Tag Archives: Predictive Analytics

You need to discover fire before you can launch a rocket!

I was recently in Germany attending the Gartner Data and Analytics Summit in Frankfurt. I had a chance to meet many clients and prospects who are struggling to get their analytics projects off the ground. Every client wants a moon shot – to leapfrog immediately to space flight without having even discovered the wheel.

I also had the opportunity to meet with an amazing thought leader in analytics, Alexander Thamm and we discussed this topic in depth.

Alex Thamm, the CEO of Alexander Thamm GmbH has helped some of the leading companies in Germany achieve amazing results with analytics. In our discussion, Alex shared what he is seeing: organizations want the nirvana promised by AI and Cognitive technologies. The challenge with many of these organizations is that they do not even have the basics down. A recent HBR article backs this view up. Many organizations either do not understand, don’t appreciate or even worse, are in denial of what is really needed to be successful to move up the analytics curve.

So, as I sat there at Alex’s headquarters in Munich, Germany I thought about an analogy I have shared multiple times because I have too frequently experienced vendors pitching the magic of artificial intelligence and analytics only to see a deer the headlights type of gaze.

So I shared the following analogy with Alex to see if this is what he is also seeing.

Explaining artificial intelligence and advanced analytics is like explaining the space shuttle launch to cave men. First you start a fire. Once the fire is going and strong enough you need to get yourself a tin can. You take that tin can and fill it with some type of combustible fuel and set it on top of the fire. Enough pressure will build up and send the tin can into the sky. The tin can will go around the earth. Voila! You have space travel. The cave men tend to sit there completely astonished…..amazed really. After overcoming their initial amazement….they eventually say: “Wow, that is absolutely amazing!!!!! Incredible! So, how did you get the fire??

Alex’s reaction was clear. This is what he is seeing too. He like the analogy so much he recommended I share it as a blog post.

I am not saying that most organizations are filled with cave men. I AM saying that we as vendors owe it to our clients to help the progress along the analytics curve. If we are truly dedicated to our clients’ success, sometimes we need to have the tough conversation with the client – Yes, I understand you want to have artificial intelligence across your organization (moon shot). I recommend you start with getting your data collection and management processes (fire) down first and then we can build up to your moon shot.


Part 3 – 4 ways analytics can deceive you

Last week I shared how two organizations have suffered from lack of the right data and incorrect data.  This week I finish off the 3 part series…… (Part 1, Part 2)

The last two examples relate to analytics governance. Analytics adoption began as IT-led, coordinate projects but as desktop and cloud applications became more accessible to individuals, their use spread across the enterprise. Michael Goul is an associate dean for research and professor of information systems at Arizona State University’s W. P. Carey School of Business, has spent the last few decades studying artificial intelligence and business analytics. While he agrees that data science has the potential to revolutionize commerce, he also thinks too many companies are rushing headlong into the field without putting proper governance systems in place, and in some cases this has led to disaster.

In our third example, the lack of Analytics governance provided an opportunity for fraudulent behavior. A media company was going through a rapid growth phase. They expected that their customer satisfaction, measured by their NPS score, would likely dip during this time. They were watching close for an indication that they need to ramp up their investment in customer service. However, as they grew, their NPS remained curiously constant – which seemed rather odd. After 9 months, the CEO was so concerned that they might be missing something that he brought in an outside consulting firm to help them understand how they delivered the same level of customer service while they expanded so rapidly. It turns out that the person who created the dashboard for the Executive team was using an ungovernable desktop BI solution and was being paid a bonus based on the NPS score performance. When the score started to slip, the individual who created the dashboard was faced with losing his bonus so he opened up the spreadsheet that feeds the dashboard and nudged the number up manually so that he still got paid his bonus. What started out as a small adjustment, grew to a 40% variance over the 9 month period. How many customers got frustrated with the declining level of service in that period and switched providers? Despite their attention to this metric, they got blindsided by not knowing. (misrepresented)

Our final example is the opposite side of the same coin – related to analytics governance. Executives have become reliant on dashboards as single-screen “snapshots” of performance. But dashboards are not the magic view some managers treat them as. Although they can convey important measures, dashboards cannot always provide the nuance and context necessary for effective data-driven decisions. The data can be 100% correct but the visualizations can be very misleading. Here is an example from Harvard Business Review. It is for a large package delivery company who wanted to reduce vehicle accidents by offering drivers the option to upgrade their GPS to a system that would help them avoid high-risk traffic areas. After monitoring drivers’ behaviors, a frontline manager checked the dashboard and found, to her surprise, that the accident rate was actually higher with the upgrade.Picture1

At first glance, it appears that drivers who upgraded their GPS were in more accidents, which might lead someone to suggest they downgrade back to what they had. In reality, the upgrade was actually quite effective and the manager would have seen this had they compared accident rates for “safe” drivers versus “accident prone” drivers.


