Comments on: Predictive analytics in marketing decisions http://jtonedm.com/2009/02/10/predictive-analytics-in-marketing-decisions/ James Taylor on Everything Decision Management Mon, 14 May 2012 18:05:17 +0000 hourly 1 http://wordpress.org/?v=3.3.2 By: James Taylor http://jtonedm.com/2009/02/10/predictive-analytics-in-marketing-decisions/comment-page-1/#comment-13144 James Taylor Wed, 11 Feb 2009 15:59:03 +0000 http://jtonedm.com/?p=1407#comment-13144 Mark I guess that using analytics to decide yes/no on a specific offer would add value but it seems a small return on the analytics investment. Unless you at least have the control to send one of several offers based on the analytics I would argue your ROI will be small JT Mark
I guess that using analytics to decide yes/no on a specific offer would add value but it seems a small return on the analytics investment. Unless you at least have the control to send one of several offers based on the analytics I would argue your ROI will be small
JT

]]> By: Mark Eastwood http://jtonedm.com/2009/02/10/predictive-analytics-in-marketing-decisions/comment-page-1/#comment-13143 Mark Eastwood Wed, 11 Feb 2009 14:45:24 +0000 http://jtonedm.com/?p=1407#comment-13143 James,    I think I disagree with part of what you said. You said "If you assume one marketing decision for everyone (sending the same newsletter, for instance), then knowing how to segment your customers won’t help."   I argue that analytics can help you know to whom to send and not send the offer. If a big credit card company has the data they can create an analytics profile for the kind of prospect that is more likely to respond to their offer. Then they filter a list of names using the score so that only the more likely people are sent the solicitation.    Doing so means they can send fewer mail pieces (lower costs) and achieve a higher response rate (if the analytics are accurate). I'd call this a kind of "marketing ROI" or perhaps decision-yield improvement. Perhaps this is what you intended with your thoughts on segmentation, but I took it to mean different offers by segment. Cheers, mee James,
   I think I disagree with part of what you said. You said “If you assume one marketing decision for everyone (sending the same newsletter, for instance), then knowing how to segment your customers won’t help.”
  I argue that analytics can help you know to whom to send and not send the offer. If a big credit card company has the data they can create an analytics profile for the kind of prospect that is more likely to respond to their offer. Then they filter a list of names using the score so that only the more likely people are sent the solicitation.
   Doing so means they can send fewer mail pieces (lower costs) and achieve a higher response rate (if the analytics are accurate). I’d call this a kind of “marketing ROI” or perhaps decision-yield improvement. Perhaps this is what you intended with your thoughts on segmentation, but I took it to mean different offers by segment.

Cheers,
mee

]]>