SlideShare a Scribd company logo
Blaze a Trail to Predictive
Selling with Einstein Intent
Daniel Peter
Salesforce MVP
Lead Applications Engineer,
Kenandy
22x Salesforce Certified
twitter: @danieljpeter
instagram: @realdanpeter
email: dan@danpeter.com
Einstein Intent API
A Natural Language Processing classification service
Marketed primarily for lead routing and case
routing.
But could it be used to predict other things?
Sure!
Use Einstein Intent on real customer data!
Tent products
Add ons
You bought a Mastertent, you should consider buying footplates.
Recommendations
How do you know what people might want before they do?
historical data + algorithm = predict the future!
model
Tent, Sidewall => Lighting
Inflatable Tent => Flag
Pavilion, Truss => Wind Dancer
Lighting
Training Data
(independent => dependent)
independent
to be predicted
predicted
dependent
Einstein Intent API
A Natural Language Processing classification service
Examples Labels
10x15 MASTERTENT Transit Bag 66lb Footplates Wheel Set MASTERTENT Stake Kit 4 Stakes
10x10 MASTERTENT Transit Bag 6 PowerStretch Table Cover MASTERTENT Stake Kit 4 Stakes
10x15 MASTERTENT Transit Bag Sandbag MASTERTENT Stake Kit 4 Stakes
Feather Flag Large Double Sided Same Art Large Flag Pole Hardware w Carry Bag Square
Water Bag
Cross Base
Feather Flag Small Double Sided Same Art GYBE 4000 13 Wall Double Sided Same Art
Groundspike Gybe Counter w Digital Print Scoprega Rechargeable Gybe Pump Small Flag Pole
Hardware w Carry Bag Square Water Bag
Cross Base
6 Table Cover Feather Flag Medium Double Sided Same Art Groundspike Medium Flag Pole
Hardware w Carry Bag
Cross Base
10x10 MASTERTENT Transit Bag MASTERTENT LED Lighting Kit MASTERTENT Stake Kit 4
Stakes Wind Dancer LT Flag Double Sided Wind Dancer LT Hardware
66lb Footplates
10x10 MASTERTENT Transit Bag MASTERTENT Stake Kit 4 Stakes Peak Flag Double Sided
Same Art Peak Flag Hardware 10 Series Raingutter 10 x 10 Tent Connecting Clamps Wheel
Set
66lb Footplates
13x20 MASTERTENT Transit Bag MASTERTENT Stake Kit 4 Stakes Wheel Set 66lb Footplates
How to train the model with real data
Separate items into core and add-on.
Find items ordered together. These products create a bag of words
which we use as our independent variable.
Clean data: Special case customers, products which aren’t products,
product with insufficient examples, duplicate examples.
Data Prep
Predicting from within Salesforce CPQ
Predicting from within Salesforce CPQ
Predicting from within Salesforce CPQ
Architecture
TRAINING
PREDICTING
Fuzzier than using an item id primary key. This means new products which aren’t in the model
can still work.
Can use any words that might be good predictors — Such as customer demographics, season.
Data Prep is your job, and very important!
Using Natural Language Processing
for product recommendations
Code Demo
Generating JSON file from Salesforce data and
submitting it to Einstein Intent to train the model.
Sending in the bag of words from a CPQ quote to
get a prediction for an upsell.
Interpreting model accuracy metrics
Overall Model Percentage Accuracy Train / Test (0-100%)
Test is the important one.
F1 score: accuracy of the label (0-1)
Confusion Matrix: granular strengths and weakness of the model
NEW! Learning Curve
Confusion Matrix
(source:
wikipedia)
Resources
Einstein Intent API Basics on Trailhead
https://trailhead.salesforce.com/en/modules/einstein_intent_basics
Main Einstein Site (signup, docs, more)
https://einstein.ai/
Github repo with the code from this demo
https://github.com/danieljpeter/DF17IntentDemo
Blaze a Trail to Predictive Selling With Einstein Intent

More Related Content

Similar to Blaze a Trail to Predictive Selling With Einstein Intent

Decoding Data Science
Decoding Data ScienceDecoding Data Science
Decoding Data Science
Matt Fornito
 
