Hi! Thanks for dropping by my blog! It is much appreciated.
In my last blog post I talked about the critical roles that SeETL and BI4ALL play in our ability to guarantee Outstanding Success for a Data Warehousing project. If you do not have these two tools at your disposal on your projects then you should have something very like them.
If you want us to review the tools you are using and give you an honest appraisal?
We would be happy to do that.
Today’s blog post is about the next step.
What do you do once you have got your data warehouse?
After all? A data warehouse does nothing for you until you make some decisions, make some investments to change the business, and see how that works out.
If you have not watched the video below on “What is Data Warehousing All About” ? Please stop reading this blog post and catch up on this video. I will presume you know the content of that video in this blog post.
There is one for business managers below
Ok. You are here. I assume you know all about “The Management Decision Making Process” discussed in the videos.
Simply put. Your data warehouse has no value until decisions are being made based on the data in the data warehouse that can ONLY be made based on the data in the data warehouse. Then and ONLY THEN can it be said that the data warehouse was responsible, indeed invaluable, in making the decision possible.
And it is only when the Data Warehouse is invaluable and necessary for the business to run that it will be widely adopted and used. If it is just “another reporting tool” then it might well be sidelined further down the line.
The #1 way to make sure that the data warehouse becomes indispensable and guarantees Outstanding Success is to use if for sales and marketing work.
If you put your data warehouse to work in sales and marketing and you make it mandatory for your marketing staff to build their marketing plans using the data warehouse you will almost certainly guarantee Outstanding Success.
I am reminded of a client where the Director of Marketing of the subsidiary was “being nice” to his marketing staff and not making usage of the data warehouse mandatory. The marketing analysts could build their marketing plans using whatever tools they liked and still get their budgets approved.
After about 6 months of this it was clear that the marketing staff were in great fear of putting the marketing campaigns in to the data warehouse for fear the campaigns might not work out how they forecast. It was much easier to hide failed marketing campaigns if they were not in the data warehouse. It was also the case that those who were successful were trying to hide their success from their colleagues so as not to share.
Eventually the Marketing Director held an “town hall” meeting and told his staff that he absolutely would never make anything mandatory as an “edict” because that was not his style.
Everyone was free to use whatever tools they wanted. It was just that if any member of staff did not copy the forecast for their campaigns to the data warehouse they would not get their budgets approved.
Not mandatory. Just no budget for your campaigns.
Everyone got the message.
Of course, as the saying goes, “then a funny thing happened”.
What had been happening was that marketing analysts had been competing with each other. So they kept their “secrets” of how to create successful campaigns to themselves and let the other guys fend for themselves. They were not sharing. It is easier to be seen as successful if your colleagues are failing, right?
Once we had the forecasts and budgets of the campaigns in the data warehouse as well as the results we were able to more easily see who was doing well. The Director of Marketing started making a point of congratulating those people who were doing really well in front of all their colleagues.
Of course, now the guys who were struggling knew who the best marketing analysts were and so they pestered them with questions and asked for tips. This new social status worked very well and soon the performance of ALL the marketing analysts was improving.
So. If you are going to make your data warehousing project an outstanding success?
You want to aim it at “Sales And Marketing” in the first instance. Unless there is a compelling reason Version 1.0 of your data warehouse should be all about “Sales and Marketing” to get the early wins on the board.
Anything where the data warehouse can contribute to “sell more stuff” whatever is the “stuff” your company sells is where you should put the bulk of your early exploitation projects.
Now, I know that the Chief Financial Officer is very likely going to want to use the data warehouse for “finance” and is going to be looking for “cutting costs” using the data warehouse. There is no better way to snatch defeat from the jaws of Outstanding Success than to make the early exploitation projects all about finance or all about cost cutting.
People will soon come to hate the data warehouse as a symbol of “cost cutting”.
You can not cut costs to Outstanding Success.
You do not get Outstanding Success by producing a P&L a few minutes faster.
You get Outstanding Success by “selling more stuff”.
So what you should work on is Campaign Management, Sales and Marketing, Customer Profiling and Customer Segmentation. Those are the areas where you can take some of the Marketing budget and you can put it in to analysis using the data warehouse to develop algorithms for “next best offer” and then you can run those “next best offer” campaigns with some form of coupon or feedback technique so you can tell the effectiveness of your campaigns.
I have seen some truly astonishing results from doing this.
The very first time I saw a 20% purchase rate from direct mail for something so boring as an accidental death insurance policy back in 1993 I nearly fell out of my chair. And that 20% purchase rate was from people who had NEVER bought a product from us..EVER!
We all thought it was a mistake at first. We actually went and checked with the data entry people to see the letters that were coming back to verify for ourselves that these were, indeed, the forms we had mailed out come back to us.
Sure enough. We were getting 20% purchase rates. So we tested again. And again. And again. The response rates stayed stubbornly at 20% purchase rates for direct mail.
So we rolled out the campaign and through the life of the campaign rates averaged 18% purchase rates for the letters sent out. It was only a year later that we discovered that the “return to sender” mail was being thrown in the garbage making our results even better.
It took us a full year to question the fact we were getting no “return to sender mail”. We knew the guys in the call centre worked hard to keep the addresses of customers up to date. But no “return to senders” in a whole year was not believable. So we investigated.
Sure enough. There was a guy in the mail room who grabbed all the “return to senders” and was just throwing them in the bin. There were hundreds of these coming back every week and he was simply throwing them in the bin because no one told him that we wanted them so we knew which customers had moved and not told us! Sigh!
