Analytical Engine

John Sviokla recently wrote in his Harvard Business blog about our first analytics product called DemandEstimator. He cites an example of a client situation in insurance industry where we have used the product to understand agent profitability, and shares some of his other ideas on how business executives can use DemandEstimator.

Please read his full post

Interesting things are taking place in pharmaceutical industry, which I believe has the potential to change pricing landscape of high cost drugs.

In June’07, the National Institute for Health and Clinical Excellence (NICE) of UK decided that Johnson and Johnson’s multiple myeloma drug, the one year treatment of which stands at around USD 48,000 a year, can be defrayed for if the company is willing to refund the cost in case of not showing a pre-decided level of cure.

This move towards sharing the risk of treatment outcome did not go unnoticed. Providers are all the more willing to share the risk of no-outcome with the drugs manufacturers. Biogen Idec, maker of Avonex for Multiple sclerosis drug that costs around $18,000 a year has made a dynamic pricing agreement with UK’s NICE. Under the plan, 5.000 multiple sclerosis patients are being followed for 10 years to see the effect of the drugs in slowing the progression of the disease. The prices of the drugs will be adjusted based on improvement of performance. Merck-Serono has offered to refund the primary care cost if its drug Erbitux if the patient does not respond within 6 weeks.

Although Europe seems to be leader in this development, but US companies are definitely taking note. United Healthcare have made an agreement with Genomic Health with United paying for the $3,460 genetic test to determine whether a woman with early-stage breast cancer would benefit from chemotherapy. If tests don’t have the intended impact of lowering chemotherapy United Healthcare will negotiate a lower price for the tests.

Essentially this development adds a new dimension to the pharmaceutical pricing, i.e, risk. Before pricing a drug, the point of paramount interest to both manufacturer and providers is what drugs are the ideal candidates for outcome-based pricing? The high cost drugs which are new in the market or have relatively low success rate seem to be the ideal target from the developments so far. The principal risk that companies should take into account is financial risk, which in turn will depend upon probability of no- outcome and amount of dollar exposure. Apart from regular variables such as physician’s ability, patient’s specific variables etc. there may be exogenous factors such as pollutions level for certain treatment which will affect the probability of outcome.

There are several interesting questions already. Will the companies have different refund policies for different locations? Will there be selection criteria? How will regulatory environment affect the risk-sharing?

We are at an interesting corner here. I will wait and watch how the pricing territory shapes up in the near future.

We recently interviewed Bill Dorion, who is the Director of IT at PepsiAmericas. He has been a pioneer of championing and executing key analytics initiatives for PepsiAmericas. Bill has been in the Information Technology sector for 30 years, which includes 22 years in the CPG industry. Bill, in his current role, manages IT applications development team for strategic solutions selection, development and implementation.. Reproduced below, are some of the excerpts from our conversation with Bill.

Question: Can you tell us something about the type of problems you are solving in the organization using analytics?
Bill: Given our current strategic priorities, we are using analytics for reducing out of stock at retail locations. There are three key components that we are focusing on:
• Better Inventory forecasting (having the right product at the right delivering location)
• Divisional Dispatch & Voice Pick (having the ordered product picked and delivered correctly)
• Suggested Order Solution (ordering the right product and quantity at the right time to fill the customer demand)

Question: How do you quantify the value realized by an analytics project?
Bill: Analytics has several quantifiable and measurable benefits for us, as we have realized during the learning process:
• Reduced out of stocks at retail locations
• Reduced inventory levels are at the retailer back room
• Optimized delivery schedules suitable to better meet demand
• Less time for sales reps. in the ordering process and more time focusing on selling

Question: What are some of the key challenges that you face in analytics focused projects?
Bill: PepsiAmericas did not have a lot of experience in handling analytics projects which presented us with some challenges. The key learning for us was to understand that analytics projects have a strong R&D component inbuilt into them. While our organization is very comfortable with execution projects, the test and learn cycles inherent in an R&D process took some time to get used to. Another challenge for us was selecting the assistance of right resources that bridge the gap between business, statistics and common sense, and can communicate between these groups effectively.