Dances with unicorns
Dances with unicornsDances with unicorns
Dances with unicorns
EspritAgile
 
Simon Wardley at SpringOne Platform 2017
Simon Wardley at SpringOne Platform 2017Simon Wardley at SpringOne Platform 2017
Simon Wardley at SpringOne Platform 2017
VMware Tanzu
 
Data science and Machine learning Booklet
Data science and Machine learning BookletData science and Machine learning Booklet
Data science and Machine learning Booklet
Vellore Institute of Technology
 
PyData: Past, Present Future (PyData SV 2014 Keynote)
PyData: Past, Present Future (PyData SV 2014 Keynote)PyData: Past, Present Future (PyData SV 2014 Keynote)
PyData: Past, Present Future (PyData SV 2014 Keynote)
Peter Wang
 
Be a Top Notch PM Using Data Science by Farfetch Product Leader
Be a Top Notch PM Using Data Science by Farfetch Product LeaderBe a Top Notch PM Using Data Science by Farfetch Product Leader
Be a Top Notch PM Using Data Science by Farfetch Product Leader
Product School
 
Introduction to python
Introduction to pythonIntroduction to python
Introduction to python
Rajesh Rajamani
 
Data Science With Python | Python For Data Science | Python Data Science Cour...
Data Science With Python | Python For Data Science | Python Data Science Cour...Data Science With Python | Python For Data Science | Python Data Science Cour...
Data Science With Python | Python For Data Science | Python Data Science Cour...
Simplilearn
 
Solve User Problems: Data Architecture for Humans
Solve User Problems: Data Architecture for HumansSolve User Problems: Data Architecture for Humans
Solve User Problems: Data Architecture for Humans
mark madsen
 
Data Science.pptx
Data Science.pptxData Science.pptx
Data Science.pptx
TrainerAnalogicx
 
bigmartsalespridictionproject-220813050638-8e9c4c31 (1).pptx
bigmartsalespridictionproject-220813050638-8e9c4c31 (1).pptxbigmartsalespridictionproject-220813050638-8e9c4c31 (1).pptx
bigmartsalespridictionproject-220813050638-8e9c4c31 (1).pptx
Harshavardhan851231
 
Adarsh_Masekar(2GP19CS003).pptx
Adarsh_Masekar(2GP19CS003).pptxAdarsh_Masekar(2GP19CS003).pptx
Adarsh_Masekar(2GP19CS003).pptx
hkabir55
 
Detailed Report on Basics Of Pandas of Python
Detailed Report on Basics Of Pandas of PythonDetailed Report on Basics Of Pandas of Python
Detailed Report on Basics Of Pandas of Python
anushaashraf20
 
Einstein GPT in CRM.pptx
Einstein GPT in CRM.pptxEinstein GPT in CRM.pptx
Einstein GPT in CRM.pptx
Atocloud
 
The 3 Key Barriers Keeping Companies from Deploying Data Products
The 3 Key Barriers Keeping Companies from Deploying Data Products The 3 Key Barriers Keeping Companies from Deploying Data Products
The 3 Key Barriers Keeping Companies from Deploying Data Products
Dataiku
 
mm-ADT: A Virtual Machine/An Economic Machine
mm-ADT: A Virtual Machine/An Economic Machinemm-ADT: A Virtual Machine/An Economic Machine
mm-ADT: A Virtual Machine/An Economic Machine
Marko Rodriguez
 
Boosting Product Categorization with Machine Learning
Boosting Product Categorization with Machine LearningBoosting Product Categorization with Machine Learning
Boosting Product Categorization with Machine Learning
Amadeus Magrabi
 
The Most-Awaited Data Science Career Track Is Here!.pptx
The Most-Awaited Data Science Career Track Is Here!.pptxThe Most-Awaited Data Science Career Track Is Here!.pptx
The Most-Awaited Data Science Career Track Is Here!.pptx
SynergisticIT
 
