Miglautsch Marketing has pioneered automated modeling since the early 90's. Many of our clients produce four or more models per month, we have built more customer models than virtually any other direct marketing company. MMI is considered one of the world's leading modeling companies. With clients such as Guitar Center, Adobe The Bay and dozens of catalog companies, our systems have proven themselves in the real world of direct mail marketing for some of the biggest mailers in North America.  In the past 6 years we won 11 blind tests against 8 of the world's best modelers.  Click->John's Welcome to hear more.

Good Data in Big Data?

Created on Friday, 27 July 2012 00:00
How to Find the Good Data in Big Data

Guest Blog by Ray Schutz, TellAllMarketing

There’s one thing wrong with Big Data. A lot of is is “a little bit like overhearing a conversation,” says John Miglautsch, chairman of Miglautsch Marketing Inc., in the second in a series of videos on data. .

“When my wife is talking with a friend and it’s clear that someone died, I’m starting to feel very emotional about it,” he says. “It turns out it’s someone different from what I thought—it was their parakeet. Not that I have anything against parakeets.”

 

So how do you sift through the overload and find actionable data (i.e., data you can use to make money)?

Go back to the direct marketing basics. Start with RFM, then move on to other forms of information.

“Direct mail 101 says there’s two ways to make money,” Miglautsch says. “One is to refine your list. Either eliminate people that are not going to be interested or find people that look bad and bring them in because of something else you know.”

Miglautsch started doing this years ago on a TSR 80 that didn’t even have a hard drive. He noticed that clients like Gurney’s Seed & Nursery, they had boxes of paper containing customer information.

He recommended that they key the data in, and they handed the boxes to him so he could it himself.

Some firms knew how many times a person bought. But they didn’t always know what the person bought. So Miglautsch started adding more nuanced item data to that the accumulator data.

Yes, there were mistakes. Hudson’s Bay Co.’s data showed that a customer bought a pair of jeans for $92,000.

“Somewhere between Alberta and Toronto, a bolt of lightning added a few zeros to that transaction,” Miglautsch laughed. “We could find it, they couldn’t find it.”

It wasn’t so easy to categorize items. Thompson Cigar might classify a cigar as “a Maduro, a Dominican, 7 inch, 42 ring,” Miglautsch continues. But Johnson Murphy Shoes, which also sold cigars, might describe it as an accessory item.

Once that’s been done, though, you can add contact data and other forms of information.

“That’s really the heart of where additional data is going to help us,” says Miglautsch, the author of Spinning Straw Into Gold, A Guide to the Magic of Turning Data Into Money.

He concludes with this: “One analyst said, very poignantly, that the advantage of Big Data is we might find little pockets of good data.”

 
 
 
10
 
10
 
10

BigData Interest Explosion!!

Created on Monday, 28 January 2013 04:28


BIGdata Video

Because I frequent analytical groups, blogs and virtual cafe's, I have been inundated by the term “BIGdata”. And as the Greek poet Menander says: “Bad company corrupts good character.” my vision could be a bit distorted. I got so excited I even did two videos about it – my most watched. http://bit.ly/BIGdataJRM So when I decided to explore Google's new Search Trends I thought it would be fun to see the history of this term.
 
Chopping the scale off makes it look more impressive
This also gives you a nice view of how this new tool works. As you can see, BIGdata has gone from basically non-existent into the stratosphere! I was so excited to be part of this 'trending' (here's the trend on that). Well, as with most analysis, if it looks too spectacularly exciting, you might want to do some comparisons. I noticed that the scale was 0-100 and that seemed like a pretty low number... then I realized that this was a relative scale. So it can be millions and millions of searches – or not... hard to tell.
Chopping the scale off makes it look more impressive

There is an additional feature which allows you to compare a second, third or fourth (not sure the limit) search term. I thought that might give me some greater perspective on this exponential growth in BIGdata (of course it is BIG, look at how it is growing).
 
 When we compare it with 'Database Marketing' we see that there has been a huge decline in what once was a trending phrase. I remember discovering that I could charge almost double if I talked about 'database' instead of the tired 'Direct Marketing'. Google doesn't offer trending before 2004 but this graph suggests that it may have peaked around that time. What the comparison graph shows is that even compared to a pretty 'niche' term like 'Database Marketing', BIGdata hasn't really gotten much use outside of our analytics cafe group.
 
As I mentioned in my video Data Science video, the first and most important character trait is curiosity... so of course I had to go deeper.
 
Even though most of my friends and readers will recognize the “Database Marketing” term, I decided to compare still a more general phrase to see what its scope might be.
 
You can see that the simple term, “Database” while declining, crushes the other two terms to such a large extent that both completely disappear. What this suggests to me is that many many more people are doing things with data and databases than are concerned about 'Database Marketing' and/or BIGdata. Which is a super great illustration of the problem.
 
We want to build data, work with data, pile up database but we really do not want to take much time letting marketing do much with it. Which is also why we find it far easier to make money with data when we do not work within the traditional database environment. And that 'out of the box' thinking is why we beat the best modelers in the world. So keep your attention on your work, your data and get some real data scientist to help you make sense of the back end.

Page 1 of 4

John's Blog

John's Best Tweets

Powered by Supremo Hosting
Scroll to Top