Week 4: Augmented Analytics

Unit 1: Search to Insight

Search to insight
Introduction

SAP Analytics Cloud Search

to Insight helps information
workers ask questions of their
data in natural language and
get a result that is visualized
in a way that they easily
understand it.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 2

Search to insight
Benefits

Fast – users get data-backed answers to

common business questions in an instant without
need for support from analyst or admins

Convenient – available on web and

mobile and get answers on the go

Enterprise-Ready – works with acquired,

and live data sources like SAP HANA, BW, and
S/4HANA

Secure – it respects the data security as

set up in the source system. No additional
authorizations required
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 3
Search to insight
Using search to insight

Key search to insight features Example questions users can answer

▪ Access search to insight from the home page with search to insight
or stories ▪ Basic questions / filtered measure
▪ Search across models or within a specific ▪ Part to whole / contribution questions
model ▪ Top n / bottom n sort and rank questions
▪ View other measures and dimensions that are ▪ Change over time questions
searchable within a model
▪ Questions restricting / filtering a measure
▪ Specify a chart type when asking questions
(advanced) ▪ Measure correlation questions
▪ Measure variance / comparison questions

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 4

Search to insight
Basic questions / filtered measure

The most basic question is just asking for the ‘Sales revenue’
value for a measure

A measure can be filtered to help users focus

on sub-populations in the data

‘Sales revenue for Canada in Q3 2020

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 5

Search to insight
Part to whole / contribution questions

Allow users to compare the ‘show revenue by sales manager’

contributions of individual dimension
members (the ‘parts’) to a measure
(the ‘whole’)

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 6

Search to insight
Top n / bottom n / sort and rank questions

Help users focus on the top (or bottom) ‘show top 10 sales revenue by product’
contributor dimension members to a
measure

Recognizes expressions like

▪ Top / highest / largest / best n
▪ Lowest / least / smallest / worst n

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 7

Search to insight
Questions restricting / filtering a measure

Allow users to ask questions that restrict ‘Revenue by Sales Manager > 5M and < 10M’
the results based on a measure value

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 8

Search to insight
Change over time questions

Help users understand how a key measure or ‘show quantity sold and sales revenue by month
KPI is changing over time. for the last 2 years’

Capable of understanding time expressions like

▪ Last month / year / quarter
▪ Between (for example, Q1 and Q2)
▪ Specific dates (for example, Thurs, Nov 25)
▪ …

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 9

Search to insight
Correlation questions with multiple measures

Help users compare how 2-3 measures ‘show gross margin, discount and sales revenue
relate to and affect one another and if by manager’
so, in what manner and why

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 10

Search to insight
Variance between version questions

Help users compare their actual values ‘compare sales revenue by actual & forecast’
for a measure to other versions of the
same measure (for example, Plan,
Forecast)

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 11

Search to insight
Overview of key steps when enabling search to insight

1. 5.
2.
Identify info 3. 4. Training &
Index model
worker use Create Test rollout to end
with search
cases / synonyms search users / info
to insight
questions workers

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 12

Search to insight
Identify use cases to enable with search to insight

Before you start (not in SAC) User Group: HR Users

Use Case: Asking questions of workforce data
▪ Identify user groups / information workers that
Example NL Questions:
search to insight can enable
▪ Show headcount by department
▪ Identify key questions they want to be able to ▪ Exits over time
ask in natural language
▪ Identify the models / views needed to answer
these questions

User Group: Sales Managers

Use Case: Asking questions of sales order
data
Example NL Questions:
▪ Revenue this quarter by sales manager
▪ Gross margin by product

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 13

Search to insight
Indexing live models with search to insight

▪ Search to insight needs to index the master data in

the live model to support querying and typeahead at
search time.
▪ Only the names of measures, dimensions, and
members (that is, the master data) are added to the
index. The raw data from the live connection is not
duplicated in SAC.
▪ One index is generated per model. All users searching
this model use the same index.
▪ Search results are not returned from the index.
Search results are returned from the live connection
based on the user’s data access rights as set up
for SAP HANA, BW, and so on.
▪ The index is securely stored in your SAC tenant.

