Analytic tools are evolving. They're expanding from specific applications that measure such things as website traffic and content consumption to data aggregators for visualization and advanced analysis.
There are multiple new tools that build on this trend, including Watson Analytics, IBM’s cloud-based analytics service,
Watson Analytics is a natural language-based cognitive service that provides business professionals access to predictive and visual analytic tools. Through IBM’s relationship with Twitter, Watson Analytics also lets customers to analyze social sentiment for insights around programs, products, trends and more.
Watson Analytics has been out of beta since late 2014, with free accounts available. I took a look at the tool to evaluate its features and learn how it helps turn data into useful, relevant visualizations.
How It Works
The UI for Watson Analytics displays multiple segments for users to choose:
- Exploration, for data discovery based on the terms provided within the dataset. For example, users can ask a question “what is the relationship of data in column one to data in column two?” They will then see a visualization of the possible result.
- Prediction, for determining influences and drivers among the data.
- Assembled, creates a dashboard and visualization report
- Refine, for evaluating data
Users can upload data into a spreadsheet workbook. But data within the sheet should be correctly formatted.
Uploaded data should be in a comma-separated values (CSV) file or Excel format with tables up to 50 columns. The data should be contained in a file smaller than 10MB. Once uploaded, users can select one of the analysis segments.
Why Use It
Watson Analytics lets you examine data with less emphasis on query language protocol and more on visualization technique. A search box with sort and filtering is available to search the datasets. The search is conducted by forming parameters into a question.
Different visual outputs are available, such as a decision tree or a circle.
If another visualization is preferred, users can select one of several visualization choices using the icons on the right side of the screen.
This simplicity of data usage has been a trend in business intelligence tools. Many business professionals do have much experience with databases languages. Thus providers like IBM seek to offer query options that do not require deep programming knowledge.
Fits With Broader Goals
IBM’s release of Watson Analytics is part of a larger strategy to increase the company’s market share of business analytics application.
For example, IBM announced a healthcare analytics program based on Watson. IBM’s strategy also reflects an interest in flexible cloud computing models that permit customers to move data and applications across cloud environments — public, private or hybrid.
IBM also released a professional edition of Watson Analytics that adds several upgrades such as increased collaboration capabilities on a single dataset, expanded database interoperability and connectors to cloud storage providers.
There are options that expand feature choices, such as the ability to upload spreadsheets from two cloud storage providers, Dropbox and Box.
Watson Analytics Pro users also gain expanded access to Twitter data with access to 50,000 tweets per dataset.
The professional version is $80 per user per month, so affordability for small businesses is still possible with an upgrade.
Real Life Use
Which brings me to the point about tools such as Watson Analytics. Innovators more intent on inventing strategies than predicting from data can now look harder at how to computationally massage their data in tools such as Watson.
The competition for sales among these tools will be high.
The new players will compete with established digital tools such as Google Analytics and Adobe Analytics as well as business intelligence tools with programming capability. Overall professionals are entering a terrific time to sort their needs and choose solutions that permit predicted opportunities from their data.
Simpler Media Group, 2015