4-Cs Series: Complexity and Campaign Marketing (it’s harder than a Rubik’s Cube)

Veteran marketing managers can tell war stories of battles fought to secure marketing budgets – the pitches and cajoling to focus C-suite attention on the strategic and the tactical importance of effective marketing campaigns.  Getting something close to the budget you want may be just cause for heaving a big sigh of relief, but these days few marketing managers will be found clinking glasses of Veuve Clicquot in celebration.  Once the budget is in hand the real work begins.  The economic downturn has put constraints on the total number of dollars you have to spread among competing projects, but it has done nothing to constrain the nearly limitless ways those dollars can be allocated.   “Do more with less” is the mantra of the day.  To make those scarcer dollars go further means relying on more than traditional finger-in-the-wind gut instincts to tell you what campaigns will work and what campaigns won’t work.  Campaign marketing – the art of pulling together targeted messages for specific geographic markets, consumer segments and product types – is in need of a healthy dose of scientific rigor.

A mere three dimensions of complexity

Remember the Rubik’s Cube?  That delightfully maddening cultural relic of the 1980s was challenging because you had to configure the right sequence of moves in three dimensions.  One small misstep – one rotation along the wrong axis – and the whole strategy would fall apart.  Now, think of trying to solve a Rubik’s Cube-like puzzle, not in three dimensions but in at least five!  Our visual cortex regions boggle trying to imagine what this hypercube would even look like.  Yet that is the gauntlet thrown down to campaign marketing managers: configure the (1) right customer with the (2) right product and the (3) right promotional offer using the (4) message via the (5) right channel.  A typical challenge of this nature presented itself to one of our clients recently: configure eight potential messages to 50 customer segments in 70 regional markets concerning 50 product categories and four distribution channels:

8 x 50 x 70 x 50 x 4 = 5.6 million unique campaigns   for budget consideration!

That is obviously a larger number of alternative spending choices than the unaided human brain can reasonably analyze.  But the complexity doesn’t end there.  With the old Rubik’s Cube there was only one objective: get each face of the cube to be one single color.  Not so with the Marketing Hypercube (pictured in the diagram).  There are in fact multiple potential target outcomes of any given campaign.  Is the objective to build initial awareness of the company or the product?  Or is it to instill preference among an audience already familiar with the product?  Or, alternatively, is it to maximize actual purchases through targeted prices, promotional incentives, penetration opportunities, and/or purchase timing strategies?  In effect, the targeted sales & marketing outcomes themselves represent yet another dimension of complexity.

Multi-dimensional campaign marketing challenges

So how do we solve a problem of this magnitude of complexity?

Perhaps it is somewhat counterintuitive, given that we have called for a strong dose of scientific rigor, but the first order of business is to put aside the mathematics, take a step back and employ some good old-fashioned human judgment (don’t worry, we’ll shortly come back to the mathematics when we start to build predictive models around customers, messages and objectives).  Let’s start by remembering what we are trying to accomplish: to configure a campaign that will most effectively resonate with the target customer segments and accomplish our specified performance objectives.   We want to be able to predict the effect of the campaign before it is even launched.  This requires making some basic assumptions – but before your analysts integrate these assumptions into predictive models they need to obtain bottom-up business insights. These insights come from experience gained by your sales associates through interaction with their customers.  For example, they can be gleaned from short 30-45 second surveys and similar diagnostic tools built around particular initiatives (e.g. price, penetration, wallet share, loyalty, and general awareness-familiarity-preference survey templates).

The next challenge is to align these insights with the right segments.  Don’t think of this as a “once-and-done” event.  You have hundreds of thousands of customers and there are near-limitless ways to segment them.  The segments around which you build your first campaign iteration may not be the segments you employ in the end – or perhaps you will learn that those segments require different campaign strategies.  This is an iterative process – sampling, inputting new insights into existing predictive models, aligning campaigns to segments, resampling, revising segment strategies, updating model assumptions and constraints, and repeating.