Fig 2. Indexing settings are found in Model Preferences

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 14
Search to insight
About prompts / variables when indexing live models

▪ Prompts / variables set the scope of what data is

indexed in the live model and is ultimately
searchable by end users
▪ If no variables are set, search to insight will attempt
to index all data

Best Practice
▪ Generate the index using a technical user account with admin access to the underlying data to set the
largest index scope possible
▪ Don’t set variables for optional prompts and avoid mandatory prompts if possible

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 15

Search to insight
Specifying dimension members to be indexed

Steps to complete
▪ Select which dimension members should be
indexed and made searchable

Best Practice
▪ Don’t index dimension members if users are
unlikely to use ask questions with them (for
example, IDs, lat/long…)
▪ Don’t index sensitive dimension members as the
typeahead can expose more data than the user
has access to
▪ Remove high cardinality dimensions

Note: If the dimension name IS indexed but the members are NOT indexed then users
▪ can ask questions to group by the dimension (e.g. ‘Show Revenue by Product’)
▪ can’t ask questions that filter on dimension members (e.g. ‘Show Revenue for Bikes’)
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 16
Search to insight
Create synonyms for popular measure and dimension names

Steps to complete
▪ Create synonyms / acronyms for the names of
common measures / dimensions that are
frequently referred to in questions

Best Practice
▪ Restrict access to manage synonyms to a few
people so that they are consistently managed
▪ Acronyms used for measures can also be
captured (for example, AOV = Average Order
Value)

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 17

Search to insight
Demo

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 18

Search to insight
Demo Resources

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 19

Search to insight
Resources

Help Documentation: Augmented Analytics

Detailed help guides

SAP Analytics Cloud Community Page

Up-to-date resources on SAP Analytics Cloud

SAP Analytics Cloud Augmented Analytics

Focus on augmented analytics

Ask Questions on Search to Insight

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 20

Thank you.
Contact information:

[email protected]
Follow all of SAP

www.sap.com/contactsap

© 2021 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of
SAP SE or an SAP affiliate company.
The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its
distributors contain proprietary software components of other software vendors. National product specifications may vary.
These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or
warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials.
The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty
statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional
warranty.
In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or
any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation,
and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/or platforms, directions, and
functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason
without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or
functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ
materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they
should not be relied upon in making purchasing decisions.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered
trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names
mentioned are the trademarks of their respective companies.
See www.sap.com/trademark for additional trademark information and notices.
Week 4: Augmented Analytics
Unit 2: Improve Your Data Understanding with
In-Story Augmented Insights
Improve your data understanding with in-story augmented insights
Smart insights – immediate insights in context

Get instant, contextual insights into your data

▪ Explore data independently without create or edit
rights
▪ Dynamic explanations that describe points of interest
to business users in easily understandable natural
language
▪ Understand the data underneath a story by
explaining “Why are sales high in region X compared
to other regions?”, “How has the data changed?” or
“What contributes to a calculated value?”

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 2

Improve your data understanding with in-story augmented insights
Smart insights configuration

The smart insights configuration enables

story designers to help ensure that
insights provided are more useful to
business users by:
▪ Excluding a dimension from consideration as
a top contributor in smart insights
▪ Excluding a specific dimension member or
members from consideration as a top
contributor

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 3

Improve your data understanding with in-story augmented insights
Recommended analyst actions

▪ Add a comment to stories inviting users to right-click

and interact with the insights
▪ Embed smart insight dynamic text in your most
consumed stories
▪ Configure included measures to improve insight
relevance

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 4

Improve your data understanding with in-story augmented insights
Enable smart insights for live SAP HANA connection

▪ Configure cloud connector and technical user

▪ Enable indexing on relevant connections

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 5

Improve your data understanding with in-story augmented insights
Automatic time series forecasting

Predict future values of a series

based on historical data to help
you make more confident
decisions
▪ Enable automatic forecasting on
time series charts, line charts, or
planning tables with the click of a
button
▪ Apply advanced options such as
linear regression, triple exponential
smoothing, and additional inputs
https://www.sapanalytics.cloud/resources-your-guide-to-time-series-forecasting/
▪ Forecast quality is based on MAPE
and displayed as 0-5 rating

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 6

Improve your data understanding with in-story augmented insights
Demo

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 7

Improve your data understanding with in-story augmented insights
Demo Resources

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 8

Improve your data understanding with in-story augmented insights
Resources

Help Documentation: Augmented Analytics

Detailed help guides

SAP Analytics Cloud Community Page

Up-to-date resources on SAP Analytics Cloud

SAP Analytics Cloud Augmented Analytics

Focus on augmented analytics

Ask Questions on Smart Insights

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 9

Thank you.
Contact information:

[email protected]
Follow all of SAP

www.sap.com/contactsap

© 2021 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of
SAP SE or an SAP affiliate company.
The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its
distributors contain proprietary software components of other software vendors. National product specifications may vary.
These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or
warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials.
The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty
statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional
warranty.
In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or
any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation,
and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/or platforms, directions, and
functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason
without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or
functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ
materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they
should not be relied upon in making purchasing decisions.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered
trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names
mentioned are the trademarks of their respective companies.
See www.sap.com/trademark for additional trademark information and notices.
Week 4: Augmented Analytics
Unit 3: Automated Data Exploration with Smart
Discovery
Automated data exploration with smart discovery
Viable machine learning for the business analyst

1. Bridge the knowledge gap between

machine learning and BI and allow
analysts to safely use machine learning
in BI
2. Automatically leverage data contained
in existing BI models, eliminating
manual data preparation requirements
3. Map the output cleanly to the business
question and make it simple to then
refine and improve

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 2

Automated data exploration with smart discovery
Specifying the business question

The user can refine

the question by
specifying the target
and entity.
Identifying the entity
allows the
automated machine
learning to provide
much more focused
analysis.

