My First Predictive Customer Journey Analysis: 2005

In the analytics media right now, “customer journeys” are cool and topical, but predictive customer journeys are even more inspiring, because they go well beyond telling you what’s already happened, into the zone of what’s likely for each of your customers to do next, and what you can do about it to present a timely strategy or treatment. This helps you improve offer timing, relevance and profitability.

You might have read some of our case studies in this area such as Five Lessons Learned From Predictive Customer Analytics or What Does a Predictive Customer Journey Look Like. Or maybe you’ve read about how we execute these strategies in Detecting Customer Events and Business Responses to Customer Events.

What you probably don’t know is, while customer journeys just became cool in 2017, we were building predictive customer journey strategies back in 2005 with a customer loyalty decision sciences program for one of the country’s prominent payment platforms.

Here’s how the program worked. The payments platform marketers selected a nationwide merchant for a sales stimulation offer, like $25 cash back at the point of purchase on a purchase of $100 or more. The strategy was that these could only be cardholders that hadn’t purchased anything using that card at that merchant’s stores anywhere in the country in the last six months. This dramatically increased the direct attribution of these offers towards cardholder response.

In terms of targeting, it would have been easy to simply contact those customers who had purchased at the merchant’s direct competitors, or had high RFM (recency, frequency, monetary) behavior in that merchant’s broader category (e.g., retail apparel, or airline travel, or small office/home office, or retail technology). Of course, we did use that sort of information as an input in our scoring and segmentation approach.

Looking back, we did something else that turned out to be even more powerful. We had access to 13 months of individual transactions for over 60 million cardholders. We built behavioral profiles of customers that ought to have purchased with that merchant (but didn’t), based on their purchasing similarity to other customers that actually had purchased from that targeted merchant. This meant looking at purchasing patterns across virtually every merchant category, the timing of those purchases, and the sequencing of relevant purchasing events by week and by month.

Using those advanced approaches, and comparing to several comparison approaches (i.e., using only purchasing history with the merchant’s direct competitors, or using random selection with the same cash-back incentive), we found our approach usually beat every other approach by a wide margin. Performance analytics revealed that most importantly, we could get the offer timing right based on who was primed for such a purchase.

We repeated this set of market tests for over 18 months, building more than a dozen of these major campaigns, improving the approaches, and generating tens of millions of dollars in increased spend incremental to the other test cells in our campaigns. We also leveraged this approach with some other credit card issuer clients of ours, seeing whether these approaches would work as well for customer retention tests as it did for sales stimulation (it did).

This is one of the client experiences that led us to create the Corios tagline, “We tell the story numbers can’t”, because we found we were literally telling a story about consumer purchasing patterns that perhaps nobody else had discovered yet. We were lucky to have access to such a massive repository of transaction data was so difficult to put in one place, given the technology capabilities of the time.

Since then, we’ve generalized the approach to go well beyond credit cards, and applied it in all sorts of other financial services, manufacturing and retail domains, for marketing, risk and financial crimes behaviors, and focusing on telling stories about the customer journey across the omni-channel experience.

Do you want to know more about predictive customer journeys? Send us a note to to start your own journey.

Humanizing Next Best Offers to Maximize their Effectiveness as Field Sales Leads

Next best offers: As marketers we have many offers we can target and deliver, each to a large pool of customers. The question remains, what are the best offers to make, not only measured by what’s best for your company, but more importantly what’s best for your customers. To qualify as “best”, these offers should be customer-relevant at that time in your customer’s relationship, and address their unmet needs.

The most conventional (but least effective) approach to select the best offers relies on sorting by the most profitable products and executing those offers to the company’s best customers. A more effective approach is to sort and target the best customers first; this is better but far from the ideal strategy. We’ve found the ideal strategy in almost all situations is to target individual offers to individual customers through the customer’s preferred channel at the best time for that customer. Sure, this is analytically and operationally challenging, but the results we’ve seen in more than 80 cases with our clients has proven this to be the winning strategy. We’ve also found that the technical and operational challenges are really not the limiting factor. Instead, it’s the change management required inside the company to fundamentally alter the way offers and channels are prioritized, how success is measured, and to convert the organization from a product-centric mentality to a customer and contact strategy focus.

In order to overcome this cultural challenge, we’ve found it is the most fruitful to put this capability right in the hands of relationship managers in the sales field, because their top-line revenue contribution is vitally important and they are always willing to consider prudent ways to spend more time with the customers who are actually interested and in the market. Correspondingly, we converted our approach to next best offers to not focus only on the needs of marketers, but also to focus on sales management and relationship managers.

How did we do this? First, we delivered our next best offer strategies through CRM platforms like Salesforce and Dynamics, which are the most common platforms our clients use.

