A BLAST FROM THE PAST

Nielsen’s New Fused Data May Be a Better Way to Target Television

By Erwin Ephron

Before the death of Apollo, the Arbitron/Nielsen single-source project, had been properly mourned, Nielsen was out in heels romancing a new partner.  Worse yet, a first cousin. 

The Company will fuse data from the NTI TV ratings service with measures of streaming video and data on consumer purchase behavior, including packaged goods, pharmaceuticals and automotives.  This will create all-Nielsen single-source database to help advertisers make better informed TV buying decisions.  The announcement takes us back to 2002 when Fusion, that year’s next big thing, was strangled by suspicious researchers.

DATA FUSION

I don’t know how Nielsen plans to do it, but the fusion concept is simple. Database A is consumer records of product purchase. Database B is consumer records of TV viewing.  Database A is 'married' to Database B by matching respondents on key characteristics.  When this is done, the fused database behaves as if a single set of respondents had participated in both studies.  That makes the media math calculations, like ratings and Reach/frequency, familiar and very simple.

If this seems spooky, you're giving fusion too much credit. The NTI panel member's ascribed purchases will not be the same as his or her actual purchases, but in aggregate the fused database will produce the same numbers as appear in both the purchase and TV viewing files.

MORE PREDICTING VARIABLES

The usual question this raises, "how good is a fusion?" is the wrong question. We should be asking "is using fusion better than what we are currently doing?"  It probably is.  Fusion works on the same statistical assumptions used when we target Women 18-49 for shampoo commercials because, per-capita, that group uses more shampoo.  But the fusion match, ascribing shampoo purchasing to TV viewers is likely to be better than the demo match because the linkage uses more than sex and age as the predicting variables.

BACK TO THE FUTURE

Fusion has a long history in the US.  A good example is the current MARS/NTI fusion (first done by Kantar and Nielsen in 2002). Here the individual respondent records of ailment sufferers from the MARS pharmaceutical magazine readership study were joined to those of the Nielsen Peoplemeter TV panel.  The links included age, sex, a number of household variables, geography and volume of TV viewing by genre divided into Cable and non-Cable.

The resulting database reports national estimates of TV viewing among the sufferers of specific ailments as if they were collected by a single-source survey using the Nielsen Peoplemeter to measure TV.  This has substantial benefits over demo match planning.

WHY USER MATCH IS BETTER

TV planning and buying have always relied upon simple demo matching to target potential buyers. The problem with this is age/sex targets seldom define consumer markets. They simply show concentration of buyers. This results in "targeting error" of two kinds.  Many in the demo target will not be product users (false positives) and many users of a product will not be in the demo target (false negatives).

For example, the TV target for acid reflux remedies is Adults 35+ because close to 75 percent of acid reflux sufferers are over 35. But only 19 percent of adults over 35 suffer from acid reflux. That means that 81 percent of the dollars spent targeting Adults 35+ may be wasted. 

Fusing MARS with NTI allows agencies to select TV networks, genres and programs, based upon the ascribed viewing of acid reflux sufferers rather than the simple age/sex profile of that sufferer group. This in theory can produce a major improvement in TV planning and buying.

DIFFERENT BUYING DECISIONS

Moving from theory to practice, it seems to work. Since the MARS records are fused to NTI it is possible to run optimized schedules comparing the cost consequences of using fused data instead of the demo target.

For Adult Sinus Headache Sufferers (a large user group) the demo target is Adults 18-49. To achieve a 65 reach of Adults 18-49 required a 166 target point schedule.  But the fused database shows that schedule generates 190 Sinus Sufferer GRP's, many more than are needed. To reach 65% of Sinus Sufferers requires not as much of TV distributed differently and costs substantially less than the original plan.

This and many other examples suggest that user targeting through fusion could help a wide range of advertised products.  Especially those where product use correlates with TV viewing better than sex and age do.

- June 13, 2008 -

^ top of page ^      < back <