Comments on: The benefits of segmentation, part I http://blog.webgenomeproject.org/the-benefits-of-segmentation-part-i/ Thu, 01 Jul 2010 19:42:57 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Kaila Colbin http://blog.webgenomeproject.org/the-benefits-of-segmentation-part-i/comment-page-1/#comment-2028 Kaila Colbin Tue, 04 Nov 2008 01:55:57 +0000 http://blog.webgenomeproject.org/the-benefits-of-segmentation-part-i/#comment-2028 Hello there Rachel, Thanks so much for your input. I totally agree with you. Like most questions, this one gets more complex the more you try to scale it. So as a small business that sells maternity wear, I might leave flyers in obstetricians' offices, and that would be appropriate segmentation. On the other hand, a large company in a saturated market whose shareholders demand continual growth, I might seek unfair advantage by crossing the line of that sweet spot. The other difficulty lies in the phrase 'what data is available'. All data needs to be collected in some way; the most basic form of segmentation would involve using only data that you collect anyway in the course of doing business. But of course that's generally not enough to make significant predictions, so further data gets collected/mined/purchased. At what point should data be off limits? I welcome your thoughts and ideas. Hello there Rachel,

Thanks so much for your input. I totally agree with you.

Like most questions, this one gets more complex the more you try to scale it. So as a small business that sells maternity wear, I might leave flyers in obstetricians’ offices, and that would be appropriate segmentation. On the other hand, a large company in a saturated market whose shareholders demand continual growth, I might seek unfair advantage by crossing the line of that sweet spot.

The other difficulty lies in the phrase ‘what data is available’. All data needs to be collected in some way; the most basic form of segmentation would involve using only data that you collect anyway in the course of doing business. But of course that’s generally not enough to make significant predictions, so further data gets collected/mined/purchased. At what point should data be off limits?

I welcome your thoughts and ideas.

]]> By: Rachel http://blog.webgenomeproject.org/the-benefits-of-segmentation-part-i/comment-page-1/#comment-2027 Rachel Fri, 31 Oct 2008 11:11:47 +0000 http://blog.webgenomeproject.org/the-benefits-of-segmentation-part-i/#comment-2027 You're so right about the "sweet spot" - it's so easy to take masses of data and try to find a link between this current behaviour and some obscure predicted behaviour. Businesses have to be realistic about what's relevant - to them and to their customer/prospect. I guess I've always seen segmentation as a fairly simple way of using what data is available to group customers or prospects together. Once grouped, then making some "educated guesses" based on that grouping about what offers, products, services, media choices and or response mechanisms are going to be relevant to the customer and what is going to give a return to the business. But I am still astounded at the low level of even basic segmentation that some businesses are practicing, when they have appropriate data available to be much smarter and accurate. You’re so right about the “sweet spot” – it’s so easy to take masses of data and try to find a link between this current behaviour and some obscure predicted behaviour. Businesses have to be realistic about what’s relevant – to them and to their customer/prospect.

I guess I’ve always seen segmentation as a fairly simple way of using what data is available to group customers or prospects together. Once grouped, then making some “educated guesses” based on that grouping about what offers, products, services, media choices and or response mechanisms are going to be relevant to the customer and what is going to give a return to the business.

But I am still astounded at the low level of even basic segmentation that some businesses are practicing, when they have appropriate data available to be much smarter and accurate.

]]>