Smart discovery will

automatically
prepare the data to
match the question
by aggregating it at
the level of the entity.
The system guides the Business
Analyst through the question to
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC
answer process. 3
Automated data exploration with smart discovery
Overview

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 4

Automated data exploration with smart discovery
Key influencers

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 5

Automated data exploration with smart discovery
Unexpected values

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 6

Automated data exploration with smart discovery
Simulation

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 7

Automated data exploration with smart discovery
Demo

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 8

Automated data exploration with smart discovery
Demo Resources

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 9

Automated data exploration with smart discovery
Resources

Help Documentation: Augmented Analytics

Detailed help guides
SAP Analytics Cloud Community Page
Up-to-date resources on SAP Analytics Cloud
SAP Analytics Cloud Augmented Analytics
Focus on augmented analytics
Ask Questions on Smart Discovery

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 10

Thank you.
Contact information:

[email protected]
Follow all of SAP

www.sap.com/contactsap

© 2021 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of
SAP SE or an SAP affiliate company.
The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its
distributors contain proprietary software components of other software vendors. National product specifications may vary.
These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or
warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials.
The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty
statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional
warranty.
In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or
any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation,
and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/or platforms, directions, and
functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason
without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or
functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ
materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they
should not be relied upon in making purchasing decisions.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered
trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names
mentioned are the trademarks of their respective companies.
See www.sap.com/trademark for additional trademark information and notices.
Week 4: Augmented Analytics
Unit 4: Prepare the Right Data for Predictive
Scenarios
Prepare the right data for predictive scenarios
Understand predictive scenarios

Predictive for Planners and Business Analysts

▪ Bring the power of automated machine learning to end users
▪ Generate with the push of a button​ and without the intervention
of a data scientist:
− revenue or cost forecasts (time series)
− business outcomes (regression)
− anticipate customer or employee behavior (classification)
▪ Produce transparent, understandable, and trustworthy
predictions that can be easily integrated in stories

Totally Integrated in Analytical Workflows

▪ Predictive planning: create time series forecasts at large scale
to support planning needs
▪ Acquired datasets can be used as a data foundation for
predictive scenarios, supported by dataset wrangling
capabilities
▪ Live datasets: Predictive scenarios can use SAP HANA tables
and SQL views as data sources and delegate predictive
processing to SAP HANA on-premise systems
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 2
Prepare the right data for predictive scenarios
Supported data models: planning models

Planning
Planning Models Models
Source data Source data
(actuals) (actuals)
▪ Planning-enabled models can be used as data sources
for time series forecasting scenarios. They cannot be
used in classification and regression scenarios.
▪ Historical data is acquired into planning models from the Predictive Planning
source systems. Data Time Series
Sources Forecasting
▪ All data sources that can be leveraged by planning Scenarios
models are supported (SAP BW, SAP S/4HANA, …),
except SAP Business Planning and Consolidation (BPC). Planned Predictive
data forecasts
▪ Any version (public or private) can be used as a source of
data to create the time series forecasting models. Stories
▪ Predictive forecasts can be written back to private
versions, promoted to public versions, and consumed
into stories to help planners.
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 3
Prepare the right data for predictive scenarios
Gathering the right data in planning models

How much data history is required to get the best forecasts?

▪ Make sure you have enough data and that you have quality data. The rule of
thumb is 5 to 1 as the ratio between actuals and predictive forecasts.
▪ Example: If you want to forecast one year ahead at a monthly level
(12 months), you’ll need to retrieve ideally 5 years of data history into your
planning models (60 months).
▪ If you only have 3 or 4 times more actuals than predictive forecasts, you can
still forecast, but in a less optimal way.

Should I include influencing factors?