For both groups, the upgrade made them safer. So why did the accident rate increase for the entire fleet of drivers while decreasing for each group? Because in this case almost all of the accident-prone drivers chose to use the upgraded device and almost all of the safe drivers kept the old device. Preexisting driver behavior was confused with the effectiveness of the upgrade.

THE REALLY INTERESTING THING in this case? The visualizations were accurate as was the data displayed.  They just did not show the whole picture when looking at all factors leading to accidents. Joel Shapiro, Executive director data analytics program at Northwestern’s Kellogg School of Management, says “Perhaps the greatest danger in using dashboards for decision making is in misattributing causality when comparing elements on the dashboard.”

These scenarios are real and they are happening every day.  Might they be happening in your organization? …..

So what the heck do you do with Watson Analytics?

It has been a little over a year since IBM’s Watson Analytics was released.  Watson Analytics, a smart data discovery tool, continues to deliver easy analytics for everyone.  Some competitors are trying to catch and imitate what IBM has delivered but frankly, their latest news is more “snooze worthy” than newsworthy.

Because Watson Analytics is so easy to use and visually appealing, IBM has seen strong adoption (1 million plus users!).  Over the past couple of months there have been some amazing additions to an already fantastic product (see Watson Analytics for Social Media).

Watson Analytics is so powerful and can address so many different uses cases.  I am frequently faced with the following query:  So, I get it!  I have a ton of data and I need a way to make sense of it.  But what exactly do I DO with it??

This really depends on what your role is.  In our work lives we each fit into some type of role:

  • sales guy
  • marketeer
  • finance professional 
  • operations director
  • supply chain manager
  • head of strategy

So, as you read this and think about what YOU can do with Watson Analytics, think about the data you need to better visualize.  Think about what finding patterns within your data could do for you.  Finally, think about how adding social media data to your analyses can make you analyses event more effective.

But most importantly let’s put this in context of some of the major roles and use cases that will matter to you:

Screen Shot 2016-04-14 at 9.52.01 PM.png

Over the next several weeks, I will be sharing the details behind some of these specific use cases and providing a “how to do this yourself” approach.  Check back to hear more about how #watsonanalytics can help address many of these use cases.



Analytics for everyone? For free? Get out!

Making analytics accessible to all users has been a challenge for years.  Companies like Tableau and Qlik have done a decent job of reaching finicky business users by providing visualizations and data discovery capabilities.  They are largely one trick ponies in each of their areas of strength.  That is just not going to work for me. I am an only child with some strong Veruca Salt tendencies (Think, I want it all and I want it NOW!).  So I am particularly happy to talk about  the Watson Analytics announcement today.

Here is what the market is saying:

As a former marketing operations leader, I am thrilled to see these types of capabilities being offered directly to marketing, sales, human resources, and finance professionals.  For the first time these business users can have all of the analytics capabilities they need in one location without ANY technical knowledge required.  From snazzy visualizations to predictive analytics Watson Analytics offers the analytics you need all in one single location to answer questions like:

  • How can I make sure I target the hottest leads?
  • How can I increase the value of the customers I already have?
  • How can I create more successful campaigns?
  • How can we attract and retain the right employees with the right skills?
  • How can we create and keep top performers?

I have neither the time nor the patience to play around with analytics tools.  I need to get the answers to these from my data and make decisions NOW.  Watson Analytics gives me what I need and it is cloud based….so I don’t need to download and install anything!  Woohoo!

Watson Analytics is starting with a beta program.  I am looking forward to getting my grubby little mitts on it over the next several weeks.  You too can get an early glimpse by registering for the beta here!

Analytics is not a thing you “do”…it’s a way of life.

I recently attended conference in Singapore

A beautiful view of Singapore at night.
A beautiful view of Singapore at night.

where I had the chance to listen to thought leaders discuss the need for analytics to address what Gartner refers to as the “nexus of forces” – CloudSocialMobile and Information.   Mychelle Mollot took a deeper dive on the “information”  force in the Big Data and Analytics keynote.