Pie for ai
Pie for aiPie for ai
Pie for ai
Rabimba Karanjai
 
DATANOMIQ Qualification & Certification
DATANOMIQ Qualification & CertificationDATANOMIQ Qualification & Certification
DATANOMIQ Qualification & Certification
DATANOMIQ GmbH / AUDAVIS GmbH
 

Similar to Blaze a Trail to Predictive Selling With Einstein Intent (20)

Decoding Data Science
Decoding Data ScienceDecoding Data Science
Decoding Data Science
 
Dances with unicorns
Dances with unicornsDances with unicorns
Dances with unicorns
 
Simon Wardley at SpringOne Platform 2017
Simon Wardley at SpringOne Platform 2017Simon Wardley at SpringOne Platform 2017
Simon Wardley at SpringOne Platform 2017
 
Data science and Machine learning Booklet
Data science and Machine learning BookletData science and Machine learning Booklet
Data science and Machine learning Booklet
 
PyData: Past, Present Future (PyData SV 2014 Keynote)
PyData: Past, Present Future (PyData SV 2014 Keynote)PyData: Past, Present Future (PyData SV 2014 Keynote)
PyData: Past, Present Future (PyData SV 2014 Keynote)
 
Be a Top Notch PM Using Data Science by Farfetch Product Leader
Be a Top Notch PM Using Data Science by Farfetch Product LeaderBe a Top Notch PM Using Data Science by Farfetch Product Leader
Be a Top Notch PM Using Data Science by Farfetch Product Leader
 
Introduction to python
Introduction to pythonIntroduction to python
Introduction to python
 
Data Science With Python | Python For Data Science | Python Data Science Cour...
Data Science With Python | Python For Data Science | Python Data Science Cour...Data Science With Python | Python For Data Science | Python Data Science Cour...
Data Science With Python | Python For Data Science | Python Data Science Cour...
 
Solve User Problems: Data Architecture for Humans
Solve User Problems: Data Architecture for HumansSolve User Problems: Data Architecture for Humans
Solve User Problems: Data Architecture for Humans
 
Data Science.pptx
Data Science.pptxData Science.pptx
Data Science.pptx
 
bigmartsalespridictionproject-220813050638-8e9c4c31 (1).pptx
bigmartsalespridictionproject-220813050638-8e9c4c31 (1).pptxbigmartsalespridictionproject-220813050638-8e9c4c31 (1).pptx
bigmartsalespridictionproject-220813050638-8e9c4c31 (1).pptx
 
Adarsh_Masekar(2GP19CS003).pptx
Adarsh_Masekar(2GP19CS003).pptxAdarsh_Masekar(2GP19CS003).pptx
Adarsh_Masekar(2GP19CS003).pptx
 
Detailed Report on Basics Of Pandas of Python
Detailed Report on Basics Of Pandas of PythonDetailed Report on Basics Of Pandas of Python
Detailed Report on Basics Of Pandas of Python
 
Einstein GPT in CRM.pptx
Einstein GPT in CRM.pptxEinstein GPT in CRM.pptx
Einstein GPT in CRM.pptx
 
The 3 Key Barriers Keeping Companies from Deploying Data Products
The 3 Key Barriers Keeping Companies from Deploying Data Products The 3 Key Barriers Keeping Companies from Deploying Data Products
The 3 Key Barriers Keeping Companies from Deploying Data Products
 
mm-ADT: A Virtual Machine/An Economic Machine
mm-ADT: A Virtual Machine/An Economic Machinemm-ADT: A Virtual Machine/An Economic Machine
mm-ADT: A Virtual Machine/An Economic Machine
 
Boosting Product Categorization with Machine Learning
Boosting Product Categorization with Machine LearningBoosting Product Categorization with Machine Learning
Boosting Product Categorization with Machine Learning
 
The Most-Awaited Data Science Career Track Is Here!.pptx
The Most-Awaited Data Science Career Track Is Here!.pptxThe Most-Awaited Data Science Career Track Is Here!.pptx
The Most-Awaited Data Science Career Track Is Here!.pptx
 
