Re: Ranking customers

This WebDNA talk-list message is from

2003


It keeps the original formatting.
numero = 48273
interpreted = N
texte = You need to determine their average purchasing per day * average number of days and when that equals their average purchase cost, they are likely to buy. This only works for clients who have purchased more than once. You'd have to bug one-time clients. This is basically just a tool to gague how much a returning customer is going to purchase Today... You can send out promotions based on that dollar value, or on the likelyhood that they will purchase soon.Variables:[date] Today's Date[early_date] Earliest Purchase Date[last_date] Last Purchased Date Generated variables:[ave_purch_day] Average Purchase / Day[ave_next_purchase] Average Revenue next Purchase[early_to_today] Days from Earliest Purchase until today.[last_to_today] Days from Last Purchased Date until today[ave_purchase] Average Purchase[ave_days] Average days between purchases How to figure it out:ave_purchase=(average of purchases)ave_days=Average( Purchase 2 date - Purchase 1 date (Days between 2nd purchase and 1st, etc) Purchase 3 date - Purchase 2 date Today's Date - Last Purchase date ) early_to_today=[math]{[date]}-{[early_date]}[/math]last_to_today=[math]{[date]}-{[late_date]}[/math]ave_purch_day=[math][ave_purchase]/[ave_days][/math]ave_next_purchase=[math][ave_purch_day]*[ave_days][/math] When [ave_next_purchase] is greater than or equal to [ave_purchase] then the customer is most likely to buy something else and should be sent a promo. This can be based on how much the average purchase is. you can get them to try to buy a more expensive item, or you can offer them a deal based on how much they spend on average.I don't know if this helps at all, but it's what I came up with quickly. I have an excel spreadsheet that outlines it fairly well if you would like it.-- Matthew C. Bohne Web Developer Sandusky Register 314 W. Market St. Sandusky, OH 44870 419-625-5500 ext. 253 matthewbohne@sanduskyregister.com http://www.sanduskyregister.comOn Sunday, March 2, 2003 9:11 PM, WJ Starck wrote: >We'd like a way to rank our customers so that we can target our better >customers with select promotions. One way would be to rank solely based >on $ amount, but that probably doesn't tell the whole story. For >example, someone might have purchase $750 from us on one order, and >then never returned. > >Accordingly, I imagine it should be a weighted mixture of total $ >amount purchased (+ weight), average $ purchase amount (+ weight), >total number of purchases made (+ weight), frequency of purchases (+ >weight) and time since last purchase in days (- weight) > >Any statisticians out there care to weigh in (no pun intended)? > >Comments, ideas, suggestions appreciated... > > >-- > >Will Starck >NovaDerm Skincare Science >http://www.novaderm.com >wjs@novaderm.com > > >------------------------------------------------------------- >This message is sent to you because you are subscribed to > the mailing list . >To unsubscribe, E-mail to: >To switch to the DIGEST mode, E-mail to >Web Archive of this list is at: http://webdna.smithmicro.com/ >------------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://webdna.smithmicro.com/ Associated Messages, from the most recent to the oldest:

    
  1. Re: Ranking customers (Matthew Bohne 2003)
  2. Ranking customers (WJ Starck 2003)
You need to determine their average purchasing per day * average number of days and when that equals their average purchase cost, they are likely to buy. This only works for clients who have purchased more than once. You'd have to bug one-time clients. This is basically just a tool to gague how much a returning customer is going to purchase Today... You can send out promotions based on that dollar value, or on the likelyhood that they will purchase soon.Variables:[date] Today's Date[early_date] Earliest Purchase Date[last_date] Last Purchased Date Generated variables:[ave_purch_day] Average Purchase / Day[ave_next_purchase] Average Revenue next Purchase[early_to_today] Days from Earliest Purchase until today.[last_to_today] Days from Last Purchased Date until today[ave_purchase] Average Purchase[ave_days] Average days between purchases How to figure it out:ave_purchase=(average of purchases)ave_days=Average( Purchase 2 date - Purchase 1 date (Days between 2nd purchase and 1st, etc) Purchase 3 date - Purchase 2 date Today's Date - Last Purchase date ) early_to_today=[math]{[date]}-{[early_date]}[/math]last_to_today=[math]{[date]}-{[late_date]}[/math]ave_purch_day=[math][ave_purchase]/[ave_days][/math]ave_next_purchase=[math][ave_purch_day]*[ave_days][/math] When [ave_next_purchase] is greater than or equal to [ave_purchase] then the customer is most likely to buy something else and should be sent a promo. This can be based on how much the average purchase is. you can get them to try to buy a more expensive item, or you can offer them a deal based on how much they spend on average.I don't know if this helps at all, but it's what I came up with quickly. I have an excel spreadsheet that outlines it fairly well if you would like it.-- Matthew C. Bohne Web Developer Sandusky Register 314 W. Market St. Sandusky, OH 44870 419-625-5500 ext. 253 matthewbohne@sanduskyregister.com http://www.sanduskyregister.comOn Sunday, March 2, 2003 9:11 PM, WJ Starck wrote: >We'd like a way to rank our customers so that we can target our better >customers with select promotions. One way would be to rank solely based >on $ amount, but that probably doesn't tell the whole story. For >example, someone might have purchase $750 from us on one order, and >then never returned. > >Accordingly, I imagine it should be a weighted mixture of total $ >amount purchased (+ weight), average $ purchase amount (+ weight), >total number of purchases made (+ weight), frequency of purchases (+ >weight) and time since last purchase in days (- weight) > >Any statisticians out there care to weigh in (no pun intended)? > >Comments, ideas, suggestions appreciated... > > >-- > >Will Starck >NovaDerm Skincare Science >http://www.novaderm.com >wjs@novaderm.com > > >------------------------------------------------------------- >This message is sent to you because you are subscribed to > the mailing list . >To unsubscribe, E-mail to: >To switch to the DIGEST mode, E-mail to >Web Archive of this list is at: http://webdna.smithmicro.com/ >------------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://webdna.smithmicro.com/ Matthew Bohne

DOWNLOAD WEBDNA NOW!

Top Articles:

Talk List

The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...

Related Readings:

Re:FYI: Error message (1996) newbie question about zip code search (2003) Loss in Form (1998) [WC 2.1.30] Stack overflow. (2000) Listserver problem (1997) Meet your competition (2000) How can I Add several Items into the cart at once? (1997) WebCatalog & htdig (2000) Running 2 two WebCatalog.acgi's (1996) Expiration of Carts (1997) WebCat2 - Getting to the browser's username/password data (1997) [Announce]: Web server security and password protection (1997) firewall ports for Mac OS X/WebCatalog (2001) including an image through WebDNA (2005) RE: Error: template (1997) test (2002) Setting up shop (1997) Cross OS Compatible HTML emailing (2003) ShowNext (1997) Tcp connect (2002)