December 12th, 2019

As business analytics coaches it is our passion to talk about latest innovations in the financial planning and analysis space. Today we're going to talk about some of your pain points and discuss some ways Lodestar Solutions can help you redesign your TM1 planning system.

TM1 Model Issues

We all have models or work with models in the past that we either inherited or have built a long time ago. Overtime, TM1 models can become sluggish and start to perform poorly. Some of this is due to poor design and or accumulation of large amounts of data. A lot of this also has to do with not fully understanding how multidimensional models work.

If you have a relational mindset you will take a little bit of retraining of the mind in order to adjust to multidimensional models. People who come from a relational background and built multi-dimensional models in a lot of cases added dimensions that were not necessary in the cube. These dimensions could have easily been attributes off a dimension and not an actual dimension within the cube. As you may know, adding more dimensions to a cube introduces complexity and likely makes the cube sparse and subsequently will perform poorly. As time goes on and more data is added to these cubes they start performing poorly and you end up with a sluggish environment. Add feeders to the problem and in a lot of cases data points are overfed and this results in higher memory consumption overtime. This all snowballs into problems that some of you may be familiar with, such as complaints from end users. Lengthy maintenance cycles in the shape of drawn out data loads, reconciliations and poorly performing reports.

So now that we have outlined all your pain points, how will Lodestar help you fix your TM1 Planning System? In the latest version of TM1 which is called Planning Analytics, IBM has introduced certain features that address the problems that we have outlined. We're going to attempt to provide certain ideas on how you can improve and enhance your current planning and forecasting environment by upgrading to the latest version of Planning Analytics. By putting in a little bit of work up front, you can not only fix the issues outlined above you can also take your planning and forecasting to the next level by employing latest technology and utilizing our expertise to create models that perform better and use the latest industry standards in creating a predictive forecast that ends up in a solid business plan. Here are some methods we can use to do it.

Ideas to Enhance and Improve your TM1 Environment

In Planning Analytics there is a feature called hierarchies. This is probably one of the biggest enhancements to TM1 to date. Dimensions in Planning Analytics are a collection of hierarchies. You can have unlimited amount of hierarchies in Planning Analytics. You can not only use hierarchies as an alternate consolidation of elements but you can also use hierarchies as virtual dimensions. What do we mean by virtual dimension you may ask? It's just a dimension that is created either temporarily or permanently and can be used in a cube for analysis. Any attribute within a dimension can be converted into a hierarchy. If you have cubes with dimensions that only have a few elements in them, you can actually remove those dimensions from your cube and use them as attributes for another dimension. Subsequently these attributes can be used as a virtual dimension and now you can do the same analysis as previously done but with a dense cube that uses less memory. It also gives you the flexibility to include or exclude this dimension at will.

Opportunity to Redesign

Using hierarchies as a virtual dimension will require a redesign of cubes and will provide the opportunity to look at the environment and do a deep health check. This may possibly include, removing redundancies in the form of cubes, dimensions or elements. Combining cubes and reducing the memory footprint. A health check will also provide the opportunity to create new models, possibly create a predictive model that provides a better forecast using statistical analysis, or maybe use algorithmic artificial intelligence to create models that help in decision optimization.

What Else Do You Get?

With Planning Analytics you get Workspace (PAW) which is a state of the art modeling, dashboarding and reporting environment. This is the new platform for all your needs. You do not need Architect to create models anymore. You also get the updated version of Perspectives which is called Planning Analytics for Excel (PAX). In PAX you can do everything that you are able to in Perspectives and more. Staring a new report takes seconds and within minutes you can have something presentable for any audience. All cubes are available at your fingertips and simple drag and drop capabilities make things a breeze. There is a huge improvement from years ago when the only reporting was Perspectives, TM1 Web or Cognos BI. Now you have modeling, reporting and dashboarding all in on space.

Here are some links to information regarding Planning Analytics and Hierarchies

November 11, 2019

As business analytics coaches it is our passion to talk about latest innovations in the financial planning and analysis space. Today we're going to talk about Planning Analytics and the ability to create complete models using Planning Analytics Workspace.

With Planning Analytics version 2.0, IBM has added capabilities to create fully functional models that were only possible by using TM1 architect in the past. This approach worked fine for seasoned developers and modelers that knew their way around TM1, (legacy name for Planning Analytics) and its components. New modelers and developers had to familiarize themselves with all the components of TM1 which created a bit of a learning curve.

IBM's go forward strategy is to take Planning Analytics to a place where all model development and reporting can be done in one platform.

Key Features

Let's talk about some of the key features that should encourage seasoned developers to consider modeling and developing in Planning Analytics Workspace.

1. Processes - At the top of this list is writing complex Turbo Integrator processes. In Planning Analytics Workspace, you can use your TI processes from your existing code library and with little modifications you can use them in Workspace. Once you have converted these TI processes you can continue to use them in Workspace for future projects. The intellisense and code completion capabilities within Planning Analytics Workspace is very robust and helps you write extensive lines of code in a short period of time. The 4 advanced sections of a TI process are available on one page and it is extremely easy to follow the flow of the Prolog, Metadata, Data and Epilog when it is on a single page.
2. Rules - Rule writing is equally easy and has the same intellisense and code completion capabilities as Turbo Integrator processes. You can drag and drop items from the object tree into the rule editor. All the find and replace features are still there along with a multitude of other short cuts that make rule writing a breeze.
3. Control Objects - All control objects like Element attribute cubes, Element security cubes and Picklist cubes are available to modelers just like they were in architect. They all appear under the object tree in Workspace and can be used normally for data manipulation and or reference.  
4. Administration - In our opinion this is the best feature and capability in Planning Analytics Workspace. Modelers can create specific books just for administration. Each tab can represent an aspect of administering either a dimension update, data loads or element attribute updates. When organized in specific tabs along with documentation, it becomes very easy to manage a model. We have embedded specific videos regarding certain aspects of administration within the book to help a novice or new modeler manage the model.
5. Documentation – We are finding ourselves doing less documentation in external places and we all know how much we love to do documentation. Workspace is the perfect place to document procedures, definitions and descriptions regarding the planning process right where it is relevant. As an example, a workforce planning book can have steps listed in relevant tabs as to how to add an employee, or transfer or select benefits for a new employee. These instructions are available to the planner right where they are needed and makes for a quicker, productive user experience.
6. Security - Just like administering a model you can have a specific book just for security, and each tab can represent a different area of security. You can have a specific tab for users, groups, cubes and dimensions. And this way a security administrator can go from tab to tab checking security for a certain user or group.
7. Server Administration - IBM recently added administrative capabilities for planning analytics servers. You can manage the actual servers, control threads and conduct most of the activities that were previously done through the operations console. In in every new version IBM is adding more capabilities to manage the servers and have a central place for all administrative activities.
8. Workflow - IBM shared their plans to bring workflow capabilities into planning analytics at the 2019 Data and AI form in Miami FL. This new workflow capability is going to be centered around specific steps of the planning process like Workforce planning, capital planning, etc. It is going to be highly customizable and, in our opinion, will fill the gaps that exist currently in Planning Analytics today.

In Conclusion we think IBM’s emphasis on using Workspace for model development is the right step in reducing the number of tools needed to create a fully functional, robust planning model. This is going to be enormously helpful for our sales team when compared to competing industry products.