
  • 22.
    #wpewebinar AWS for WordPressPersonalize How to use Amazon Personalize with your WooCommerce Shop
  • 23.
    #wpewebinar Don’t...if you arealready using WooCommerce, you can add machine-learning powered personalization using the AWS WordPress plugin. Setting up recommended products the hard way? Install the AWS WordPress Plugin Enter your AWS account ID (this is required) Enable the Amazon Personalize integration Train an Amazon Personalize solution Deploy recommendations to your WooCommerce site Set up AWS services directly from the Admin console Personalization AWS Wordpress Plugin Setup
  • 24.
    #wpewebinar AWS for WordPress Plugin •Set your AWS credentials and you’re ready! • Details on setting up AWS for WordPress can be found here.
  • 25.
    #wpewebinar Setup • “New Productsin Recommendations” ▪ for when new products are added after training is complete. • Manual cycle for training is recommended for lower traffic shops.
  • 26.
    #wpewebinar New Products in Recommendations StartTraining createPersonalizeFullAccessPolicy createPersonalizeS3BucketAccessPolicy createPersonalizeS3BucketAccessPolicyBatch createS3BucketAccessPolicy createS3BucketAccessPolicyBatch createRole createDatasetGroup createInteractionsSchema createItemsSchema createInteractionsDataset createItemsDataset uploadCustomersCSV uploadProductsCSV uploadInteractionsCSV createDatasetImportJob getDatasetImportJobStatus createSolution getSolutionStatus createSolutionVersion getSolutionVersionStatus createCampaign
  • 27.
    #wpewebinar Setup process Start Training Now Collectinteraction data And upload when enough is gathered
  • 28.
    #wpewebinar When training completes,recommendations will start AWS Cloud AWS for Wordpress Plugin Amazon Personalize Datasets Train Models Recommendation Function User Events Event Tracker WooCommerce User Interactions Product Recommendations
  • 29.
  • 30.
    #wpewebinar Recommendations Results • Asimple way to track the success and details of your campaigns. • On submit of a purchase the total is added to the campaign total. • More to come as we refine tracking results!
  • 31.
  • 32.
    #wpewebinar Use Case: MLAssisted Shopping Experience
  • 33.
    #wpewebinar Use Case: Use Case:ML Assisted Shopping Experience CRM Chat/Support Product Recommendations From userID Or hide ID! AWS Cloud AWS for Wordpress Plugin Amazon Personalize Datasets Train Models Recommendation Function User Events Event Tracker WooCommerce User Interactions Product Recommendations
  • 34.
    #wpewebinar ML assisted customerexperience • Easy to pass data from your WooCommerce shop to other apps. • In this example, a chat app is sending personalized data about the shopper to the agent while hiding the identity of the shopper. • Personalization can be a tool to closer engage with your customers. ML assisting Humans! Chat Agent Dashboard
  • 35.
  • 36.
  • 37.
    #wpewebinar Passing $recommended_products toJavascript PHP / Plugin Root PHP File ./js/chat-attributes.js wp_localize_script() to pass the response as `personalize_info`
  • 38.
  • 39.
  • 40.
    #wpewebinar Timing and nextsteps! Target beta mid February. Want early access? Tell us about your shop at wptech@wpengine.com
  • 41.
    #wpewebinar Slides and recordingwill be made available shortly after the webinar QUESTIONS AND ANSWERS
  • 42.
    #wpewebinar Resources. Want to getinvolved? Join the beta! AWS for WordPress Plugin Download Setting up the AWS for WordPress Plugin
  • 43.
  • 44.
  • 45.