Question: How do you see analytics growing within your organization and within the CPG industry?
Bill: If you ask me about the present, I see analytics growing in two main areas, optimizing the ordering process and helping in pricing and trade promotion spend

Question: Since selection of the right analytics resources is a key component for the success of the project, what do you look for in an ideal analyst that you hire into your team?
Bill: From my perspective, there are three key ingredients to a key analyst hire in our team. First and foremost, the analyst should fit into the culture of the team . He/She should be good at understanding business problem and incorporating it into the statistical tools to solve the problem. Last but certainly not the least, effective communication skills are very important. I talked about it earlier when I mentioned the need to bridge the gap between business, statistics and common sense through effective communication.

Mobile phone operators are facing a rapidly saturating market, thus facing customer retention as the primary challenge.  Mobile operators, like almost all big retail operations, have their own predictive churn models. Moreover, mobile operators have the luxury of having a wider array of variables. The “Know Your Customer” questionnaire and the usage behavior of the mobile phone typically provide access to a few thousand variables.  Modelers frequently add some off-the-shelf segments (from providers such as Claritas or Acxiom) in the mix to make a better sense of and add more predictive power to the model.

But the large number of variables adds at least two complexities.  First, even after performing variable reduction, modelers usually are still left with a high number of components/ factors, a number high enough to make a concise business presentation difficult.  Second, the high number of variables also affects the actionability of model insights.  No single variable affects churn in isolation, so it is hard to determine the independent impact of individual variables in customer churn.

An alternative approach to comprehend and act on the model recommendations will be to bundle the variables in logical groups.  For example, the variables related to time can indicate when a customer is more prone to churn and can be grouped as a ‘when’ group of variables. This ‘when’ group will not necessarily tell us why a customer is more prone to attrition, and would rather tell us when she is more prone to attrition.  Similarly, the variables’ indication of issues faced by a customer (her interactions with call center might be a good source for this) such as handset problem or network issues can be grouped as ‘why’ variable.  The demographic variables that are significant in predicting churn help the modeler to profile the customers likelihood to churn and can be grouped as ‘who’ variables. An integrated view of these three groups is more palatable, and from a modeler’s perspective, it is easier to convey to the business ‘who’ are more likely to churn, ‘why’ they are more likely to churn and ‘when’ they are more likely to churn.

The above approach does not only make the models easier to comprehend, but also increases actionability by providing a structure to the model findings.

This idea of “timeshare” as a “financial product” has been floating in our den for several weeks now. There are several definitions and models for timeshare. However, for the purpose of this post, let’s use a simpler definition. A timeshare “owner” is entitled to a defined vacation (pre-defined location and timing) every year for an extended duration of time. In effect, she gets property rights over “the vacation” by making an advance payment.

Let’s consider the table below:

Here, Ms. X buying an economy vacation might have ended up buying Vacation A. Let’s say, Vacation A allows 3 days and 4 nights of Deluxe Room stay at a Country Resort in Malibu. She is allowed to use any 3 day window between 4^th and 10^th January of every year (until 2020).

In 2008, however, she can exchange her Vacation A for Vacation B, which has the same price. Going further ahead, in 2010, her economy vacation of 2006 has become super premium and she can easily avail a longer version of Vacation B or C.

However, at any point, the ownership of vacation A allows Ms. X a chance to “exchange” the right over the room with another vacation of equivalent value. The pre-condition for this exchange is the “price.”

The price assignment, here, is done by a timeshare exchange, such as RCI, Interval, Marriott Timeshare, Starwood Vacation Ownership, etc. These exchanges factor the demand and supply for a particular vacation, location, quality of the property, length of vacation, seasonal variations in demand, etc., to arrive at the price of a vacation. The interesting thing is that this “price” is not constant, and depends on the “perceived” value of this “vacation” at a certain point in time.