▪ Think about calendar events, economic conditions, sanitary conditions,
budget… Multiple factors can have an influence on the predictive forecasts.
▪ Make sure to include these influencing factors as part of the planning model.
The values should be provided for actual periods (past) but also for the
forecasted periods (future).
▪ Tip: seasonality-related factors (MonthOfYear etc) are already created
automatically.
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 4
Prepare the right data for predictive scenarios
Source data for time series forecasting

▪ In its most basic form, a time series

corresponds to the evolution of a given
indicator over time.
▪ This example shows the evolution of
the European Union Population
(indicator) over time (year).
▪ Historical data corresponding to the
years 1970 to 2020 is used to forecast
the period 2021-2030.

Source data: The World Bank: SP.POP.TOTL (CC BY 4.0)

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 5
Prepare the right data for predictive scenarios
Source data for time series forecasting

▪ Some factors can influence the evolution of our main

indicator. These are called influencers.
▪ While it’s possible to generate predictive forecasts
without using influencers, in some cases influencers
can help increase forecasting accuracy or help
perform what-if simulation scenarios.
▪ As an example, ice cream sales could be influenced
by the outside temperature on the corresponding day.
▪ In order to give the predictive model a chance to
leverage the explanatory power of these influencers,
they need to be included in the source data.

Source data: Math is Fun

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 6
Prepare the right data for predictive scenarios
Source data for time series forecasting

▪ In some cases, the time series will vary depending on

specific entities (the entity is the unique identifier of the
time series – location in this example).
▪ In this example, we see the evolution of the percentage
of the population that is fully vaccinated in France and
Germany in early October 2021.
▪ When gathering the data, and to create one time series
forecasting model per entity, dates and values will be
repeated across the different entities.
▪ Entities can be defined using multiple fields. For instance,
entities could be defined by combining countries
(for example, France) and population age ranges
(for example, [18;25]) to forecast the progress of
vaccination per age ranges in different countries.

Source data: ourworldindata.org (full citation) (CC BY-SA 3.0 AU)

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 7
Prepare the right data for predictive scenarios
Source data for time series forecasting

When dealing with the most advanced use cases, all Planning
Models
relevant information can be included in planning models:
Dates
▪ Dates
Properties Versions
▪ Values to forecast (target variable)
Accounts Influencers
▪ Influencers
▪ Entities Measures Hierarchies

The tight integration between time series forecasting and

Dimensions Currencies
planning facilitates this:
▪ Planning models natively handle dates with different time
granularities.
Source data Predictive
▪ Values and influencers can be modeled as accounts / measures. (actuals) forecasts
▪ Entities can result from the combinations of dimension, property,
and hierarchy values. These values can be also be filtered
upfront.

All these concepts can be leveraged by time series Time Series

forecasting scenarios. Forecasting
Scenarios
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 8
Prepare the right data for predictive scenarios
Supported data models: datasets

Datasets

▪ Datasets are simple data collections, organized as Datasets

tables. Source data Source data
▪ Data can be acquired from a variety of data sources,
and it’s also possible to create live datasets on top of Delivering predictions
SAP HANA on-premise systems. into stories with
▪ Datasets contain the information required to create datasets
Data Predictive
classification and regression scenarios (data types, Sources Scenarios
statistical types…) and to deliver predictions.
▪ Datasets can also be used as data sources to create Predictions
time series forecasting scenarios.
▪ Predictive scenarios deliver predictions into datasets,
and datasets can in turn be used in stories.
Stories
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 9
Prepare the right data for predictive scenarios
Using datasets for time series
Date Target Variable Date Target Variable Forecasted Sales Confidence Intervals

Predictive
scenario
–
Time series
forecasting

Training dataset (used to create Output dataset (containing the predicted forecasts)
the predictive model)

Note: it’s also possible to use entities and influencers in datasets.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 10
Prepare the right data for predictive scenarios
Using datasets for classification
▪ When creating and using classification scenarios, we need two datasets:
− The first one is used to create the classification model – the training dataset. It uses historical data.
− The second one is used to apply the classification model – the application dataset. It’s the data we want to
predict.
▪ The classification scenario will generate predictions into a third dataset – this is the one used in stories.

1. Trains using
Classification
Scenario
2. Applies using

Training dataset – Employees that left or stayed at the end of 2021 3. Generates

Application dataset – Employees as of beginning of 2022 – we want to predict if they might leave Generated dataset – end of 2022 predictions
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 11
Prepare the right data for predictive scenarios
Using datasets for regression
▪ When creating and using regression scenarios, we need two datasets:
− The first one is used to create the regression model – the training dataset. It uses historical data.
− The second one is used to apply the regression model – the application dataset. It’s the data we want to
predict.
▪ The regression scenario will generate predictions into a third dataset – this is the one used in stories.

