#RFM in Social Media

I've been taking classes in Social Media (believe it or not) and they suggest that I should find a search term that others might be looking into. Since I've had major wins in catalog circulation I tried that. Well both terms are a bit too general I think. I use predictive modeling but I find tweets by super models. If I use quotes to combine the terms I find people in financial analysis which I really know nothing about. In the shower this morning I thought about #RFM. It stands for Recency, Frequency, Monetary. It is the foundation of almost all customer scoring. Unfortunately, with most formulas, RFM masks the potential of the best customers and also smudges the the worst into one big glop. We virtually always beat #RFM by working a lot harder, creating a host of additional variables. So, I'm going with #RFM (or e-RFM) and we'll keep exploring with Social Media.

The Tyranny of Time-Saving Social Media

OK, so now I'm engaging on Social Media - as you all know, my tweets feed my hoots which post to my site and my reading list comes up on my blog which now should post to my site as well. I look away from my screen and there are 99+ new tweets from all across the world - on every topic imaginable. These are people I'm "following" so I guess I should read it. I've managed to limit Linkedin (I wonder if this will post there?) to actual posts (not who's now connected to whom) but nevertheless it is endless. TV commercials happen when I'm not watching... but these are friends and people who I want to "connect" with. Is it OK to ignore their precious broadcasts. I guess it is... if they really wanted me to hear about the latest battery-saving Droid tips, they would drop over with a case of beer. LOL John <><

The Post that Ate the Internet?

I am very concerned.  My post feeds my blog, to RSS into other pages, twitter, facebook, linkedin, onto my site which may feed my blog - could this take over the web?

The Adventure Begins (again)

The adventure began a few weeks ago, with the end of a 6 year gig. A very large catalog hired us to pull all their customer names, to relieve the burden from their IT while converting their order entry system. In the process, our analytics were compared with some of the best modelers in the world. We won eleven tests against eight systems but that is another story. Anyway, I took another assignment - trying to grow an e-commerce company with very little spending (something all marketers wish to accomplish). They have great products and a small loyal following. "Lets try Social Media..." My head is starting to swim... partly from staying up until 2am a few nights (because that was the only time the training videos I'd purchased would play)...partly from trying to program joomla, hootsuite, twitter, facebook and linkedin to talk with each other... now that I have this working for myself, I just have to figure out what to say :)

Windows 7 and the death of Microsoft

I'm not foolish enough to predict the end of Microsoft but perhaps, Lord willing, it will shrink back to a more fitting approximation of its name. I have been a Micro-phobe for decades so my previous laptop, purchased four years ago, initially arrived with no operating system... plus the promise of the vendor to help me load Linux onto it. Though that project was a moderate success, many of the accessories, like the DVD player and cam could not be activated. I caved and purchased WinXP and have been relatively pleased with it.

Having learned my lesson, 6 months ago I purchased a shiny new ThinkPad with Win7 64 bit installed. Much to my surprise, since this is my second Win7, little of my legacy work software worked. Repeated efforts to install virtual machines and dual boot XP failed. My shiny machine was basically worthless for actual work. Scanning the usual sources of help, I found many many people in similar straights. (Now in defense of MS, my most important software is an old statistics package from SPSS that runs on 16bit Windows.) So first take-a-way - your key software may not run on Win7.

After successfully hosing up my system, I unpacked the recovery disks - yes ThinkPads still include disks. I installed the Win7 operating system - but this time, it happened the disks were for 32bit. I wouldn't think that mattered, but now everything works.

My point... Microsoft is not interested in supporting your legacy software and they are moving away from it. Make plans...

Just for fun I tried running my SPSS on Linux... and even without any Windows software, it runs just fine. So, if you are interested in preserving your legacy Windows environment, you probably want to look into Linux.

Next time, The Cloud... :)

2010 News

Wow! It has been quite a while since I've posted. We've been busy reinventing cement (http://recocement.com) and beating modeling tests (two of the top modeling companies in the past 3 years) plus several others. A Ph.D. statistician told me this past week that we have a 'modified bayesian system'. He also said, "that happens to be what the consensus now admits achieves the most successful results... lucky you stumbled onto it." Well, I like to think it was more than a stumble... we just kept trying to make it better and better and now its the best. People ask me if it is better than RFM... please :). We had one client that retested and re-tuned their RFM system for 18 months, incorporating what we were teaching them. Nevertheless, when we would meet at the DMA in New York, they would buy dinner. "Why are you buying dinner?" "I'm the vendor!"

"Well, every time we use your system, we find about 20% more names than we would find without it... and they make money."

One of the tests which we contested recently, we generated twice the unique names of the competition. Now often, if you have more names, they aren't as good, but in this case, even the worst segments of ours beat the best of theirs. What this means is that our system is finding the good names in the bad spots and ignoring the bad names in the high segments - that incidentally is the best that modeling can do for you.

John <><

Marketing Database Interface and System for Junction Solutions/MS DAX (Microsoft Dynamics AX)

One of our major clients with more than 11 million customers is successfully migrating from MACS by Smith-Gardener/Ecometry/Escalate to Junction Solutions MS DAX system.

We have been handling their circulation modeling and list pulls for several years. There are many significant challenges migrating from one legacy system to another. An area often neglected is the connection between the old data and the new. Not only does a new file structure require a new interface into the marketing database, but also, new product categories, customer types, division codes, group codes, suppression codes all have to be unified on the database. Most importantly, if customer segmentation is required, old product classifications must be converted to new systems so that customer preferences will include both new and old (so we can figure out what kinds of products they prefer).

Jim Berry, our database guru, said, "Most people don't realize that conversion of an existing marketing database to a new operations system is more difficult than building one from scratch."

Bottom line, if you have either, MACS or DAX, we have an interface and database ready for you.