Understanding Model Types in SAP Analytics Cloud

What’s a model?

In SAP Analytics Cloud, models are the foundation for data exploration. Consisting mainly of measures and dimensions, models provide a framework for the data visualizations in your stories.

Stories are where you analyze your data. It’s here where you build your charts, tables, graphs, and use other data visualization tools to tell the story of your business.

What are the benefits of a model?

While you can create a story without a model, having an underlining model gives you more options. You can clean your data to allow for easier or more accurate data analysis; you can create hierarchiesso you can drill down to different levels of granularity; you can create new measures based on formulas that can be used by multiple stories; and so much more.

Put simply, here are some of the benefits of models:

  • Clean your data — separate or combine columns; fix typographical errors; ensure proper measures and dimensions are assigned; and more
  • Customize your data — geo-enrich your data so that you can include maps in your stories; create multi-level hierarchies; create custom calculations; set units and currencies; add formulas; and so on
  • Control your data — share your data with others and assign different permissions settings; tell multiple stories with one model

What are the different types of models?

Home Page Model SAP Analytics Cloud

There are two types of models in SAP Analytics Cloud:

  • Analytic models
  • Planning models

Analytic models allow you to clean your data and prepare it for story mode. Here, you can define measures and dimensions, create calculations, set up hierarchical relationships, geo-enrich your data, and more.

Planning models allow you to do everything analytic models do, but they also give you a more control with your data such as setting up budgets and forecasts, creating your own versions of model data, copying and pasting data, and using spreading, distribution, and allocations features.

Within a planning scenario, individuals within your organization can create public, private, and shared versions of data that will not overwrite a public version until approved. These versions enable individuals to change values without compromising public data, giving planners the ability to try different what-if scenarios within their data before making final decisions.

NOTE: only planning licenses with allow you to access planning models.

enable planning model SAP Analytics Cloud

In planning models, you need to define the settings for the Time, which determines the range of your data such as how far into the past and how far into the future you can plan.

planning time

To establish your time dimension, first, select the Lowest Granularity; you can choose from Year, Quarter, Month, Day. Choosing the level of granularity determines how deep you can view into your data. For example, you may have daily sales transactions, but see little value in looking at that level of granularity. You therefore may decide that ‘Month’ is the lowest granularity that you require.

Next, select the start and end dates for your model. For example, you may have ten years of data, but are only interested in looking at the last year. In this case, you can set the start date to be one year in the past.

You can also define what types of data you can manage in your model.

In planning models, there are different categories — money you actually made (Actual), money you’re going to make (Forecast), and money you are going to spend (Budget). SAP Analytics Cloud allows you to choose the granularity of each of these categories.

category planning model SAP Analytics Cloud

Take budget for example, you can select whether you see a daily budget, monthly, quarterly, or yearly.

The standard categories available in your planning models are:

  • Actual — actual values, or money you made
  • Budget – how much you are allowed to spend
  • Planning – what is the goal (financial/non-financial) you are trying to achieve
  • Forecast – what is your expectation of the financial data
  • Rolling Forecast — the range for how far back and forward you look

You can also define your model preferences.

  • Description — insert a name for your model
  • Default Currency — the currency your company uses to report their results on the corporate level
  • Data Audit — allows you to track when changes are made to your data
  • Privacy — provides a way for user to define who can access the model data
  • Currency Conversion — allows you to see the values in your model in different currencies
  • Preconverted Actuals — when importing actual data from datasources into SAP Analytics Cloud, currencies will remain as is and not be converted

Which model type is right for me?

By default, everyone who purchases SAP Analytics Cloud or uses the trial version will have access to analytic models. Planning models require a separate license. Depending on your role and business needs, you can determine which model type is most suitable to you.

model type comparison SAP Analytics Cloud

Analytic model use case

Analytic models are great for measuring success and looking for opportunities. For example, suppose you run an ice cream parlor and have the following data:

  • Flavor
  • Sales price
  • Quantity sold
  • Date
  • Etc.

You could easily tell a story with your data showing what are the most popular flavors, what are your peak periods, what are the long-term sales trends i.e. Month over Month, or Year over Year. From this analysis, you could make decisions about promotions, introducing new flavors, and even expansion opportunities.

Planning model use case

Planning models can do everything analytic models can do, but they are used more by large companies looking to do sales forecasting, budget allocation, ‘what-if’ analysis, and so on.

Keeping with the ice cream example, imagine a large manufacturer of ice cream looking to save costs and increase profit margins. They may use a value-driver tree to test different ‘what-if’ scenarios such as switching all cow milk for soy milk, increasing marketing spend in underperforming regions, or changing store hours.