Pie for ai
Pie for aiPie for ai
Pie for ai
 
DATANOMIQ Qualification & Certification
DATANOMIQ Qualification & CertificationDATANOMIQ Qualification & Certification
DATANOMIQ Qualification & Certification
 

More from Daniel Peter

Salesforce Slack Demo Cactusforce 2022
Salesforce Slack Demo Cactusforce 2022Salesforce Slack Demo Cactusforce 2022
Salesforce Slack Demo Cactusforce 2022
Daniel Peter
 
Rules-based Record Generation with Custom Metadata Types
Rules-based Record Generation with Custom Metadata Types Rules-based Record Generation with Custom Metadata Types
Rules-based Record Generation with Custom Metadata Types
Daniel Peter
 
Save Millions of Clicks! Easily migrate complex schemas from SQL to Salesforce.
Save Millions of Clicks!  Easily migrate complex schemas from SQL to Salesforce.Save Millions of Clicks!  Easily migrate complex schemas from SQL to Salesforce.
Save Millions of Clicks! Easily migrate complex schemas from SQL to Salesforce.
Daniel Peter
 
No Refresh Needed
No Refresh NeededNo Refresh Needed
No Refresh Needed
Daniel Peter
 
Using Custom Permissions to Simplify Security
Using Custom Permissions to Simplify SecurityUsing Custom Permissions to Simplify Security
Using Custom Permissions to Simplify Security
Daniel Peter
 
DF Global Gathering PuneWIT
DF Global Gathering PuneWITDF Global Gathering PuneWIT
DF Global Gathering PuneWIT
Daniel Peter
 
Dreamforce Global Gathering Bangaluru 2017
Dreamforce Global Gathering Bangaluru 2017Dreamforce Global Gathering Bangaluru 2017
Dreamforce Global Gathering Bangaluru 2017
Daniel Peter
 
Hyperbatch (LoteRapido) - Punta Dreamin' 2017
Hyperbatch (LoteRapido) - Punta Dreamin' 2017Hyperbatch (LoteRapido) - Punta Dreamin' 2017
Hyperbatch (LoteRapido) - Punta Dreamin' 2017
Daniel Peter
 
HyperBatch - Snowforce 2017
HyperBatch - Snowforce 2017HyperBatch - Snowforce 2017
HyperBatch - Snowforce 2017
Daniel Peter
 
LDS salesforce saturday
LDS  salesforce saturdayLDS  salesforce saturday
LDS salesforce saturday
Daniel Peter
 
Tree Traversal #SalesforceSaturday
Tree Traversal #SalesforceSaturdayTree Traversal #SalesforceSaturday
Tree Traversal #SalesforceSaturday
Daniel Peter
 
HyperBatch
HyperBatchHyperBatch
HyperBatch
Daniel Peter
 
Forcelandia 2016 PK Chunking
Forcelandia 2016 PK ChunkingForcelandia 2016 PK Chunking
Forcelandia 2016 PK Chunking
Daniel Peter
 
PK chunking presentation from Tahoe Dreamin' 2016
PK chunking presentation from Tahoe Dreamin' 2016PK chunking presentation from Tahoe Dreamin' 2016
PK chunking presentation from Tahoe Dreamin' 2016
Daniel Peter
 
Lightning Reports - Dreamforce 2015
Lightning Reports - Dreamforce 2015Lightning Reports - Dreamforce 2015
Lightning Reports - Dreamforce 2015
Daniel Peter
 
Callout architecture
Callout architectureCallout architecture
Callout architecture
Daniel Peter
 

More from Daniel Peter (16)

Salesforce Slack Demo Cactusforce 2022
Salesforce Slack Demo Cactusforce 2022Salesforce Slack Demo Cactusforce 2022
Salesforce Slack Demo Cactusforce 2022
 
Rules-based Record Generation with Custom Metadata Types
Rules-based Record Generation with Custom Metadata Types Rules-based Record Generation with Custom Metadata Types
Rules-based Record Generation with Custom Metadata Types
 
Save Millions of Clicks! Easily migrate complex schemas from SQL to Salesforce.
Save Millions of Clicks!  Easily migrate complex schemas from SQL to Salesforce.Save Millions of Clicks!  Easily migrate complex schemas from SQL to Salesforce.
Save Millions of Clicks! Easily migrate complex schemas from SQL to Salesforce.
 