1. Trains using
Regression
Scenario

3. Generates
Training dataset – Used cars with their selling prices

2. Applies using

Application dataset – Used cars for sales today Generated dataset – price predictions
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 12
Example source: https://blogs.sap.com/2020/03/31/hands-on-tutorial-sap-smart-predict-used-car-pricing/
Prepare the right data for predictive scenarios
Recommendations when creating datasets for classification / regression scenarios

▪ Gather enough variables in your training dataset

The more your input variables relate to the target variable, the more accurate your predictive model will be. This is
evaluated by the predictive power (classification) or RMSE (regression).
The first step is to gather all the base information that relates to the entity (employee characteristics for employee churn,
car characteristics to predict car price sales, and so on). Beyond the base information, think how you can enrich the base
dataset with derived information (feature engineering). Keep in mind that personal information is protected and focus on
building trustworthy and ethical models.
▪ Gather enough observations in your training dataset
You want to make sure you have enough observations to build a robust enough model (as evaluated by the prediction
confidence). In classification, the rule of thumb is to have at least several hundred examples of the less represented class.
Regression models will also require a minimum of several hundred rows of data.
You should also be careful when collecting data not to introduce bias in the data (selecting a specific employee population
or specific time periods). Your data should reflect the business question you want to see answered.
▪ Don’t mess with time!
There is usually a delay between the moment we collect input variables (employee or car characteristics) and the target
variable (employee left, car was sold at a given price). When building your datasets, don’t mess with time!

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 13

Prepare the right data for predictive scenarios
Demo

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 14

Prepare the right data for predictive scenarios
Resources

Help Documentation: Augmented Analytics

Detailed help guides

SAP Analytics Cloud Community Page

Up-to-date resources on SAP Analytics Cloud

SAP Analytics Cloud Augmented Analytics

Focus on augmented analytics

Ask Questions on Augmented Analytics

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 15

Thank you.
Contact information:

[email protected]
Follow all of SAP

www.sap.com/contactsap

© 2021 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of
SAP SE or an SAP affiliate company.
The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its
distributors contain proprietary software components of other software vendors. National product specifications may vary.
These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or
warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials.
The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty
statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional
warranty.
In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or
any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation,
and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/or platforms, directions, and
functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason
without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or
functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ
materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they
should not be relied upon in making purchasing decisions.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered
trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names
mentioned are the trademarks of their respective companies.
See www.sap.com/trademark for additional trademark information and notices.
Week 4: Augmented Analytics
Unit 5: Predict Future Outcomes with Smart Predict
Forecasting
Predict future outcomes with Smart Predict Forecasting
Understanding data sources for time series forecasting

Time series forecasting scenarios can be created using two types of data sources:
▪ Planning-enabled models
▪ Datasets (either acquired or live on SAP HANA on-premise systems)
The major differences are listed below.

Planning Models Datasets

▪ Creating a planning model requires a Planning ▪ Creating a dataset can be done using any BI or
Professional license. Planning license.
▪ Planning models are semantically rich. Metadata can ▪ Datasets are simple collections of data. The data is
help improve forecasting, for instance to predict at a always handled at the level provided in the dataset and
higher-level of aggregation. cannot be aggregated.
▪ It’s possible to write the predictive forecasts directly into ▪ Separate datasets containing the predictive forecasts
the planning models (in private versions). are created. They need to be integrated with models
▪ Data preparation can be automated (data refresh and stories.
schedules, data actions, multi-actions). ▪ Lightweight data preparation is offered, and data
refreshes should be handled manually.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 2

Predict future outcomes with Smart Predict Forecasting
Creating predictive scenarios

▪ Predictive scenarios can be

created and retrieved:
– In the Files area
– In the Predictive Scenarios area
▪ A predictive scenario answers a
dedicated question and can
contain multiple predictive models
(experiments).
▪ Predictive scenarios can be
secured, moved, shared… like
other files in SAP Analytics
Cloud.
▪ Note: predictive scenarios cannot
be copied inside a tenant. They
cannot be transferred across
tenants.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 3

Predict future outcomes with Smart Predict Forecasting
Creating time series forecasting models

Time Series Data Source

Time Series Data Source
▪ Dataset
▪ Planning model
▪ Version
Forecasting using planning models

Forecasting using datasets

Predictive Goal
Predictive Goal ▪ Target
▪ Target ▪ Date variable
▪ Date variable ▪ Number of forecast
▪ Number of forecast periods
periods ▪ Entities help create
▪ Entities help create multiple models on
multiple models on dimension combinations
dimension combinations
▪ Entities can be filtered to
focus the forecasting
Predictive Model
scope Training
▪ The data history
Predictive Model Training window that is used
▪ The data history window ▪ Influencers
that is used
▪ Influencers

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 4

Predict future outcomes with Smart Predict Forecasting
Evaluating time series forecasting models

