peshkova - Fotolia


When is your data 'good enough' for predictive analytics projects?

Companies often wait to initiate a predictive analytics project because they want data perfection first. Why is that misguided?

In today's age of the customer, companies are striving for richer customer insights, and the value of data has never been stronger. But in some cases, they shy away from predictive analytics projects because they are Waiting for Godot: that is, waiting for data perfection before they begin.

In Bluewolf's latest "The State of Salesforce" report, 81% of Salesforce customers cited increasing the use of predictive analytics as the most important initiative for their sales strategies. But many business leaders approach the use of predictive analytics warily, because they misguidedly believe that pristine data is a requisite for successful use of these tools. In truth, no business has perfect data.

Rather than hinder you from implementing predictive tools, the existence of imperfect data gives an opportunity to identify points of weakness in your data sets and mature your strategy. The dramatic benefits of predictive analytics can be seen at every level of their use -- the key is to start work on a predictive analytics project.

Use it or lose it

Predictive has a wide definition with many use cases. At a basic level, predictive analytics can be used to identify a key customer segment in your database. Using predictive analytics tools doesn't have to be an astronomical project -- you can find insight even in the poor data. If you don't start using your data, you won't have the opportunity to further iterate it and make it more useful to you.

For example, if you begin sorting service cases by product to identify trends in customer dissatisfaction and find that the product field is blank in many of these records, you've uncovered a core problem with the data: incompleteness. These kinds of insights are the building blocks of a more robust and insightful data-driven business model.

Predictive data offers an edge

From basic to advanced uses, predictive analytics projects enable business decisions to be made proactively rather than reactively. Examples of business insights include identifying customers who are most likely to attrite, discover opportunities likely to have the most immediate impact, or predict which accounts carry the highest lifetime value. Data's power comes from its versatility. Almost any business decision can be made smarter with data as its foundation.

We recently worked with a warehouse company and a media corporation; despite their being different businesses, both optimized their product pricing with Salesforce's Wave Analytics to gain insight into their most valuable assets. By moving this process out of Excel and uniting their data sources, both companies drastically improved their revenue by predicting value more efficiently.

Empower all parts of business with data

Data can add intelligence to every part of your business. With Salesforce Wave, you can build dashboards and set up alerts to deliver these intelligent insights to each team in your business. When customers likely to attrite are identified, your marketing team will be notified and enabled to tailor campaigns to those customers to improve those relationships.

An automatic service message can be configured to be sent to a database segment, or certain contacts can be sent to a queue to be called by your service team to make sure those customers have a positive customer experience. Sales teams can be aided with predictions of the best products to sell or contracts to target for renewal. Big business goals for price optimization, retention, white space analysis and targeted marketing all begin with data.

About the author
John Hope is the director of analytics for Bluewolf, an IBM company. He enables customers to make educated data-driven decisions to become more effective and relevant in their business. Prior to Bluewolf, Hope helped pilot a Salesforce Wave project for GE Capital and implemented and managed NBC Universal's CRM and Analytics teams.

Next Steps

Do you need data scientists for a predictive analytics project?

Predictive analytics projects may need a reality check

Using data from predictive analytics software

Dig Deeper on Customer analytics