No Refresh Needed
No Refresh NeededNo Refresh Needed
No Refresh Needed
 
Using Custom Permissions to Simplify Security
Using Custom Permissions to Simplify SecurityUsing Custom Permissions to Simplify Security
Using Custom Permissions to Simplify Security
 
DF Global Gathering PuneWIT
DF Global Gathering PuneWITDF Global Gathering PuneWIT
DF Global Gathering PuneWIT
 
Dreamforce Global Gathering Bangaluru 2017
Dreamforce Global Gathering Bangaluru 2017Dreamforce Global Gathering Bangaluru 2017
Dreamforce Global Gathering Bangaluru 2017
 
Hyperbatch (LoteRapido) - Punta Dreamin' 2017
Hyperbatch (LoteRapido) - Punta Dreamin' 2017Hyperbatch (LoteRapido) - Punta Dreamin' 2017
Hyperbatch (LoteRapido) - Punta Dreamin' 2017
 
HyperBatch - Snowforce 2017
HyperBatch - Snowforce 2017HyperBatch - Snowforce 2017
HyperBatch - Snowforce 2017
 
LDS salesforce saturday
LDS  salesforce saturdayLDS  salesforce saturday
LDS salesforce saturday
 
Tree Traversal #SalesforceSaturday
Tree Traversal #SalesforceSaturdayTree Traversal #SalesforceSaturday
Tree Traversal #SalesforceSaturday
 
HyperBatch
HyperBatchHyperBatch
HyperBatch
 
Forcelandia 2016 PK Chunking
Forcelandia 2016 PK ChunkingForcelandia 2016 PK Chunking
Forcelandia 2016 PK Chunking
 
PK chunking presentation from Tahoe Dreamin' 2016
PK chunking presentation from Tahoe Dreamin' 2016PK chunking presentation from Tahoe Dreamin' 2016
PK chunking presentation from Tahoe Dreamin' 2016
 
Lightning Reports - Dreamforce 2015
Lightning Reports - Dreamforce 2015Lightning Reports - Dreamforce 2015
Lightning Reports - Dreamforce 2015
 
Callout architecture
Callout architectureCallout architecture
Callout architecture
 

Recently uploaded

SMS API Integration in Saudi Arabia| Best SMS API Service
SMS API Integration in Saudi Arabia| Best SMS API ServiceSMS API Integration in Saudi Arabia| Best SMS API Service
SMS API Integration in Saudi Arabia| Best SMS API Service
Yara Milbes
 
Neo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j - Product Vision and Knowledge Graphs - GraphSummit ParisNeo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j
 
GraphSummit Paris - The art of the possible with Graph Technology
GraphSummit Paris - The art of the possible with Graph TechnologyGraphSummit Paris - The art of the possible with Graph Technology
GraphSummit Paris - The art of the possible with Graph Technology
Neo4j
 
What is Augmented Reality Image Tracking
What is Augmented Reality Image TrackingWhat is Augmented Reality Image Tracking
What is Augmented Reality Image Tracking
pavan998932
 
UI5con 2024 - Keynote: Latest News about UI5 and it’s Ecosystem
UI5con 2024 - Keynote: Latest News about UI5 and it’s EcosystemUI5con 2024 - Keynote: Latest News about UI5 and it’s Ecosystem
UI5con 2024 - Keynote: Latest News about UI5 and it’s Ecosystem
Peter Muessig
 
Introducing Crescat - Event Management Software for Venues, Festivals and Eve...
Introducing Crescat - Event Management Software for Venues, Festivals and Eve...Introducing Crescat - Event Management Software for Venues, Festivals and Eve...
Introducing Crescat - Event Management Software for Venues, Festivals and Eve...
Crescat
 
Oracle Database 19c New Features for DBAs and Developers.pptx
Oracle Database 19c New Features for DBAs and Developers.pptxOracle Database 19c New Features for DBAs and Developers.pptx
Oracle Database 19c New Features for DBAs and Developers.pptx
Remote DBA Services
 