The expected MAPE (Mean Absolute Percentage

Error) indicates how accurate a predictive model is.
When creating the predictive model, an estimation
of the expected MAPE is done.
The expected MAPE is averaged across all entities
(see 1).
You also get an expected MAPE per entity (see 2) –
the model that predicts revenue for E-bikes in China
has an expected MAPE of 11.44%. 1

Finally, you get the record count per entity (see 3).
This is useful if you have sparse time series.
2 3
You can also evaluate the accuracy of your
predictive models in stories.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 5

Predict future outcomes with Smart Predict Forecasting
Evaluating time series forecasting models

The Forecast report includes:

▪ The actuals
▪ The predictive forecasts
▪ The outliers
▪ The confidence intervals (error
min/max)
It helps you compare the model
and the actuals for the past
periods.
Tip: if your training window
stops before the last known
observation (using a “travel back
in time” approach), you’ll be able
to compare predictions and
known actuals in this report.
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 6
Predict future outcomes with Smart Predict Forecasting
Evaluating time series forecasting models

Smart Predict proposes a white-

box approach for predictive
modeling.
The time series forecasting model is
presented on the Explanation tab.
The best model to forecast revenue
for E-bikes in the Pacific U.S. region
is made of:
▪ a non-linear trend (in red)
▪ and quarterly cycles (in orange)
The relative impact of each
component is provided.
Actuals appear in light blue and can
be compared to the time series
forecasting model.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 7

Predict future outcomes with Smart Predict Forecasting
Using predictive forecasts in stories

The use of planning models or datasets as data sources changes the way you integrate predictive
forecasts in stories.

Planning Models Datasets

Predictive Consumed Predictive Consumed

forecasts into forecasts into

Time Series Private Stories Time Series Datasets Stories

Forecasting versions Forecasting
Scenarios Scenarios

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 8

Predict future outcomes with Smart Predict Forecasting
Refreshing predictive forecasts

The use of planning models or datasets as data sources changes the way you refresh predictive
forecasts in stories.
Planning Models Datasets
Planning
Retraining the Datasets Retraining the
Models
Data model on the Data model on the
refreshes latest actuals refreshes latest actuals
(scheduled) (manual)* (manual) (manual)

Delivering predictions
into stories with
Predictive Planning
datasets
Data Time Series Data Predictive
Sources Forecasting Sources Scenarios
Scenarios
Delivering the
Export data Writing back predictive
refreshes the predictive forecasts to
(when forecasts datasets
applicable) (manual)* (manual)
Stories
Stories Datasets
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC *automation is planned as part of the product roadmap 9
Predict future outcomes with Smart Predict Forecasting
Demo

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 10

Predict future outcomes with Smart Predict Forecasting
Resources

Help Documentation: Augmented Analytics

Detailed help guides
SAP Analytics Cloud Community Page
Up-to-date resources on SAP Analytics Cloud
SAP Analytics Cloud Augmented Analytics
Focus on augmented analytics
Predictive Planning Blogs (SAP Community)
Retrieve all predictive planning blogs
Hands-on Tutorial
Experience predictive planning!
Infuse Intelligence into Finance (Roche session during SAPPHIRENOW 2021)
Hear the Roche Predictive Planning story
2021 is the Year of Predictive Planning
More on why predictive is key for FP&A professionals
Best Practices
Ask Questions on Augmented Analytics
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 11
Thank you.
Contact information:

[email protected]
Follow all of SAP

www.sap.com/contactsap

© 2021 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of
SAP SE or an SAP affiliate company.
The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its
distributors contain proprietary software components of other software vendors. National product specifications may vary.
These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or
warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials.
The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty
statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional
warranty.
In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or
any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation,
and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/or platforms, directions, and
functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason
without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or
functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ
materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they
should not be relied upon in making purchasing decisions.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered
trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names
mentioned are the trademarks of their respective companies.
See www.sap.com/trademark for additional trademark information and notices.
Week 4: Augmented Analytics
Unit 6: Predict Event Probabilities with Smart Predict
Classification
Predict event probabilities with Smart Predict Classification
Understand classification

What is it?
▪ A classification model helps to classify observations based on historical data.
▪ Using classification, you can predict the probability that a specific event will
happen.

Which use cases?

▪ These questions can be addressed by classification.
Who is likely to <event> based on <context>?

Examples: Classification
▪ Who is likely to win a horse race based on their recent form?
▪ Who is likely to buy this product based on their buying habits?
▪ Who is likely to leave a company based on their employee profile?
▪ Who is likely to stop using a service based on their usage and satisfaction?