Second, we converted the format of each next best offer into the structure that relationship managers want to consume. Specifically we added a sales narrative to each next best offer. Sure, we still perform our selection of offers using sophisticated analytics approaches, but frankly, the details of those approaches aren’t that relevant to our relationship manager audience. Instead, what they find useful are business-friendly English descriptions about each customer’s profile and what makes them likely to prefer that offer. This aids the relationship manager by giving them a clear set of tips to start the conversation. For example, “how other customers like you use this product, and the value they get out of it” is better than having only information from the product manager’s perspective.

We’ve also found that customers who are ranked as highly preferable to want a certain offer, score high on that offer for very different reasons. So we call out in simple business English the top 3 reasons that makes that customer a high scorer. This means the offers are not just marketing-qualified, but we can also speed up how quickly they become sales-qualified. This works for consumer offers and for business offers. Check out this example.

Want to learn more about next best offers? Download this RedPaper. Or if you want to get to work on building this strategy for your firm, contact us directly at

Predictive IOT to reduce mechanical failures and warranty claims

In pursuit of 99% vehicle fleet uptime

Corios helped a vehicle manufacturing client apply predictive algorithms to connected vehicle and service center geofence data to help them pursue fleet uptime of 99%, and to reduce mechanical failures and warranty claims. This was a fascinating application of predictive analytics to the internet of things (IOT).

Our client is a major truck manufacturer who has the strategic global objective of increasing vehicle uptime from 96% to 99%. How they intend to get there is to predictively identify vehicles at risk of mechanical problems before they happen, by combining real time data from telematics alerts, service center geofencing, repair histories and travel and road conditions.


Corios brought in a top-down view of the problem; we gathered data from a variety of sources, including vehicle telematics alerts, warranty claims, vehicle repairs, vehicle registration, vehicle service histories, and equipment supplier profiles. This became a massive collection of data, and we rationalized all this information so we could see a complete 360 degree view of every vehicle and its detailed history over time. Then we applied predictive analytics to rapidly identify the leading indicators of the equipment failures which led to the increase in warranty claims.


Corios Tempo: operationalizing analytics via a model factory process

Reducing costs and increasing throughput of analytics assets

Predictive assets aren’t academic exercises

The biggest challenge that confronts predictive modelers and data scientists is that it doesn’t matter how good any model is, if it isn’t used by other people in your organization to improve the way they run your business.

A recent Harvard Business Review article noted at a recent industry conference, that out of 150 data scientists, roughly a third had developed a model in the past year, but not a single one had deployed it into production, nor monitored that model’s effect on business value. That’s just sad.

We believe there are some common reasons that predictive models struggle to get used. They’re complex, they’re arcane, and there is no common process or vocabulary in most organizations that enables people across the domains of data science, customer-facing representatives, and technology to make daily business decisions using models. That’s a problem we are committed to solving.

Corios Tempo: our process for building an analytic model factory

We’d like to help the industry define the discipline of rapid model development and deployment. Our own internal name for this discipline is Corios Tempo, which reveals that we think maintaining a stable rhythm of activity is an important design element.

In this presentation, we’ll discuss:

  • the problem we aim to solve,
  • the design principles of a robust solution,
  • the benefits of doing so,
  • the roles and responsibilities inside the analytics organization that need to be harnessed to solve this challenge,
  • the way we organize data,
  • how we build models,
  • how we validate and deploy these models,
  • how we administer the platform on which all this asset development and management activity takes place, and
  • the best practices that we have developed over my 30-year career in analytics as well as the decades of collective experience developed by the Corios team.

If you’d like to read more about how Corios Tempo solves this problem by providing enterprises with a repeatable process for building and operating an analytic model factory, download our Corios RedPaper on Operationalizing Analytics Assets. Or you can get in touch with us to find out how you can build an analytics model factory.

Analytics investments: why cool isn’t enough

It's really about value and execution

During a recent industry briefing call, the Chief Analytics Officer of a life insurance carrier asked me an interesting question: “Should I try to get funding support for our new analytics initiative on the basis that it’s cool and sexy?”

When I prompted her about what’s sexy and cool about her proposal, she replied, “Well, machine learning is sexy, because especially in our industry, it’s the new hot thing. And it’s cool because I think we can improve how we make underwriting decisions on new policies.”

Then, I followed up with an important clarifying question, ”Do you have the technology support to implement that model, if you did indeed find a better underwriting mousetrap?” She had to admit she didn’t yet know how or whether she could implement such a new model in their production systems, which is maintained by the technology organization. She also didn’t know the best way to work with the underwriters to adopt the pricing policy changes that her upgraded model would support.

Why cool isn’t enough

Here’s why I have found cool doesn’t win the day: In many financial services firms, few executives want to be a guinea pig for an investment that is novel in their industry, especially in fiscally and operationally conservative environments like the life insurance industry. Cool simply isn’t enough.

Read More