E-commerce Development Services- Hornet Dynamics
E-commerce Development Services- Hornet DynamicsE-commerce Development Services- Hornet Dynamics
E-commerce Development Services- Hornet Dynamics
Hornet Dynamics
 
Oracle 23c New Features For DBAs and Developers.pptx
Oracle 23c New Features For DBAs and Developers.pptxOracle 23c New Features For DBAs and Developers.pptx
Oracle 23c New Features For DBAs and Developers.pptx
Remote DBA Services
 
Empowering Growth with Best Software Development Company in Noida - Deuglo
Empowering Growth with Best Software  Development Company in Noida - DeugloEmpowering Growth with Best Software  Development Company in Noida - Deuglo
Empowering Growth with Best Software Development Company in Noida - Deuglo
Deuglo Infosystem Pvt Ltd
 
Transform Your Communication with Cloud-Based IVR Solutions
Transform Your Communication with Cloud-Based IVR SolutionsTransform Your Communication with Cloud-Based IVR Solutions
Transform Your Communication with Cloud-Based IVR Solutions
TheSMSPoint
 
Unveiling the Advantages of Agile Software Development.pdf
Unveiling the Advantages of Agile Software Development.pdfUnveiling the Advantages of Agile Software Development.pdf
Unveiling the Advantages of Agile Software Development.pdf
brainerhub1
 
Energy consumption of Database Management - Florina Jonuzi
Energy consumption of Database Management - Florina JonuziEnergy consumption of Database Management - Florina Jonuzi
Energy consumption of Database Management - Florina Jonuzi
Green Software Development
 
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
mz5nrf0n
 
Measures in SQL (SIGMOD 2024, Santiago, Chile)
Measures in SQL (SIGMOD 2024, Santiago, Chile)Measures in SQL (SIGMOD 2024, Santiago, Chile)
Measures in SQL (SIGMOD 2024, Santiago, Chile)
Julian Hyde
 
What is Master Data Management by PiLog Group
What is Master Data Management by PiLog GroupWhat is Master Data Management by PiLog Group
What is Master Data Management by PiLog Group
aymanquadri279
 
openEuler Case Study - The Journey to Supply Chain Security
openEuler Case Study - The Journey to Supply Chain SecurityopenEuler Case Study - The Journey to Supply Chain Security
openEuler Case Study - The Journey to Supply Chain Security
Shane Coughlan
 
LORRAINE ANDREI_LEQUIGAN_HOW TO USE ZOOM
LORRAINE ANDREI_LEQUIGAN_HOW TO USE ZOOMLORRAINE ANDREI_LEQUIGAN_HOW TO USE ZOOM
LORRAINE ANDREI_LEQUIGAN_HOW TO USE ZOOM
lorraineandreiamcidl
 
Webinar On-Demand: Using Flutter for Embedded
Webinar On-Demand: Using Flutter for EmbeddedWebinar On-Demand: Using Flutter for Embedded
Webinar On-Demand: Using Flutter for Embedded
ICS
 
ALGIT - Assembly Line for Green IT - Numbers, Data, Facts
ALGIT - Assembly Line for Green IT - Numbers, Data, FactsALGIT - Assembly Line for Green IT - Numbers, Data, Facts
ALGIT - Assembly Line for Green IT - Numbers, Data, Facts
Green Software Development
 

Recently uploaded (20)

SMS API Integration in Saudi Arabia| Best SMS API Service
SMS API Integration in Saudi Arabia| Best SMS API ServiceSMS API Integration in Saudi Arabia| Best SMS API Service
SMS API Integration in Saudi Arabia| Best SMS API Service
 
Neo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j - Product Vision and Knowledge Graphs - GraphSummit ParisNeo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
Neo4j - Product Vision and Knowledge Graphs - GraphSummit Paris
 
GraphSummit Paris - The art of the possible with Graph Technology
GraphSummit Paris - The art of the possible with Graph TechnologyGraphSummit Paris - The art of the possible with Graph Technology
GraphSummit Paris - The art of the possible with Graph Technology
 