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 2

Predict event probabilities with Smart Predict Classification
Supported data models: datasets

Datasets

▪ Datasets are simple data collections, organized as Datasets

tables. Source data Source data
▪ Data can be acquired from a variety of data sources
and it’s also possible to create live datasets on top of Delivering predictions
SAP HANA on-premise systems. into stories with
▪ Datasets contain the information required to create datasets
Data Predictive
classification and regression scenarios and deliver Sources Scenarios
predictions (data types, statistical types…).
▪ Datasets can also be used as data sources to create Predictions
time series forecasting scenarios.
▪ Predictive scenarios deliver predictions into datasets,
and datasets can in turn be used in stories.
Stories Datasets

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 3

Predict event probabilities with Smart Predict Classification
Leverage classification results in stories

Valerie is responsible for an online shopping portal that sells beauty and pharma products. The collected data
describes usage and activities on the portal. She wants to predict which customers might reduce their
activities.

Insights from the classification model can

be included in stories. Here, the
probability for each customer is displayed.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 4

Predict event probabilities with Smart Predict Classification
Leverage classification results in stories

Prediction explanations provide increased

business context for each predicted
probability.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 5

Predict event probabilities with Smart Predict Classification
Create a classification scenario

1. Create a classification scenario

2. Define the settings
– Training data source
– Predictive goal: the variable to predict
– Input variables to include and to exclude
3. Train the predictive model
4. Explore the predictive model
– Overview (performance)
– Influencer contributions
– Confusion matrix and profit simulation
5. Apply the predictive model to new observations
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 6
Predict event probabilities with Smart Predict Classification
Evaluate the accuracy of the model

Two indicators help evaluate the model:

▪ The predictive power evaluates the
accuracy of the classification model
(model quality). It primarily depends on
input variables.
▪ The prediction confidence measures the
ability of the classification model to predict
with the same reliability on new cases
(model robustness). It primarily depends
on observations.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 7

Predict event probabilities with Smart Predict Classification
Influencer contributions

▪ Influencers contributions show

the relative importance of each
variable in the model.
▪ The contribution of the variable
Delta Revenue in the model
equals 13.58%.
▪ The bar chart Grouped Category
Influence presents the details of
this influence per grouped
categories.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 8

Predict event probabilities with Smart Predict Classification
Confusion matrix and profit simulation

▪ The confusion matrix shows the performance of the

classification model by comparing predicted values
with actual values, based on the contacted
population.
▪ The profit simulation associates a profit/cost with the
positive categories. In Valerie’s example, this would
be the cost to prevent customers from leaving and
the benefit of keeping them engaged in the portal.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 9

Predict event probabilities with Smart Predict Classification
Apply a classification model to new observations

The following outputs can be saved in a dataset:

▪ The input variables from the application dataset you
want to keep in the generated dataset (Replicated
Columns)
▪ Predicted Category is the predicted value
▪ Prediction Probability is the predicted probability for
the Predicted Category
▪ Prediction Explanations is the list of variables that
influence the target for a given observation the most
with their specific influence.
▪ There are other outputs that can be selected.

Once the dataset is created, it can be integrated

within models and stories.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 10

Predict event probabilities with Smart Predict Classification
Demo

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 11

Predict event probabilities with Smart Predict Classification
Resources

Help Documentation: Augmented Analytics

Detailed help guides
SAP Analytics Cloud Community Page
Up-to-date resources on SAP Analytics Cloud
SAP Analytics Cloud Augmented Analytics
Focus on augmented analytics
Ask Questions on Smart Predict

The following blogs are helpful if you want to explore classification in detail:
▪ Classification in SAP Analytics Cloud in Detail
▪ Prediction Explanations For Classification Predictive Models in SAP Analytics Cloud
▪ Hands-On Tutorial SAP Smart Predict, Customer Churn Analysis for online retail
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 12
Thank you.
Contact information:

[email protected]
Follow all of SAP

www.sap.com/contactsap

© 2021 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of
SAP SE or an SAP affiliate company.
The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its
distributors contain proprietary software components of other software vendors. National product specifications may vary.
These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or
warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials.
The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty
statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional
warranty.
In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or
any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation,
and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/or platforms, directions, and
functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason
without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or
functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ
materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they
should not be relied upon in making purchasing decisions.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered
trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names
mentioned are the trademarks of their respective companies.
See www.sap.com/trademark for additional trademark information and notices.
Week 4: Augmented Analytics
Unit 7: Predict Key Indicators with Smart Predict
Regression
Predict key indicators with Smart Predict Regression
Understand regression

What is it?
▪ A regression model is used to estimate the value of a measure.
▪ Using a regression model, you can predict the most likely value for a
given observation.

Which use cases?

▪ These questions can be addressed by regression.

