ROI Modeling
December 10, 2007 10:43 PM | Written by Yan Shikhvarger
Web analytics have really been in the news lately, even with The Economist covering which metrics are more valuable for advertisers (unique visitors, time spent, or page views). It's nice to see such recognition of the various issues in the space and help to educate clients.
An interesting issue that we commonly face is figuring out how many people we must reach to make a campaign successful. Especially if there is a specific goal to reach such as distribute 30,000 eCoupons or get 10,000 signups...
So the question is: how many must we reach? What should be done is to work backwards from that number and consider possible drop off points. The two drop-off points are at the registration/landing page and the click through from whatever media/viral/etc awareness-generating vehicles are out there.
There are some common factors that can be estimated such as click through and conversion rate and it is even possible to do a conservative and aggressive scenarios. If there is a dollar value of an eventual conversion, then it's even possible get a hard ROI.
The real value of this calculation is estimating how many people a campaign should try to reach in order to set proper expectations and decide if the budget allocated will be enough to reach these numbers. That in turn will even influence the chosen tactics.
See the live spreadsheet here
See the screenshot
Digg It | Reddit | De.lic.ious | Facebook
« Prev
Main
Next »