As senior vice president of worldwide field operations at SPSS, Alex Kormushoff is responsible for leading the sales, services and marketing organizations for SPSS. He joined the company in 2005 as senior vice president of global services, bringing with him more than 26 years of executive management experience in the information technology industry.

PERFORM: How are predictive analytics important to performance marketing today?

KORMUSHOFF: The global marketing landscape has changed dramatically, and forever. There’s parity now – in economics, finance, technology, accessible skills, manufacturing and engineering – on a worldwide basis. This forces organizations to compete for mind-share and wallet-share, which together point to one thing: customer relationships.

Predictive analytics can help organizations capture and evaluate data to understand customers better, predict future behavior and – most importantly – take action to drive optimal outcomes. This gives marketers in particular the ability to develop more effective programs that support a continuous dialogue with customers. This way, they can more clearly anticipate customers’ wants and needs, and be proactive in satisfying them.

In general, a predictive analytics approach to marketing puts an end to guesswork in favor of more informed and actionable decisions.

PERFORM: How are you defining predictive analytics – and how does it differ from business intelligence?

KORMUSHOFF: Predictive analytics helps connect data to effective action by drawing reliable conclusions about current conditions and future events. If we’re talking about customer data, it’s looking at the full 360-degree view of the individual: you collect information on them through a continual “dialogue” and analyze both historical and dynamic data, make predictions, take action and then complete the cycle again and again.

Although business intelligence, or BI, looks at what has been happening and what is happening right now, predictive analytics looks at what’s most likely to happen next and maps out how best to act on that prediction. Some organizations take blocks of information – say, transactional data – while we propose a more holistic and dynamic approach to driving better and more informed decisions. This means looking at all data – descriptive, behavioral, transactional and attitudinal. How do you capture the attitudes? Through a dynamic dialogue with your customers.

PERFORM: What types of organizations can benefit most from this approach?

KORMUSHOFF: There’s relevancy for all walks of life – the notion transcends industries. But there are some commonalities we see from organizations that really embrace data-driven decision-making. We’ve seen these as typically large enterprises with lots of individual customers, who want to have (or already have) a direct relationship with their customers. Although predictive analytics is probably most “needed” in fast-moving, highly competitive markets, we’ve seen it applied with remarkable results to something as conventional as the public utilities industry. In fact, we find that the public sector is embracing predictive analytics more and more – to improve security, spot possible fraud or waste, and evaluate social programs.

PERFORM: Of your customers, where has there been the most success in applying predictive analytics to improve performance marketing?

KORMUSHOFF: We at SPSS see most success when predictive analytics is applied to customer-focused processes – and when there is a commitment to the customer relationship. And by relationship, we mean just that: an enduring, two-way dialogue characterized by trust, commitment and alignment that meets the mutual expectations of the company and customer.

The ideal state is to have this “personal relationship” with a million people, on the same terms, with a similar level of intimacy. Employing technology without people doesn’t work, but employing people without technology doesn’t scale. Modern technologies enable these personal conversations with a very large number of people; they allow enterprises to engage in a real and close dialogue with thousands of their clients or prospects on a one-on-one basis. A dream for years, customer relationship management, or CRM, is now possible, and one that, obviously, fulfills the promise of performance marketing as well.

PERFORM: Are there specific steps involved in getting started?

KORMUSHOFF: Some organizations have pockets of predictive analytics, with data mining or feedback mechanisms in place – but the concept of taking this across the enterprise is relatively new. In order to inject the power of analytics to influence everyday business, you should define a plan that is iterative with quick wins that will buy you a ticket to move on to the next thing. The idea is to foster a cultural shift, or journey, that builds and evolves over time.

The key to planning is to begin with the end in mind. Try to pull away from a static model/reporting mind-set to understand how we make decisions and drive business processes. The pitfall to avoid is putting an analyst in a corner, having them build some wonderful model and then finding a use for that model. We recommend, instead, that you select an area where there’s an opportunity for success – for example, customer retention – and access and gather data around that issue to extract insight for fuller understanding. But the big thing to keep in mind even from the start is how to act – how to take the data and insight, and put it to use. Keep that in mind, right from the beginning.

Unfortunately, although there are time-tested methodologies, there is no blueprint, no road map to predictive analytics success because each organization is unique. But we have gathered the expertise and experience to effectively solve crucial business problems.

PERFORM: What kind of return on investment (ROI) can one expect from applying this approach to marketing campaigns?

KORMUSHOFF: I really love this question. We’ve had customers who’ve experienced 10-fold, 100-fold, even 1,000-fold ROI – typically in about 10 months. We’ve also had organizations that apply it to turn what was a service center into a profit center for their organization – in one case, bringing in an “extra” $200 million in the first year alone.

It’s not an exact science, and each organization differs. We did have an outside agency – Nucleus Research – interview several customers to try and arrive at a “norm,” and were pleased when they confirmed that it was one of the highest ROI scores they’ve seen from a technology company.

PERFORM: Any advice for interactive marketers out there?

KORMUSHOFF: I think the step toward performance-driven marketing is the right one, so if this publication caught your eye, you’re on the right path. The added advice I would give is to really start to apply this beyond a rules-engine approach, and consider using predictive analytics to improve the performance of your existing systems. You can assess customer lifetime value, risk or fraud in near real time and guide each interaction or online experience more appropriately. The interactive channel is a prime spot to start this, but I’d keep in mind from the start how to expand it across multiple channels, so folks have the same brand experience – and you have the same benefits – across all lines of communication.