Here, we introduce the financial twist. If vacation was tradable, then Ms. X would be able to make a $100 profit even after moving to what she would have considered a premium vacation in 2006, in lieu of her economy vacation of 2006. And suddenly, Ms. X would look like the smartest investor in town.

In the diagram above, Y-axis denotes the price of alternative vacations at three different time point.

Trading in vacations (quite similar to commodities trading, as some would point out) can help turn a timeshare into a financial product, and hence a product for a larger mass of people. This can be facilitated through a “financial timeshare exchange,” very similar to the NYSEs of the world.

However, this alternate financial market would require:

1.Data analytics capabilities to analyze customer’s behavioral preferences with respect to market forces of supply and demand

2.An understanding of the regulatory implications of completely transferring rights over a vacation, which can be facilitated only through the timeshare exchange

3.Educating customers to think of a timeshare as not just a lifestyle product, but also a financial product

A set of whitepapers for the executive who wants to leverage the data to make smart marketing investments:

1. Identifying attractive markets and understanding your customers
2. Increasing online sales
3. Focusing on profitable distribution partners
4. Improve customer experience through call center data
5. Retaining your customers

A puzzle asked - “How many ping-pong balls could you fit into the room?”

Somewhere else, there was $26,000 prize for a similar question - “How many 76 antenna balls did we fit in a Chevy Trailblazer?”

And a little far away from all this - the Japanese were inventing square watermelons, probably thinking about a very similar puzzle. Or, maybe a little variant of these questions - After putting in as many round watermelons as possible, what percentage of the container space is wasted??
A little while back, Harish wrote about the impact of shelf-space allocation on brand performance and cannibalization. As I read about the square shaped watermelons that fit the shelf-space better in retail stores better, I almost went back to Harish’s post and thought- How will the sales of round watermelons dip because of the introduction of square watermelons? I am sure this product extension will displace some of the round ones off the shelf. Is round out of shape now? Or, is round being forced out of shape?

However, all said and done, assuming demand for watermelon to be static, a one for one replacement of round ones with square ones can help ­_corner_ more profits for sure!

Societe Generale has been in the news recently and not quite for reasons they would like to be in the news for . In one of the biggest cases of Insider Trading Fraud, an SG employee managed to cause a dent of over $7bn. And this is hardly the first ever case of such use of material non-public information. In one of our earlier posts, we talked about a project where we had built filters to tag suspect insider trades to help our client with identifying potentially suspect trades for manual evaluation. The system has been used retroactively so far, and is being considered for proactive tracking as well.

The efficacy of what we proposed was based on the simplicity of filters that we built. Using tribal knowledge for identifying suspect transactions, and augmenting it with statistical rigor and tests, we were able to generate efficiencies in the compliance/audit process. These same rules can be implemented in a real time fashion to flag suspicious activities for monitoring.

Update: Julie wrote a note to us and pointed out that SoGen case is not a case of insider trading but one of outright fraud. Thank you for catching it and pointing it out to us. Nonetheless, a set of logical and statistics based filters can be a useful part of effective control environment which may have been able to detect such fraudulent activity.

Investment Banks, with activities encompassing trading as well as deal making, are susceptible to information leakage across activities.  Traders often have access to the material non public information (MNPI) i.e., insider knowledge from the deal makers, which if used can generate significantly higher (illegitimate) profits for the trader or the employer. In recent past, a few traders from leading investment banks have been found guilty of using MNPI and punished. It is beneficial for the banks from a pecuniary as well as reputation point of view to track insider trading activities and take actions against offenders. 

However, it is difficult and resource consuming (financially and intellectually) to manually eyeball millions of trades done by employees and flag potentially fraudulent activity. What may help, though, is a set of robust logical and statistical filters that are designed to reduce the overall number of trades/traders subject to manual review, thus reducing the overall cost of regulatory compliance.

