Get our case study

Smart Predict is an Augmented Analytics feature in SAP Analytics Cloud that helps you generate predictions about future events, values, and trends.

The predictive experience in SAP Analytics Cloud is simple. Smart Predict guides you step by step to create a predictive model based on historical data. The resulting model can be used to make trusted future predictions, providing you with advanced insights to guide decision-making.

Smart Predict accelerates the prediction and recommendation creation process by focusing on business outcomes.

Before using Smart Predict for the first time, it really helps to understand a few basic concepts of predictive modeling. So, here they are!

The different types of predictive scenarios

There are currently 3 types of predictive scenarios available in Smart Predict:

• Classification
• Regression
• Time Series

Defining the business problem or business question you want to address will help you choose the right type of predictive scenario.

Classification Scenario: If you’re trying to determine the likelihood of whether something will happen, you’re dealing with a classification scenario.

Ex: You want to predict membership of categories such as Yes/NO, Customer is likely to churn or not, replacing intervals within short or long for the manufacturing process, Binary (0 or 1).

Regression Scenario: If you’re trying to predict a numerical value and explore the key drivers behind it, you’re dealing with a regression scenario. 

Ex: Predict the price of an imported product based on projected transport charges and tax duties.

 Time Series Scenario: If you’re trying to forecast a future numerical value based on fluctuations over time, seasons, and other internal and external variables, you’re dealing with a time series scenario.

Predictive Scenario based on Regression:

Step 1. Loading the data

• Before we load the dataset into SAC, we cut out a few records from the original dataset. We will use this to apply our model later. I cut out 20 records of each wines from the dataset for red and white wines, each to use later. Now I have 4 datasets as follows.

A.         1580 red wines for training

B.         20 red wines for prediction

C.        4879 white wines for training

D.        20 white wines for prediction

At first create a folder in your files to save all the files at one place.

Inside your newly created folder, create a new dataset by clicking ‘+’ icon → Dataset .

Now you will be asked how you would like to begin – load data from a local file or from a data source. Since we have data in csv files, click on local data source. Select your source file. Load all 4 datasets. Your folder should look like the last screenshot.

Step 2. Training the model

1. Let us now build the predictive scenario. This is where our models to predict wine quality will be built and trained. On the main menu, click on Create >> Predictive Scenario.
2. For this problem, our predicted entity is an integer between 0 to 10. So, we will build a regression model. Select regression, give the scenario a suitable name and description.

In the Predictive Scenarios page click on ‘+’ to add a scenario.

Select a scenario that best suits to your analytic dataset.

Fill the details and click on OK.

Select the Input Dataset as show below.

• If you want edit the edit metadata, Click on Edit variable metadata (under the input dataset field) to understand how SAC has interpreted the dataset, what is the storage and type of each of the variables, what should SAC do with missing values, is the variable the key of the dataset, etc.

• We now need to define the variable we wish to predict. In our case this is quality of the wine, so click on Target and select quality.

• If there are variables in the dataset, you would like to exclude from modelling, declare them here as show below. This helps to Improve the results.

• Now that SAC knows what it needs to do, we can get started with training. Click on Train at the bottom.

SAC will take a while to train with the data set, and identify the best model for this problem statement. Then you now have results of your model as shown below.

Step 3. Understanding the results

• After you click on Train, and SAC completes the training process, it will show you 2 tabs of information.
• Overview tells you about the quality of the results
• In our cases it is 99% confident about its results.
• It also says that the error is 0.8. This means that the true value is ±0.8 from our prediction. 
• The influencer contributions explain the results
• Density of the wine and sugar understandably have the highest correlation with wine quality, followed by the other variables.

Step 4. Applying the model

• Now that our model is ready, we can apply it on the dataset we had carved out earlier.
• Click on the apply model option (icon at far right).

• At Input Dataset Variables selects the variables that you would like to see in the output.

• Fill all the details as shown below and click on Ok.

• Then go to the folder to check the predicted model and click on the model. You can see, in the predictions file, see the column at far right called Predicted Value.

Repeat the same steps to know the quality of the Red wine.

• I found the model for red wines had an error of 0.69 with a confidence of 95%. Alcohol seems to be the dominant predictor.
• I can see both my models in the predictive model’s section at the bottom. I can see status of models (trained / applied).

Create a model based on the Output Dataset of the Predictive Scenario.

Click on Menu → Create→ Model→ Click on Get data from datasource and in the Acquire Data select Dataset.

• Select the Dataset.

• Click on the Create Model on the bottom.

Or You Can directly create a story on Dataset.

Go to Create → Click on Story

• Click on Access & Explore Data as shown below.

• Click on Datasource and click on the Dataset as shown below.

• Select the Dataset.

• Click on the Story and start creating story.