What is the <measure value>?

How many/How much <measure value>?

Examples:
▪ What is the delay in each customer paying their invoices?
▪ How many products will a customer buy in the next quarter?
▪ How much will a customer spend on my e-commerce on average?

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 2

Predict key indicators with Smart Predict Regression
Supported data models: datasets

Datasets

▪ Datasets are simple data collections, organized as Datasets

tables. Source data Source data
▪ Data can be acquired from a variety of data sources
and it’s also possible to create live datasets on top of Delivering predictions
SAP HANA on-premise systems. into stories with
▪ Datasets contain the information required to create datasets
Data Predictive
classification and regression scenarios and deliver Sources Scenarios
predictions (data types, statistical types…).
▪ Datasets can also be used as data sources to create Predictions
time series forecasting scenarios.
▪ Predictive scenarios deliver predictions into datasets,
and datasets can in turn be used in stories.
Stories Datasets

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 3

Predict key indicators with Smart Predict Regression
Include and use regression outputs in a story

Dan is a sales manager. He wants to have a dashboard that presents estimated sales revenue for the next 6
months. The goal is to get predicted insights of the expected revenue and drill down at the customer and
product level.

Insights from regression can be displayed in

stories. Here we see how the predicted
opportunity value for a selected customer is
influenced by the opportunity characteristics.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 4

Predict key indicators with Smart Predict Regression
Create a regression scenario

1. Create a regression scenario

2. Define the settings
– Training data source
– Predictive goal: the variable to predict
– Input variables to include and to exclude
3. Train the predictive model
4. Explore the predictive model
– Overview (performance)
– Influencer contributions
5. Apply the predictive model to new observations
© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 5
Predict key indicators with Smart Predict Regression
Evaluate the accuracy of the model

Two indicators help evaluate the model:

▪ The root mean square error (RMSE) measures the quality of the regression model.
The smaller the RMSE, the more accurate the model. It primarily depends on input variables.
▪ The prediction confidence measures the ability of the regression model to predict with the same reliability on
new cases (model robustness). It primarily depends on observations.

The regression model relies on a gradient boosting technique, similar to the classification model.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 6

Predict key indicators with Smart Predict Regression
Influencer contributions

▪ Influencer contributions show

the relative importance of
each variable in the model.
▪ The contribution of the
variable Customer_Status to
the model equals 14.77%.
▪ The bar chart Grouped
Category Influence presents
the details of this influence
per grouped categories.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 7

Predict key indicators with Smart Predict Regression
Apply a regression model to new observations

The following outputs can be saved in a dataset:

▪ The input variables from the application dataset you
want to keep in the generated dataset (Replicated
Columns)
▪ Predicted Value is the value predicted for the target
variable
▪ Prediction Explanations is the list of variables that
influence the target with their influence the most
▪ There are other outputs that can be selected.

Once the dataset is created, it can be integrated within

models and stories.

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 8

Predict key indicators with Smart Predict Regression
Demo

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 9

Predict key indicators with Smart Predict Regression
Resources

Help Documentation: Augmented Analytics

Detailed help guides
SAP Analytics Cloud Community Page
Up-to-date resources on SAP Analytics Cloud
SAP Analytics Cloud Augmented Analytics
Focus on augmented analytics
Ask Questions on Smart Predict

The following blogs are helpful if you want to explore regression in detail:
▪ Regression in SAP Analytics Cloud in Detail
▪ Prediction Explanations for Regression Models in SAP Analytics Cloud
▪ SAC Smart Predict – What goes on under the hood

© 2021 SAP SE or an SAP affiliate company. All rights reserved. ǀ PUBLIC 10

Thank you.
Contact information:

[email protected]
Follow all of SAP

www.sap.com/contactsap

© 2021 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of
SAP SE or an SAP affiliate company.
The information contained herein may be changed without prior notice. Some software products marketed by SAP SE and its
distributors contain proprietary software components of other software vendors. National product specifications may vary.
These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or
warranty of any kind, and SAP or its affiliated companies shall not be liable for errors or omissions with respect to the materials.
The only warranties for SAP or SAP affiliate company products and services are those that are set forth in the express warranty
statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional
warranty.
In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or
any related presentation, or to develop or release any functionality mentioned therein. This document, or any related presentation,
and SAP SE’s or its affiliated companies’ strategy and possible future developments, products, and/or platforms, directions, and
functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time for any reason
without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or
functionality. All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ
materially from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, and they
should not be relied upon in making purchasing decisions.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered
trademarks of SAP SE (or an SAP affiliate company) in Germany and other countries. All other product and service names
mentioned are the trademarks of their respective companies.
See www.sap.com/trademark for additional trademark information and notices.