What is Augmented Reality Image Tracking
What is Augmented Reality Image TrackingWhat is Augmented Reality Image Tracking
What is Augmented Reality Image Tracking
 
UI5con 2024 - Keynote: Latest News about UI5 and it’s Ecosystem
UI5con 2024 - Keynote: Latest News about UI5 and it’s EcosystemUI5con 2024 - Keynote: Latest News about UI5 and it’s Ecosystem
UI5con 2024 - Keynote: Latest News about UI5 and it’s Ecosystem
 
Introducing Crescat - Event Management Software for Venues, Festivals and Eve...
Introducing Crescat - Event Management Software for Venues, Festivals and Eve...Introducing Crescat - Event Management Software for Venues, Festivals and Eve...
Introducing Crescat - Event Management Software for Venues, Festivals and Eve...
 
Oracle Database 19c New Features for DBAs and Developers.pptx
Oracle Database 19c New Features for DBAs and Developers.pptxOracle Database 19c New Features for DBAs and Developers.pptx
Oracle Database 19c New Features for DBAs and Developers.pptx
 
E-commerce Development Services- Hornet Dynamics
E-commerce Development Services- Hornet DynamicsE-commerce Development Services- Hornet Dynamics
E-commerce Development Services- Hornet Dynamics
 
Oracle 23c New Features For DBAs and Developers.pptx
Oracle 23c New Features For DBAs and Developers.pptxOracle 23c New Features For DBAs and Developers.pptx
Oracle 23c New Features For DBAs and Developers.pptx
 
Empowering Growth with Best Software Development Company in Noida - Deuglo
Empowering Growth with Best Software  Development Company in Noida - DeugloEmpowering Growth with Best Software  Development Company in Noida - Deuglo
Empowering Growth with Best Software Development Company in Noida - Deuglo
 
Transform Your Communication with Cloud-Based IVR Solutions
Transform Your Communication with Cloud-Based IVR SolutionsTransform Your Communication with Cloud-Based IVR Solutions
Transform Your Communication with Cloud-Based IVR Solutions
 
Unveiling the Advantages of Agile Software Development.pdf
Unveiling the Advantages of Agile Software Development.pdfUnveiling the Advantages of Agile Software Development.pdf
Unveiling the Advantages of Agile Software Development.pdf
 
Energy consumption of Database Management - Florina Jonuzi
Energy consumption of Database Management - Florina JonuziEnergy consumption of Database Management - Florina Jonuzi
Energy consumption of Database Management - Florina Jonuzi
 
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
原版定制美国纽约州立大学奥尔巴尼分校毕业证学位证书原版一模一样
 
Measures in SQL (SIGMOD 2024, Santiago, Chile)
Measures in SQL (SIGMOD 2024, Santiago, Chile)Measures in SQL (SIGMOD 2024, Santiago, Chile)
Measures in SQL (SIGMOD 2024, Santiago, Chile)
 
What is Master Data Management by PiLog Group
What is Master Data Management by PiLog GroupWhat is Master Data Management by PiLog Group
What is Master Data Management by PiLog Group
 
openEuler Case Study - The Journey to Supply Chain Security
openEuler Case Study - The Journey to Supply Chain SecurityopenEuler Case Study - The Journey to Supply Chain Security
openEuler Case Study - The Journey to Supply Chain Security
 
LORRAINE ANDREI_LEQUIGAN_HOW TO USE ZOOM
LORRAINE ANDREI_LEQUIGAN_HOW TO USE ZOOMLORRAINE ANDREI_LEQUIGAN_HOW TO USE ZOOM
LORRAINE ANDREI_LEQUIGAN_HOW TO USE ZOOM
 
Webinar On-Demand: Using Flutter for Embedded
Webinar On-Demand: Using Flutter for EmbeddedWebinar On-Demand: Using Flutter for Embedded
Webinar On-Demand: Using Flutter for Embedded
 