So we were never able to find out just how good our campaigns were because we never knew how many of our letters were “returned to sender”. Once we were marking the “returns to sender” as not a valid solicitation and not to be counted one campaign even got to 30% purchase rates for direct mail!
Sean Kelly and I did one project in Moscow where we mailed out coupons for freebie give aways for trial products. The take up of the coupons was nearly 100%. Russians sure love a freebie give away! We constructed these campaigns of three freebie giveaways for a product and then stop the giveaways.
Sure enough, we were getting retention rates of nigh on 30% for the product after the promotion ended. Fully 30% of those who got the giveaways went on to become regular users of these products.
Those are INSANE numbers.
When you have 25,000 products on your SKU list you can soon see the power and benefit of pointing your data warehouse at the “sell more stuff” problem.
Every extra sale you can get using the data warehouse is extra profit as long as you are selling products at a profit. Which is not always the case by the way.
So. If you want your data warehousing project to be guaranteed Outstanding Success?
Once you have built it and you have good data flowing in to it. Use it to figure out “next best offer” on a mass scale and improve your algorithms for “next best offer” as much as you can.
You will be making money on your data warehouse investment in no time.
True story.
I am often asked:
“Peter, what is the most successful data warehouse project you have ever done?”
The one that comes to mind as the most successful project ever was one where the data warehouse was paid for even before we finished building it. That is pretty damn successful.
If you have not read about it before on this page? The story goes like this.
Australian Consolidate Press (ACP) was one of Kerry Packers companies back in 2000. Kerry Packer being the richest man in Australia. ACP was in trouble. Kerry himself called it “a dog with fleas”. James Packer, Kerrys son, was supposed to be looking after ACP and one of his best friends was running it.
Kerry stepped in and fired Jamess friend and installed John Alexander as the new Managing Director. This happened right before the start of our project! John was not too sure what he had gotten himself in to because of the poor performance of ACP.
ACP published about 90 magazines and books and had 37% of the Australian magazine market at the time. But in 2000 the magazine market was hyper competitive and the Internet was eating in to profits as well. A perfect storm of competition that was doing ACP a lot of damage.
ACP had a sister company called National Distribution Management or NDM. NDM were responsible for distributing all the magazines to the 7,000 or so outlets around the country. Australia being a very big country!
Because NDM were already going to 7,000 outlets at least once a week they also bid to deliver other peoples products to those outlets.
As things happened Readers Digest, who were still pretty big back in 2000, had put out a tender for 5 years of delivery. NDM had bid on this and lost.
The first we knew of it was when the Managing Director of NDM came to our CIO and asked him if he had any ideas. He said no but that he should talk to myself and my lead Business Analysis Consultant, one Clint Morgan, who is a great guy and we are still great pals.
We were invited to a meeting to brainstorm ideas as to how to try winning the Readers Digest Account even after we had been told we had lost. No idea would be too outrageous so all ideas were welcome. The Managing Director of NDM had secured a promise from Readers Digest that they would give him one last opportunity to pitch. So it was desperation stakes.
Clint put forward the idea that we should promise Readers Digest that we would have our data warehouse calculate the sales volume for Readers Digest by 9am every morning and have it available on their desktops in their offices.
When the Managing Director of NDM asked Clint “can we really do that?” His answer was priceless.
“How the hell would I know. Besides, it doesn’t matter if they don’t buy it, right?!”
Sure enough. Readers Digest bought the 5 year delivery deal from NDM. Our promise to give them daily sales figures on their desktops was cited as a “major differentiator” that no one else offered. So then we had to figure out how to do what we promised.
Clint was anointed as a “marketing superstar” and when he was retrenched from Informix at the end of the year ACP were only too happy to offer him a job.
Now. I was never told how much a 5 year deal to deliver Readers Digest to 7,000 outlets was worth. I was a vendor after all. But the data warehouse project was costed at about AUD300,000 for the 3 months of development work and the first 3 months running costs.
You can be very sure that the profit from the Readers Digest deal was FAR more than the AUD300,000 ACP were paying for the project.
The results from that project were staggering even by the standards of my clients.
Over the first two years of operation with “marketing superstar” Clint Morgan at the helm of the data warehousing projects, ACP doubled their gross profit and went from 37% market share to 42% market share with the commensurate increase in advertising revenues from their magazines.
Kerry Packer, the richest man in the country, who had called ACP a “dog with fleas” just two years earlier, who had fired the Managing Director and installed his own man, called ACP “the Jewel in the corporate crown”. Kerry promoted John Alexander who eventually went on to head up the Packer group of companies.
When ACP was eventually sold off by James after Kerrys passing it was rated as one of the top 10 publishing houses in the world.
That is what I call “Outstanding Success”.
ACP was not the company that made the most money from a project that I sold and implemented. But they are the company that had the most spectacular corporate turn around and they had the most high profile of owners. Kerry Packer.
So yes. If you aim your data warehousing efforts at “sell more stuff” once you have it up and running? Your chances of “Outstanding Success” are vastly improved.
Whether you use our tools or someone elses tools?
Whether you build your own data warehouse model or buy one?
If you focus on “sell more stuff” you will have a much greater chance of “Outstanding Success”.
If you got this far? Thank you for reading and/or listening. I hope this was useful to you.
Best Regards
Peter