It may sound tedious.  But over time this iterative process will help you greatly improve the accuracy of your predictive campaign models.  You will be in the position to pinpoint the effects that a specific campaign had on improving the value of certain customers’ transaction baskets through penetration initiatives, for example, or to measure the contribution of a customer loyalty campaign to actual revenue saved through decreased contract defections.  Those 5.6 million alternative budget allocations will start to look less daunting, and you will have a higher degree of confidence in making spending decisions closely aligned at a very granular level with your demand environment (for example, our client was able to more than triple its customer conversion rate through applying science to its campaign marketing process).  In short, you will be able to do more with less – even if you are one of the many people who never did get the hang of the Rubik’s Cube!

Subscribe   |  

Density, Sparsity and the 4-Cs

Solving the micromarketing challenges of the Information Age

We live in the Age of Information, so we are told.  Never before has so much raw data existed bearing testament to every  pulsebeat of human commerce, every touchpoint between a customer and a good or service.  The problem for decision-makers, according to the conventional wisdom, is Information Overload – volumes more data to analyze than the human brain can easily digest.  But it is not that simple – there are deeper challenges below the surface.

Information is not always where you need it

While the conventional wisdom is right in the aggregate, the lush and dense information rainforest starts to turn remarkably arid and sparse as you drill down into the nuanced segments of your demand environment.  At the micromarket level, infrequent transactional activity in the long tail of customers and SKUs yields little insight to inform decision making.  Managers thus face challenges that go well beyond the simplistic construct of TMI (too much information).  They need tools for managing the real information problems in their micromarkets.  These tools need to address head-on the challenges posed by what we call the 4-Cs:

• Complexity: With near-limitless combinations – of customers, products, locations, messages and channels – managers need the ability to first aggregate and then disentangle how variables related to price, assortment, advertising & promotions, and sales mechanisms affect customer demand and thus impact firmwide performance metrics like market share or profit margin.  Not knowing what impacts sales or profits raises the risk of suboptimal performance. Advanced scientific methods can help fill in the gaps where data sparsity exists and extend the vision of key decision-makers far into the details of what moves their markets.

• Coordination: Marketing involves a series of decisions, all of which have an impact on each other – yet each decision often gets made in an organizational silo isolated from other decisions.  This can produce persistently suboptimal outcomes unless managers can overcome the limitations of organizational silos.  Holistic optimization tools that provide visibility across silos and facilitate “what-if” experimentation can help achieve a clear, coordinated understanding of each single decision in a more integrated context.

• Connection: Managers need to connect to what the market is telling them in real time.  Historical transaction data can only help so much in an environment of constant flux: customer tastes change and competitive threats emerge in a Petri dish of constantly evolving activity.  It is not enough for decision-makers to learn from their quantitative systems: the systems have to learn from them as well.  This is what it means to be market aware: intimately connecting human experience and judgment with machine-based algorithms for optimal decision guidance.

• Customization: Insightful managers know that there is no such thing as an “average” customer.  Marketing and sales messages that play to a perceived average will wind up being average themselves – in other words falling short of truly connecting in the best way with target customers.  The fact is that every enterprise’s customer base is unique – defined by a distinctive combination of tastes, wants, needs and propensity to spend.  This is true even if the product line is what most observers would view as commodities.  Customizing a value proposition down to the most granular level possible can unlock the power of micromarket monopoly and defend against the margin-eroding practices of cutthroat price competition.

If the Information Age were really all about combing through volumes of aggregate data to develop key marketing decisions for your average customers then the 4-C framework would not matter so much – you could price, advertise and sell based on their perceived wants, needs and spending propensities.  But that average customer doesn’t exist.  The more precisely you can gain the necessary insights into micromarket uniqueness, the more you can calibrate marketing and sales decisions to optimal advantage.

So, if you are a marketing manager in a highly competitive industry  like foodservices, consumer packaged goods or retail, what should you be looking for in business intelligence & analytical solutions to take on the 4-C challenges?  Ask yourself three questions:

1. Can the solution really cope with the complexity of my demand environment in a way that is commercially viable, i.e. that keeps up with the fast pace of my daily decision-making?
2. Is the solution seamlessly compatible with my company’s existing technology platform including existing ERP and other critically important business intelligence?
3. Can my sales reps continue to do their jobs effectively and impart their experience and judgment without compromising the integrity of the system’s recommendations?