ALGIT - Assembly Line for Green IT - Numbers, Data, Facts
ALGIT - Assembly Line for Green IT - Numbers, Data, FactsALGIT - Assembly Line for Green IT - Numbers, Data, Facts
ALGIT - Assembly Line for Green IT - Numbers, Data, Facts
 

Blaze a Trail to Predictive Selling With Einstein Intent

  • 1. Blaze a Trail to Predictive Selling with Einstein Intent
  • 2. Daniel Peter Salesforce MVP Lead Applications Engineer, Kenandy 22x Salesforce Certified twitter: @danieljpeter instagram: @realdanpeter email: dan@danpeter.com
  • 3. Einstein Intent API A Natural Language Processing classification service Marketed primarily for lead routing and case routing. But could it be used to predict other things? Sure!
  • 4.
  • 5.
  • 6.
  • 7. Use Einstein Intent on real customer data!
  • 10. You bought a Mastertent, you should consider buying footplates. Recommendations
  • 11. How do you know what people might want before they do? historical data + algorithm = predict the future! model Tent, Sidewall => Lighting Inflatable Tent => Flag Pavilion, Truss => Wind Dancer Lighting Training Data (independent => dependent) independent to be predicted predicted dependent
  • 12. Einstein Intent API A Natural Language Processing classification service Examples Labels 10x15 MASTERTENT Transit Bag 66lb Footplates Wheel Set MASTERTENT Stake Kit 4 Stakes 10x10 MASTERTENT Transit Bag 6 PowerStretch Table Cover MASTERTENT Stake Kit 4 Stakes 10x15 MASTERTENT Transit Bag Sandbag MASTERTENT Stake Kit 4 Stakes Feather Flag Large Double Sided Same Art Large Flag Pole Hardware w Carry Bag Square Water Bag Cross Base Feather Flag Small Double Sided Same Art GYBE 4000 13 Wall Double Sided Same Art Groundspike Gybe Counter w Digital Print Scoprega Rechargeable Gybe Pump Small Flag Pole Hardware w Carry Bag Square Water Bag Cross Base 6 Table Cover Feather Flag Medium Double Sided Same Art Groundspike Medium Flag Pole Hardware w Carry Bag Cross Base 10x10 MASTERTENT Transit Bag MASTERTENT LED Lighting Kit MASTERTENT Stake Kit 4 Stakes Wind Dancer LT Flag Double Sided Wind Dancer LT Hardware 66lb Footplates 10x10 MASTERTENT Transit Bag MASTERTENT Stake Kit 4 Stakes Peak Flag Double Sided Same Art Peak Flag Hardware 10 Series Raingutter 10 x 10 Tent Connecting Clamps Wheel Set 66lb Footplates 13x20 MASTERTENT Transit Bag MASTERTENT Stake Kit 4 Stakes Wheel Set 66lb Footplates
  • 13. How to train the model with real data Separate items into core and add-on. Find items ordered together. These products create a bag of words which we use as our independent variable. Clean data: Special case customers, products which aren’t products, product with insufficient examples, duplicate examples. Data Prep
  • 14. Predicting from within Salesforce CPQ
  • 15. Predicting from within Salesforce CPQ
  • 16. Predicting from within Salesforce CPQ
  • 18. Fuzzier than using an item id primary key. This means new products which aren’t in the model can still work. Can use any words that might be good predictors — Such as customer demographics, season. Data Prep is your job, and very important! Using Natural Language Processing for product recommendations
  • 19. Code Demo Generating JSON file from Salesforce data and submitting it to Einstein Intent to train the model. Sending in the bag of words from a CPQ quote to get a prediction for an upsell.
  • 20. Interpreting model accuracy metrics Overall Model Percentage Accuracy Train / Test (0-100%) Test is the important one. F1 score: accuracy of the label (0-1) Confusion Matrix: granular strengths and weakness of the model NEW! Learning Curve
  • 22.
  • 23. Resources Einstein Intent API Basics on Trailhead https://trailhead.salesforce.com/en/modules/einstein_intent_basics Main Einstein Site (signup, docs, more) https://einstein.ai/ Github repo with the code from this demo https://github.com/danieljpeter/DF17IntentDemo