• Administrators can now restrict who can create files and folders within dedicated subfolders of the Public folder. These users still won’t be able to create files and folders directly within the Public folder itself.
• Administrators can audit changes made to files’ sharing settings

These include:

1. The name of the object whose permissions were changed
2. The permissions that were granted/revoked and to which specific user/team
3. The user who made the change
4. The timestamp of the change

Note: The Activities log is now updated to show details for any changes that are made to the Sharing Settings of files stored under Browse > Files.   
                            

Stories

• You can now add or delete comments without saving your story.
• When data labels in your bar charts overlap, you can now prioritize the bars in visualization to show the front bar’s data labels only, making it easier to view your information.

• You can now use Search to Insight to search within live BW and S/4 HANA data models. You can choose a model and search for insights and can also create charts.

Planning

• In the calendar, you can now see if processes are related to each other or if they just happen to start at the same time. Also, in the Gantt view, you can now see the time values changing when you drag and drop a process.

• You can now set up a context for a Calendar task, defining all the relevant information for the task such as the Model, primary or Driving dimension, and assignees and reviewers.

• When you copy an aggregated account with calculated subaccounts, you can now see additional targets highlighted so you can paste the account value and its underlying records.

Analytics Designer

• As an application designer, you can now add geo map widgets to your analytic application and allow your application users to create stories from the widgets. Currently, geo map widgets don’t support data source-related APIs. Support for this is planned for future waves.

• Analytics Designer introduces a concept of names for widgets and other application entities to ease the handling effort at design time.

If you as an application designer want to rename some entity, various references to it must be adapted (mainly scripts, dynamic texts, and calculated measures). Previously, you had to do this work manually.

The new automatic refactoring automates this process as much as possible.

• You can now add a Search to Insight object to the scripting area and design its open dialog by defining the source of the questions to be asked. You can also specify whether the dialog is in simple mode or advanced mode and if you want to enable Clean History and Auto Search. Simple Dialog Mode, Advanced Dialog Mode are shown below.

• You can now use the BPC Planning Sequence widget to run planning sequences for BPC live data connection models. This way, the application user can run a planning sequence, which can contain one or more planning functions, as defined in the BPC system.

In the modern digital world customers with information at fingertips, making informed decisions to buy or lease a car.

Manufacturers can now connect with customers directly, transform the traditional selling practices by offering customized incentive programs. Using advanced analytics manufacturers can get deeper insights into customer sentiments and influencing factors for closing a deal with the customer

With this dynamically changing ownership models and competitive incentives landscapes, manufactures and dealers need more tools to manage incentive programs and budgets to dynamically modify programs by gaining insights into customer’s preferences to win deals.

Incentive Planning and Trade Promotion management process consists of activities aimed at successful automobile sales which can benefit from improved budgetary controls. Sales promotions and program activity, dealer incentives, promotional campaigns, and many other programs are accurately planned, budgeted, executed and audited for efficacy using our flexible and easy to customize Incentives / Trade Program Management solution.
 

Our flexible solution offers:

• Planning Budgets – total allocation of dollars
• Reconciliation – detailed breakout of budgets
• Programs – Add and manage initiatives
• Reports and Publications – chargebacks from distributors and vendors against the program 

Plan and Analyze Promotion Budgets:

 

Create templates and established budget plans templates. Plan Budget Allocations, Analyze program effectiveness, measure progress and ensure accurate execution which can be tracked by:

Model / VIN / Product

Regions / Zones / Market

Dealer / Zip/ Distributor

Program / Affiliations / Fleet
 

Planning & Management Workflows:
 
Create or Change programs, conduct what-if financial analysis, measure and review budget impacts. Review and track changes and manage approvals.:

Planning Workflow

Financial Workflow

Program Changes

Audit Trail and Compliance

Incentives Planning and Management through a set of standard and ad hoc business processes with workflow and analytics

Connect all process steps (activities) and decision events, such as program planning, new incentive model analysis, approval 

Trigger process approvals, notifications and communication workflows where the execution of one task results in automatic initiation of other processes
 

Seamless Data and System Integration

Add new data sets, integrate budgets from existing ERP system, connect other CRM systems, including external financial

Establish two-way integration with data feeds.

Keep your existing Financial tools for budgets and integrate with new incentive management solution automatically.

Add new programs and import from existing CRM systems and managed metadata.

Track & Manage Spending

Manage incentive payments to dealers and partners. Track and manage program expenditures, reconcile financial and maintain accurate reporting across,

Leverage the TEK Incentives & Promotion Management tool to gain full visibility and maximum control on your budgets and effective incentive programs.

Create Targeted Incentive Programs with Advanced Analytics, AI and Machine Learning

Create customized and targeted incentive plans by gaining insights into customer preferences, social trends, tailored promotional programs:

With access to open data, social media and APIs take a  giant  step from conventional incentive planning by deploying

• Prescriptive analytics driven approach
• AI in measuring customer trends and sentiments
• Digital Incentive Campaigns

 Thank you,
Bharat Sunkari.