We’re going to come back and explore each of these questions in subsequent postings.

Subscribe   |  

Brand Loyalty: The Uphill (but Winnable) Battle for Heartshare

The other day I conducted a little thought exercise, and it brought me back to a question that often comes up in my line of work: the fleetingness of brand loyalty in the age of marketing message saturation and the daunting challenge for brand managers and other decision-makers whose livelihoods depend on the existence of such loyalty among their customers.  Happily for those who walk the brand beat, there is a ray of hope in this otherwise cautionary tale.

Olay, Nivea, Neutrogena and L’Oreal are all established beauty products brands with a broad array of medium-priced product lines and multiple product offerings in each.  More to the point, for purposes of this thought exercise of mine, is that each of them offers a range of good quality facial cleansers, a product I buy on average about once every two months.  The exercise was to determine what, if any, brand loyalty existed in my facial cleanser purchases over the last 2 years.  The answer appeared to be: none.  Nada.  At some point over those past 24 months and (give or take) 12 purchases, my domestic shelf space has been occupied by at least one representative facial cleanser SKU from each of those brands.  I wondered why this was the case.  And then I remembered that it was not always thus.  Long ago (more years than I care to disclose) there was a rather splendid product by Neutrogena called the Facial Cleansing Bar. Read the rest of this entry »

Subscribe   |  

Models Didn’t Bring Down Wall Street; People Brought Down Wall Street

“Burn the mathematics” wrote economist Alfred Marshall in a letter to a friend, musing about the proper role of mathematics and scientific inquiry in the field of economics.  That 19th century cogitation would seem to be a prêt-a-porter soundbite for these latter days of the 21st century’s first decade – a time in which the mathematical infrastructure that underpins longstanding economic and financial theories stands accused of all manner of malfeasance, particularly given its presumed role in the decade’s signature economic event – the financial market meltdown of 2008.  The logic behind the accusation goes roughly thus: More complex (but not necessarily more “accurate”) models allow for more complex instruments to be created. Increased complexity means it takes more time to process and then fully comprehend what the numbers may be telling you. At the same time, though, technology allows buy and sell orders to be executed almost instantaneously through electronic trading systems. Time is of the essence, and ponderously complex computations simply won’t do.  A seemingly elegant (and fast, and commercially viable) shortcut is discovered and becomes the currency of the day. The models’ outputs come to be trusted blindly simply because there is no time to question them (and too much money to be made by using them). The impenetrable Greek letters obfuscate the sensitivity of the models to changes in important assumptions – which is fine for a few years because those assumptions (e.g. rising housing prices) don’t change – but then all of a sudden they do. The models start losing more money than they make. Then the chasm widens further as the high levels of leverage in the system make themselves felt. The losses accelerate dramatically, wiping out years of profits in just a few months. Burn the mathematics, indeed.

But let’s take a different look at this apparent tight coupling of mathematics and dire outcomes. Our recent correspondence with an author who has been widely published on the subject of Wall Street’s use of mathematical models recently offered to us an interesting opinion. His point was that the problem with the models was not so much their complexity, but rather that they were models in the first place. His argument was that you can’t ever perfectly hedge model risk.  Now, I agree with that observation: a model by definition selects some aspects of reality to represent and omits others, and the choice of what to include and what to omit is subject to human error, therefore fallible and not perfectly hedgable.  But I take issue with the idea that the fault lies in the existence of the models themselves.  Models can be misused – I think that much is clear. But the notion that models are all doomed to failure obscures a deeper truth about the goals of predictive modeling; namely that you can seek either to reduce the world or truly explain it. By trying to elegantly reduce the world to as few predictor variables as possible, you are more likely to be sowing the seeds of future failure, because complexity and actual drivers of outcomes are taken out of the equations to make them more solvable (or perhaps sellable, as in the case of the Gaussian copula function that was behind Wall Street’s demise, as we discussed in a previous posting “You Can’t Punt Away the Dimensionality Curse”). Predictive modelers don’t have to go down that road, however: they can also set out with the goal not of reducing an entire system to a single neat, tractable equation, but to quantify and explain all of the relationships that dictate outcomes to the absolute fullest extent possible. Tractability and computability are things to address later in the process, through technological means, but they should not dictate the fundamental mathematical approach at the outset. Read the rest of this entry »

Subscribe   |  

You Can’t Punt Away the Dimensionality Curse

A single mathematical formula brought ruin to the global financial markets.  What happened was not a failure of quantitative methods per se but rather a lesson in the perils of ignoring real-world complexities in favor of deceptively elegant shortcuts.

