What I Learned This Week – Nov 19th, 2017

Article 1: You May Not Need Big Data – Harvard Business Review

The article titled “You May Not Need Big Data After All” originally published in December 2013 makes a case for evidence-based decision making in organizations. The authors argue that a big data projects in and by itself won’t magically transform an organization. Companies first need to learn to use data and analysis to support its every day decisions. The authors have used the example of 7-Eleven shops to illustrate this point:

  • Toshifumi Suzuki – the first CEO – decided early on that rapid inventory turnover was the key to profitability. He placed the responsibility for ordering in the hands of mostly part-time salesclerks.
  • To support his salesclerks, he sent each store daily sales report and supplemental information such as weather forecasts.
  • The reports detailed the following:
    • What had sold the previous day?
    • What had sold the previous year on the same date?
    • What had sold the last day the weather was similar?
    • What was selling in other stores?
  • To ensure data, decision making and operations were in sync, Suzuki arranged for deliveries three times a day. Multiple deliveries also ensure that the food sold in stores were fresh.
  • He also connected the clerks with suppliers to encourage the development of items that would suit local tastes.
  • The data presented to the clerks is not big data, but little data.

There are four best practices for evidence-based decision making:

  • Agree on a Single Source of Truth
    • The entire organization relies on the same data.
  • Use Scorecards to inform people of their and their team’s performance on a daily basis
  • Explicitly manage your business rules and allow software to implement those rules
    • Use a rules engine to separate the rules from the enterprise software. Managing and changing rules may be easier with a rules engine. IT expertise may not be needed to change the rules in a rules engine.
  • Use coaching to improve performance

Article 2: Advertising Analytics 2.0 – Harvard Business Review, March 2013

How can companies attribute sales to the right marketing channel? What was the mix of marketing channels that led to a sales?

These are the type of questions that marketers are trying to answer with their advertising data. It is a challenging problem but formulating an answer may get a company closer to a better ROI from their marketing budget.

The author of this article suggest that each marketing executive need to ask these two questions:

  • How did the combination of ad exposure interact to influence a consumer?
  • Is the company investing the right amounts at the right point in the consumer decision journey to spark an action? 

Marketing professionals have access to more data than ever, but all that information is useless without proper filtering and processing. Companies can no longer just look backwards a few times a year to correlate sales with a few dozen variables.  According to the author, companies need to analyze hundreds of variables in real-time and create an “ultra-high definition” picture of their marketing performance, run scenarios, and change ad strategies on the fly. The author claims that with these data-driven insights, companies can often maintain their existing budgets yet achieve improvements of 10% to 30% in marketing performance.

The author describes an advertising analytics framework composed of three main activities:

  • Attribution: Gathering insights on how your advertising activities interact to drive purchases.
    • An attribution model can be created based on data about the following:
      • Market Conditions
      • Competitive Activities
      • Marketing Actions
      • Consumer Response
      • Business Outcomes
    • I had created a multi-channel attribution model demo using a graph database. You can read about it here:

Exhibit: Advertising Analytics Framework – Optimizing Advertising

Screen Shot 2017-11-19 at 11.25.23 AM.png

(Source: Harvard Business Review)

  • Optimization: Use predictive analytics tools to run scenarios for business planning. One result of this process is that you have arrived at an elasticity for each of your business drivers.
    • If your TV advertising elasticity in relation to sales is 0.03. Doubling your TV advertising budget will yield a 3% lift in sales.
  • Allocation: Given the progress in technology, marketers can readily adjust or allocate advertising in different markets on a monthly, weekly, or daily basis.

 

 

 

 

 

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s