Intelligent Targeting Offers New Genetic Algorithm Modeling to Create Better Direct Marketing and Customer Acquisition Outcomes
Marketing Darwinism at Its Best: Ingenious Next Generation Data Models Compete Against One Another to Produce a Better Dataset That Produces Increased Lift, Results and ROI.
Tampa, FL, September 08, 2008 --(PR.com)-- Thousands of companies buy mailing lists to reach a certain demographic they think will produce results. Other companies use predictive modeling techniques to hone their list of prospects that will probably produce ROI. And then there is next generation modeling that does all of the above but uses genetic algorithms to pit database against database to produce a dataset with optimum lift, results and ROI.
The latter modeling technique is now offered by Intelligent Targeting.
“Predicative modeling is an excellent way to forecast revenue from a given database. But, the traditional method of linear and logistic regression to create the final dataset has inherent limitations due to the extensive time and expense to draw a conclusion on the best subset data by an analyst,” states Gary Cottingham, CEO, Intelligent Targeting.
According to Cottingham, recent developments in computing power and technology have produced an inexpensive and fast approach to make machine-learning technologies such as genetic algorithms an economically viable approach. What used to take days to produce a viable subset now takes hours. And if you wanted to create and compare predictive outcomes against several others, the time is reduced from weeks to days.
“The new technology is almost Darwinian in nature,” states Cottingham. “The algorithms can produce one model based with a predicated outcome from one assumption and then create a different model and outcome for it to compete against. The strongest model, the one with the best outcome, moves on to compete against a third model and so on. In the end, one database subset beats all the others.”
Another benefit of using genetic algorithms is that all data points are utilized to create predictive models instead of a smaller sampling randomly chosen by a human for trial and error.
“Experienced marketers and analysts always think they know the best criteria and factors to produce the best database for the best possible outcome,” states Duane Royer, partner with Intelligent Targeting. “And some can after a lot of time, money and energy has been spent. Through today’s technology, however, the computer modeling program we deliver can quickly analyze all of the data to eliminate assumption-based methodologies and human-error.
“In the end, the analysis that used to take weeks or longer to complete now it takes a fraction of the time at a fraction of the cost.”
And the predicative outcomes are much higher. Adopting this approach, marketers can reduce their risk of getting inaccurate data and improve response rates, or lift, by 7.5% to 15%.
“Producing results and ROI is what it is all about, whether it is a bull or bear market,” states Cottingham. “While many will still take the risk of using linear and logistic regression to create their lists and predictive outcomes, we believe that the more accurate models now come from the next generation of customer modeling: genetic algorithms.
To read a white paper on genetic algorithms, please visit the “news section” of Intelligent Targeting’s Website… www.intelligenttargeting.com.
A full-service direct marketing and interactive firm, Intelligent Targeting is designed to provide sales and marketing executives with a unique quantitative approach to identifying their best customers, retaining the right ones and finding more like them who shop more frequently and spend more per capita than the average consumer. Customer acquisition. Simplified.
###
The latter modeling technique is now offered by Intelligent Targeting.
“Predicative modeling is an excellent way to forecast revenue from a given database. But, the traditional method of linear and logistic regression to create the final dataset has inherent limitations due to the extensive time and expense to draw a conclusion on the best subset data by an analyst,” states Gary Cottingham, CEO, Intelligent Targeting.
According to Cottingham, recent developments in computing power and technology have produced an inexpensive and fast approach to make machine-learning technologies such as genetic algorithms an economically viable approach. What used to take days to produce a viable subset now takes hours. And if you wanted to create and compare predictive outcomes against several others, the time is reduced from weeks to days.
“The new technology is almost Darwinian in nature,” states Cottingham. “The algorithms can produce one model based with a predicated outcome from one assumption and then create a different model and outcome for it to compete against. The strongest model, the one with the best outcome, moves on to compete against a third model and so on. In the end, one database subset beats all the others.”
Another benefit of using genetic algorithms is that all data points are utilized to create predictive models instead of a smaller sampling randomly chosen by a human for trial and error.
“Experienced marketers and analysts always think they know the best criteria and factors to produce the best database for the best possible outcome,” states Duane Royer, partner with Intelligent Targeting. “And some can after a lot of time, money and energy has been spent. Through today’s technology, however, the computer modeling program we deliver can quickly analyze all of the data to eliminate assumption-based methodologies and human-error.
“In the end, the analysis that used to take weeks or longer to complete now it takes a fraction of the time at a fraction of the cost.”
And the predicative outcomes are much higher. Adopting this approach, marketers can reduce their risk of getting inaccurate data and improve response rates, or lift, by 7.5% to 15%.
“Producing results and ROI is what it is all about, whether it is a bull or bear market,” states Cottingham. “While many will still take the risk of using linear and logistic regression to create their lists and predictive outcomes, we believe that the more accurate models now come from the next generation of customer modeling: genetic algorithms.
To read a white paper on genetic algorithms, please visit the “news section” of Intelligent Targeting’s Website… www.intelligenttargeting.com.
A full-service direct marketing and interactive firm, Intelligent Targeting is designed to provide sales and marketing executives with a unique quantitative approach to identifying their best customers, retaining the right ones and finding more like them who shop more frequently and spend more per capita than the average consumer. Customer acquisition. Simplified.
###
Contact
Intelligent Targeting
Gary Cottingham
813-868-7800
www.intelligenttargeting.com
Gary Cottingham
813-868-7800
www.intelligenttargeting.com
Contact
Categories