The fault, dear investor, lies not in the head of AIG’s Financial Products Group or members of the Bear Stearns Investment Committee or any other anthropomorphic entity: rather it was a single mathematical formula that apparently felled the pillars of global finance.  That’s the gist of a recent article in the 3.17 edition of Wired magazine entitled “Recipe for Disaster: The Formula that Killed Wall Street” by Felix Salmon.  The formula, known as a Gaussian copula function (when is the last time that term became a fixture of the public discourse?), purported to solve the mother of all securitization problems: establishing default correlation factors between the many constituents of the pools of mortgages and other credit obligations whose cash flows served as the underpinning for the complex derivative securities known as collateralized debt obligations (CDOs).  Awareness of the potential in this arcane formula helped power the CDO market to some $4.7 trillion in volume over the course of the housing bubble years of this decade.  As the Wired article explains, the explosive commercial viability of this formula can be explained by its use of a simple sleight of hand.  Rather than modeling out the default correlation implications of pools of thousands upon thousands of individual mortgage obligations – an extremely complex undertaking requiring powerful algorithms and massively robust computational processing technology – the CDO market’s Wall Street practitioners used a shortcut that appeared elegant but proved deadly: using the market price of credit default swaps (CDSs) as a proxy for the actual historical data.

What happened in essence was that the CDO market ran up against one of the most challenging of quantitative modeling problems: the dimensionality curse.  This refers to what happens in complex environments where numerous variables interact with each other and all of the resulting combinatorial possibilities influence the economic value.  The addition of an incremental variable to the pool exerts an exponential effect on the number of possible outcomes.  Think of a simple case: if you have a pool of two variables then the number of potential outcomes is four: add a third dimension (variable) to the mix and the potential outcomes expand to nine, and so on.  In an environment like pools of thousands of mortgage obligations or credit card receivables influenced by a bevy of macro- and micro-economic, behavioral, seasonal and other random factors there are literally billions of combinatorial outcomes that could affect the incidence, magnitude  and frequency of default events and hence the price of the CDOs whose economic value derives from those pools.   Getting to the right answers – and doing so with enough speed to satisfy the blistering pace of 24-7 investment markets every day – is a daunting challenge to say the least.  So when Daniel X. Li, a quantitative analyst at JPMorgan Chase, posited the use of CDS prices as a proxy for historical data in a 2000 paper published in the Journal of Fixed Income Securities, the CDO market rejoiced and basically punted away the dimensionality curse by using this shortcut.  The reasoning and the assumptions employed proved to be flawed and the disastrous results are entirely visible to the naked eye in all their graphic detail.

In quantitative methods as in life there are no free lunches.  You can’t simply punt away the dimensionality curse – you have to embrace it and try to achieve mastery over it using all the knowledge and technology tools at your disposal.  At Sentrana we deal with dimensionality curse problems every day – the demand markets for the products and services our clients sell are highly complex environments: tens of thousands of products for thousands of customers in hundreds of locations reachable by any number of marketing vehicles and sales channels.  Modeling these environments is not for the faint-hearted: but the problems are not impossible.  The computational technology does exist, as does the modeling science.  The critical ingredient is the will and determination of those who practice quantitative methods in business to forego the easy outs and stay focused on solving the real problems, however daunting.

Perhaps the field of quantitative methods needs a variation of the medical profession’s Hippocratic Oath: First of all, do no harm.  Clearly the Wall Street experiment egregiously failed that standard.  Let’s hope that the next time some arcane mathematical formula figures into the cultural Zeitgeist it will be for better, not for worse.

Subscribe   |