The approach described above requires a non-trivial effort up-front to acquire, aggregate, and clean the trading data. After the required cleaning of trades’ data and imputation of missing value of trades in certain time periods (assuming no trade), we build a general understanding of the traders’ behavior. We supplement this understanding with the help of logical/statistical filters which are a combination of basic measures (such as directionality of trading transactions, means and variances of trade values), trend analysis (such as correlation between trade and market movements, deviation from trader’s historical trading pattern) and statistical models (such as time-series models).

 

Fig 1: Example of correlation between market movements and trading
action

 

Fig 2: Example of analyzing directionality of trading actions

 

Some of the complexities of this approach lie in the inherent biases in any technique that the analytics as well as business teams should be aware of. Whether an OLS model will work as well as a time-series model to identify suspect transactions is a hotly debated question, for instance. The key to this problem, as in most of the advanced analytics
problems, is a strong coordination between business and analytics. Subject matter expertise around investment banking, trading, stock price movements, etc. must be built into a data-driven approach to solving the overall business problem.

Recommendation Engines are being used by many online retailers today to cross-sell and up-sell products and services to online consumers. Amazon.com, Newegg.com, Buy.com seem to use some sophisticated algorithms around the recommendation engines at the heart of their online selling portal. They are also an integral part of any choice optimization strategy.

Without even going to the technicalities of recommendation engines available, we have put together a few points which can help businesses can harness the power of recommendation engines:

• Get the Basics Right
  Understanding some salient features of your products or services can help your business classify and categorize what parameters are important to track and assign weights to and make recommendations to users on. If you are a seller of toys, the most important parameters (with varying weights) may be price, toy features (learning toys or motion-based toys etc.) and toy accessories (batteries, replacement parts etc.). If you sell high-end luxury goods, the most important parameters may not include price and may just center on niche features and custom options.

• Take Baby Steps
  Unless your business is high up on the maturity curve (a la Amazon), you need not start with algorithms that are very sophisticated. Take a baby step approach to developing algorithms. Broadly speaking, recommendation engines can be classified into 3 categories:

1. Content Based filtering engines – those that are based on the contents of an item and tries to pair user interests (if known from previous web visits) to ‘contents’ of an item (can range from books to luxury cars)
2. Collaborating filtering engines or Social filtering engines – those that try to match user profile and interests with other users with similar profile and interests. Takes into account user reviews of products and services.
3. Knowledge-based filtering engines – the most complex of the three, these try to leverage knowledge of user needs and a target product or service that may not be directly related. For example, an engine may try to recommend you a GPS unit based on your interest in books related to Roadtrips or North American geography.
   

Ample literature is available on these recommendation engines and other emerging architectures (Few links here). The choice of what recommendation engine is best for the business should be entirely based on the business goals and the products and services being offered.

• Make them Adaptive
  No matter what the nature of the business is, recommendation engines should be built around the belief that there is never a perfect algorithm for designing them. They should be constantly capturing and storing usage data and patterns and must continuously learn from user website visits, time spent on web-pages, third party websites visited and so on. Every time the same user visits your website, the recommendations that are presented will be a little more relevant than on the previous occasion. A sample starting point architecture can be found here.

• Keep it True
  Last but not the least, objectivity of recommendations is valued by the savvy consumer. An online platform for freely expressing opinions (on user experience or product performance) can feed into your recommendation engine to make it better with time. Amazon.com and Buy.com provide not only good platforms for customer feedback but also opinions on where else shoppers can get the same product they are looking at cheaper or costlier rates. The neutrality of the ratings and objectivity of the price positioning influences a customer positively and presents an image that these sellers are not trying to peddle their wares necessarily, but want you to make a good purchase decision as per your needs, preferably through their own stores but not necessarily.

Photo Credits: ieatcrayonz, scott92007, Jmb, losiek