As a marketer I grapple with explaining these nexus of forces and the need for Big Data and Analytics everyday.  Everyone seems to have an opinion on exactly what Big Data is and why it is so important.   It was in fact several customers and partners  who reminded me of several things: Continue reading Analytics is not a thing you “do”…it’s a way of life.

Finally! Predictive Analytics even I can use!

When I was running marketing operations, I had hired a statistics expert to come into the organization to help me make sense of all of our marketing data.  As I worked with this excellent “stats guy”, I realized that predictive analytics can be pretty darn complicated and required a more specialized skill set than I had originally thought.

This all changed on Tuesday June 11, 2013.  IBM announced the availability of IBM SPSS Analytic Catalyst – a statistician in the software – which makes predictive analytics on big data available to mere mortals like myself.  I was admittedly skeptical that this new offering could actual make predictive analytics accessible.  There had to be some gotcha….  well, I was wrong (not something I admit often!).

So, what does this thing do exactly? Let’s say you are a marketer and want to understand which customers would likely “churn” or leave for your competitor.  This is where Analytic Catalyst comes in.  There is a three step process to get the answers you need.

  1. Add your data (csv file)
  2. Select the field you would like to predict
  3. Review plain English results

Sounds simple?  Well it is.  Review a demo here to see how this can help marketers, customer service teams, sales and other organizations find the small data within the big data!

Marketing Win Revenue = Autopsy on a dead guy!

Marketing organizations are not exactly known for their desire or ability to measure the outcomes of their activities.  As a former director of marketing ops I ask myself:  Is it ability?  Is it lack of desire? Or, is it simply that marketers just do not know where to start?   As with any complex organizational discussion, it is likely a mix of the three.  My experience leads me to believe that most marketers just do not know where to start.  And those who have started are likely choosing all the wrong key performance indicators and metrics.

When I was still in my marketing operations role,  I would frequently hear from the marketing teams about how much win revenue we got from our marketing programs.   We were very fortunate that our organization had standardized our reportingCWprovided analysis capabilities to our program managers and the management team with standard dashboards for a single version of the truth.  Managing your business via lagging indicators is fine when you hit your target which we frequently did.

Still, something troubled me.  We knew what had happened and why but still could not answer what WILL happen.  Essentially, we could do an autopsy on a dead guy but could not predict what would kill him…Something had to change.

As a B2B marketing organization we subscribed to Sirius Decisions demand waterfalldemand waterfall In spite of this, our key performance indicators, metrics and business intelligence platform continued to reinforce the behavior that win revenue is king.  Our CMO was no longer content with seeing the results at the end of every quarter.  He wanted to know within a reasonable margin of error (+/- 10%) where we would end up each quarter.  We needed to change our behavior to do this, we needed to change our metrics…moving up the waterfall all the way to inquiries…and beyond.

Applying Predictive Analytics to the Problem

We were swimming in marketing data – contacts, inquiries, marketing qualified leads, sales qualified leads and win revenue.  What we needed to do is figure out what patterns lead to a person progressing all the way through the demand waterfall.   By applying predictive analytics, we were ultimately able to predict within +/- 5% how much win revenue would come from our marketing activities.

We took our contacts, marketing outreach data (emails, web, live events), inquiries, pipeline and win revenue and brought this data together in a predictive model. SPSS-Modeler This predictive model identified those contact job titles, marketing offers and marketing channels which would yield a successful marketing inquiry.  From here, we were able to extend the model to be able to forecast which contacts would lead to inquiries and which of those inquiries would convert to marketing and sales qualified leads.  With this knowledge in hand, we were able to conduct “what if” analyses and predict the optimal mix of contacts, tactics, and channels to meet or exceed targets.

Changing Behavior

Now that we were able to predict the mix of contacts, offers and channels, we decided to change the way we measured marketing moving WAY up the demand waterfall.  We implemented the following metrics:

  • # of contacts within each segment (industry and geo)
  • # of net new contacts added to the database (as determined by email address)
  • # of inquiries
  • Conversion % from inquiry to marketing qualified lead
  • # of marketing qualified leads
  • Net new pipeline created ($)
  • Conversion % from marketing qualified lead to sales qualified
  • Acceptance % by sales rep

Notice – no win revenue.  As a marketing organization we believed marketing is responsible for finding new names and delivering net new pipeline while sales is ultimately responsible for revenue.

We updated our reports and dashboards I spoke about in a previous post to reflect these new metrics and assigned targets to each marketing team based on our predictive models.

My thought: I would rather know early on that I was off target versus having to do the post mortem…what are your thoughts? Send me your thoughts @BrendanRGrady on Twitter.