Written by:Scott Rakestraw
3/16/2008 3:30 PM

The last few of months I have been heads down working with our development team and clients to develop UQube Geo™,  a web-based application for allowing marketers to visualize performance for geo targeted media (ValPak, Advo, Money Mailer, etc) at the market level daily.  One of our objectives is to eliminate ad-hoc analysis that takes week, empowering clients with actionable insight to impact every media buy.

Market level is the evolution of media optimization for insert and print media.  Once you are settled on your mix of marketing channels (direct response/alt media, direct mail, tv, and online), media categories (inserts, FSIs, circular, etc), and programs (Advo, Money Mailer, ValPak, SmartSource, New Movers), the next logically step is to optimize the programs at the market level. 

When looking at market level, we focused on empowering our clients to evaluate markets/geographies both by market and media.  The two perspectives are outlined below. 

Market Perspective:

The market perspective helps clients evaluate a specific market or compare markets across marketing channels and media vehicles using key metrics - circulation/impressions, cost, clicks, calls, sales, and cost per acquisition (CPA or CPS).  The market perspective helps marketers…

• Evaluate the marketing halo effect - the combination of marketing that is working or not working to drive sales
• Analysis external factors that may impact marketing performance, e.g. competitor’s promotions and marketing, weather, economic factors, etc.
• Identify over fished markets – markets where multiple media vehicles are at play
• Identify offer cannibalization – multiple media vehicles in a market with competing offers, the better offer will nullify the competing media vehicles.   For example, if you received a “Cash Incentive” in the mail and saw the product/service in a newspaper or a co-op mailing with no incentive, which one do you think would win?
• Evaluate trends at a geographically level
• Overlay economic indicators and competitive spend, to highlight external factors.

Media Perspective

The media Perspective is designed to help marketers optimize what markets to buy for a specific media and at what rate (CPM), independent of all other marketing activity.  It enables a media buyer to evaluate the media vehicles specific geographically boundaries (zones, forms, zip codes) by calculating and presenting key metrics (circulation/impressions, cost, clicks, calls, sales, response rate and cost per acquisition (CPA or CPS) at the geographically buying level.  This allows for granular levels of targeting, typically reserved for direct mail, to be applied to insert and print media.

The Challenges
When designing a scalable and automated market level analysis engine, you face several challenges. 

• Translating all the media markets into a common market definition.  This required looking at all media activity in a specific market at the geographically level like DMA, CBSA, City, client regions, etc.
• Correlating markets in a confirmation report to the level where response and sales are captured, typically zip code.
• Building data quality assurance mechanism to identify and correct data issues.  For example, if a media vendor changes a zip code or carrier route grouping, not catching it can lead to false analysis. 
• Allocating sales at a granular market level.  For example, when calculating an end date for market level analysis, you have to look at 800# assignments and program start dates not only at a program level but at a zip code level.   For example, a co-op mailer will drop throughout the month, requiring you to take into account the drop to the specific zip code.  If you list the first of the month as the start date to start allocating sales to the media, it would be incorrect because even though some markets drop at the first of the month, a large majority of the circulation drops through the entire month.  Some zips will not drop until the last the week of the month.
• When looking at newspaper, multiple newspapers cover the same area, requiring allocating sales for overlapping newspaper using allocation models.
• When analysis performance at a DMA (210), zip code (40,000), carrier route (600,000), and numerous media zones and forms levels, you want to help people get to answers fast.  No one likes having  to cycle through thousands of rows in a report or in Excel to make decisions.  This aspect of developing the product was one of the most difficult but also the most fun.  We will continue to explore creative ways to present market level analysis but I think we have good start with the use of performance bands, plot charts and visualizations.  To see the how visualized DMAs, check out the instant demo at http://www.squeezemorejuice.com, our micro website for UQube Geo.
• Convincing the media vendors it is advantageous to provide confirmations at the market level. In the end, the media vendors buy-in because they realize that market level analysis helps everyone operate in acceptable tolerances for key metrics, reducing the risk the media would be dropped from the media mix. Also, it provides an innovate means of competing with direct mail and online media. 
• Getting confirmation data from media vendors in a format that enables market level analysis but does not disturb the existing insertion order and confirmation process. Most vendors have their own systems and market definitions. Unfortunately, there is currently no standard for market level insertion and confirmation data that is shared across vendors, yet!

The Results
Using these two perspectives, we have seen some interesting results.   Below are some real life examples.

• A program was about to be dropped because it was not performing with the tolerance range but did represent a significant number of sales.  By eliminating the bottom markets, the media’s performance greatly improved and was not cut. 
• Looking at external factors such as home sales, a client was able to see that markets that performed well, regardless of media, were markets with good new home sales.  This insight provided a new variable to consider when trying to drive sales at the lowest cost. 
• A client redirected circulation from a poor performing market to the zip codes that performed well in the market to improve results.
• A client who utilized circular (newspapers) had specific markets and news papers that were outside the cost per sales tolerance level.   Instead of cutting the poor performing markets, they renegotiate a better CPM (cost per thousand) to bring the markets into an acceptable cost per sale tolerance.
• In a specific media, a client increases sales by 20% without increasing spend just by redirecting spend from poor to good markets.

This is just the start for geographically analysis for insert and print media.  It is cutting edge for some media vendors but everyone realizes that targeting at the geographical level is one way to being more competitive with direct mail and online media.  It also is a win-win for media vendors and buyers.  Instead of a media being dropped out of the mix because it does not work on a national scale, it can be optimized and priced at the market level so everyone wins.  

Copyright ©2008 Scott Rakestraw

TrackbackPrint

Location: BlogsScott RakestrawGeo Analsysis

Your name:
Title:
Comment:
Security Code Enter the code shown above in the box below
Add Comment   Cancel