Two common metrics used in almost every sophisticated analytics effort are Customer Lifetime Value (CLV) and Propensity to Buy. CLV is often seen as the golden child of marketing analytics, used by many businesses and tracked meticulously. Propensity to Buy is its lesser-known cousin. Here, we make an argument for why Propensity to Buy might be better, and when to use each.

To begin, let’s start with definitions:

The last few weeks have seen a number of interesting blog posts and news articles have lamented the term “Big Data”. VCs, bloggers, and I’m sure many others are getting worried the term is being oversold, and certainly is over-used. Questions abound: are benefits as good as they seem, and is there really a treasure trove of value within Big Data?

Arguments abound for why the term is becoming lacklustre in 2013. And while we at Canopy Labs don’t necessarily see ourselves as as Big Data startup (we prefer to see ourselves as making analytics more user friendly), we do feel the term warrants a defense. We see three, in particular.

It comes as no surprise to many that enterprise IT projects typically run over budget or are never even finished. One of the most challenging IT tasks is around customer data and analytics. Centralizing customer data is difficult, as it requires a tacit understanding of each business unit’s data structures and IT architecture. Furthermore, analyzing such data and dispersing analytic results across the enterprise requires a paradox of organizational excellence: a centralized analytics team serving a decentralized set of business units, with the analytics team actually understanding and prioritizing its work across those business units.

Canopy Labs is fortunate to work with numerous businesses across a wide range of revenue scales. From ten-person ecommerce startups to billion-dollar retail operations, businesses of all sizes struggle with the idea of centralizing customer data, standardizing analysis, and prioritizing actionable insights.

Analytics does not always mean predictive models. Sometimes a simple dashboard tracking your team’s progress is enough to spur excitement and encourage a team to perform. Oftentimes, charting a metric implicitly sets performance targets by encouraging people to be better than the days or weeks before.

To illustrate the power of charting a simple metric, see the graph below. This shows the growth in social media referrals to a corporate website. In this case, Day Zero represents the first day of charting. The social media team tracked the data daily, and simply plotted it with the goal of improving the number of social media referrals relative to the day before. For example, if there were 1,000 referrals yesterday, then today’s goal is at least 1,001.

Most companies assume customer value is directly measured by the amount of revenue generated by a customer. While a company’s accounting department should think in terms of revenue, profit, and loss, the intrinsic worth of a customer is multidimensional, and should be viewed as such. Depending on your business model, your financial state, and the maturity of your company, choosing the right dimension on which to measure customer value can mean the difference between success and failure.

So what are the dimensions? We define customer value in four parts: revenue, loyalty, sentiment, and engagement. Read below to understand how each dimension affects your company’s sales strategy.

Since creating Insights @ Canopy Labs, we’ve been reminded by customers and potential recruits about the old articles we posted on the original Canopy Labs blog. To facilitate easy reading, below are the data-oriented articles that received the most attention over the past half year:

• Analytics has a user experience problem Most analytics platforms are developed to solve as many data-related problems as possible, and are not catered to specific uses or solutions. This makes them difficult to use by non-statisticians, and hinders successful applications of analytic techniques. We outline three ways to overcome this challenge and make analytics more user friendly.
• “Is accuracy important?” and other questions in enterprise analytics Using analytics in the enterprise brings with it unique challenges not present in more academic settings. Specifically, analytics becomes a strategic decision — a balancing of pros and cons, costs, and risks. Much of this is less dependent on model performance than internal corporate constraints and budgets. Data scientists take note.
• Five data types often ignored by customer anlaytics teams If your company is using analytics to improve any aspects of its sales operations, success can mean the difference between using or not using a specific variable or data set within a model. We outline five types of data that are often ignored by companies running customer analytics projects, yet are likely to yield valuable insights and drive significant model performance.

Have other data-oriented questions? Let us know!

Welcome to the inaugural post of Insights @ Canopy Labs. This website is a service provided by Canopy Labs and aims to achieve two goals related to customer analytics:

1. To further thinking in customer analytics and data mining. While the field is moving quickly, there are few resources that provide practical insights and instructions on applying analytics in the business environment.
2. Target small and medium-sized enterprises (SMEs). SMEs, in particular, face major obstacles in finding proper solutions to customer analytics problems. These companies are big enough to need help, but too small to afford custom solutions — this resource aims to help them.

We will provide updates on research, technology, and new trends in customer analytics. We’ll also post information on Canopy Labs, but will keep this minimal.