Distribution-Based Analytics

What is distribution-based analytics?

Distribution-based analytics is a process whereby sales results are adjusted according to the distribution levels of the particular product groups being analyzed.

Why is this important?

Academic studies1, as well as 18 years of empirical work by TABS Analytics, have shown distribution to be the most important variable in the marketing mix.  Analytics conducted using this methodology offer more explanation of current market dynamics and are more predictive of future sales trends.Consumer Package Goods, CPG, Category Management, Trade Promotion, Trade Marketing, Retail, Sales, Marketing, Analytics, nielsen, symphany-iri, data, point of sale, TABS, PromoMaster, QuickTABS, Syndicated Retail Data,

Can you cite some examples of distribution-based analytics?

There are two primary areas where this method is used:

The first is sales productivity.  Let’s say that the #1 brand in the category has a 20% dollar share.  This information is somewhat helpful, but incomplete.  Compare that with being told the 20% share brand has 30% of the SKU’s on the shelf, you have a different perception of the true strength of that brand.  It is over-spaced relative to the sales that are being delivered, and retailers will likely reduce the assortment.  Conversely, if a 20% share brand has 10% of the SKU’s on the shelf, you can conclude that the brand is very productive, and it will have a relatively easy time adding incremental items at retail to further grow sales.  Sales Productivity is a more accurate assessment of consumer preference for a brand or segment.

The second area is called unit velocity change (sometimes referred to as organic growth).  It compares the units per SKU (i.e. velocity) of a brand or segment to similar yearly time periods.  Using the same 20% share brand (Brand A) to illustrate, let’s assume this brand has shown 25% growth versus last year.  Great news, right? However, the analytics also show that Brand A grew their SKU count by 50% from last year, while generating only 25% gain in unit sales.  This means that the average velocity per item (i.e. units per SKU) actually decreased by 17% [(1.25/1.50) – 1], which means it is now less productive than it was last year.  This means Brand A’s underlying trend is actually negative rather than strongly positive.  Another scenario may be Brand A generates 25% growth with only an 18% gain in SKU count.  This is very good news: velocity grew by 6% despite an 18% gain in SKU’s.  Velocity was not diluted at all.  The 6% growth is highly predictive of future trends over the next 12-24 months for Brand A.

What are some of the advantages of distribution-based analytics?

1.  More appropriate description of consumer preference.

2.  More predictive of future trends than just dollar sales trends.

3.  Powerful tool to evaluate distribution level gaps at any geographic level.

4.  Facilitates extremely fast insights into overall sales trends.  First level diagnostic on “why.”

Distribution-based Analytics

Ataman, Berk, Carl Mela and Harald van Heerde (2008), “Building Brands,” Marketing Science, Vol. 27, no.6.

 

Request Demo