Archive for the ‘Web Analytics’ Category

Tortured Data – and Analysts

Tuesday, February 9th, 2010

Fear and Loathing in WA

You may recall I wrote last year about the explicit or implicit pressure put on Analysts to “torture the data” into analysis with a favorable outcome.  In a piece called Analyze, Not Justify, I described how by my count, about 50% or so of the analysts in a large conference room admitted to receiving this kind of pressure at one time or another.

Since then, I have been on somewhat of a personal mission to try to unearth more about this situation.  And it seems like the problem is getting worse, not better.

I have a theory about why this situation might be worsening.

Companies that were early to adopt web analytics were likely to already have a proper analytical culture.  You can’t put pressure on an analyst to torture data  in a company with this kind of culture – the analyst simply will not sit still for it.  The incident will be reported to senior management, and the source of “pressure” fired.  That’s all there is to it.

However, what we could be seeing now is this: as #measure adoption expands, we find the tools in more companies lacking a proper analytical culture, so the incidents of pressure to torture begin to expand.  And not just pressure to torture, but pressure to conceal, as I heard from several web analysts recently.

One bright young analyst went “beyond the call of duty” on his analytical project.  The analyst gathered relevant data not just from the WA tool, but from Finance, Customer Service – all around the company.  The report painted a detailed picture of cost to acquire customers through various methods and campaigns, and was presented to the head of Marketing – also the analyst’s boss.

The analyst was told under no circumstances was this report to ever be produced again.  Further, the analyst was told to destroy any “evidence” this project / report ever existed.  And finally, the analyst would now be required to send all analysis through the boss first before anybody else sees it.

That’s shameful behavior for an exec.  And apparently, this kind of thing is happening more and more often.  I’ve heard plenty of “if we want your opinion, we’ll ask for it” stories, but this is the first time I’ve heard so many stories about concealing results.

Here’s a scary thought: what if the stories about web analytics not driving business value are primarily concealment stories?   What if the tool / analysts actually did provide value, which was then hidden from Senior Management?

My concern about this issue is wider than screwed up company culture and management.  What I’m more concerned about is screwed up people, analysts who may come to think this kind of behavior is normal and just part of being an analyst.

This matters because as this new generation of analysts moves to other companies and throughout the ecosystem, these pressure to torture situations could become “accepted” and even spread as “part of the game”.

It is never, ever OK to manipulate or hide the results of an analysis.  It’s not part of the job.  The role of an analyst is to analyze, not justify or conceal bad news.

Now, I realize some folks are thinking, “Yea, that’s great Jim, I’ll just get myself fired by being an analytical hero”.

I’m not saying you should respond to data torture pressure by falling on your analytical sword.  What I am saying is you – and management – need to know this kind of pressure from a superior is shameful, not a “normal” part of being an analyst.  And as soon as you can, you should get a job somewhere people respect your professional opinions.  Don’t have to agree; but must respect.

Like the company you work for?  Ask a buddy in Finance if they could use a web analyst.  Pretty sure Finance would be interested in fully-loaded cost to acquire new customers by source!

What really troubles me about this situation is it’s rarely ever talked about, so could be worse than people might think.  At the very least, Senior Management should know about the potential for this to happen and lay down some rules.  Perhaps even seek some cultural guidance on this topic (here’s a start – Fear of Analytics).

So, I want to put this message out there, perhaps create a resource for people who are looking for information on this topic.  It would be great to have examples so managers can understand and be on the lookout for these situations.  Plus, I’m sure there are some terrific stories out there about either giving in to the torture pressure or resisting it!

What about you?  Were you ever pressured to torture the data?  What happened?  Did you comply?  How did things come out?  Tell us with a Comment.  Feel free to post anonymously, leave out company names.

Control Groups in Small Populations

Friday, February 5th, 2010

The following is from the January 2010 Drilling Down Newsletter.  Got a question about Customer Measurement, Management, Valuation, Retention, Loyalty, Defection?  Just ask your question.  Also, feel free to leave a comment and I’ll reply.

Want to see the answers to previous questions?  Here’s the blog archive; the pre-blog newsletter archives are here.

Q: Thank you for your recent article about Control Groups.  Our organization launched an online distance learning program this past August, and I’ve just completed some student behavior analysis for this past semester.

Using weekly RF-Scores based on Recently and Frequently they’ve logged in to courses within the previous three weeks, I’m able to assess their “Risk Level”– how likely they are to stop using the program.  We had a percentage who discontinued the program, but in retrospect, their login behavior and changes in their login behavior gave strong indication they were having trouble before they completely stopped using it.

A: Fantastic!  I have spoken with numerous online educators about this application of Recency – Frequency modeling, as well online research subscriptions, a similar behavioral model.  All reported great results predicting student / subscriber defection rates.

Q: I’m preparing to propose a program for the upcoming semester where we contact students by email and / or phone when their login behavior gives indication that they’re having trouble.  My hope is that by proactively contacting these students, we can resolve issues or provide assistance before things escalate to the point they defect completely.

A: Absolutely, the yield (% students / revenue retained) on a project like this should be excellent.  Plus, you will end up learning a lot about “why”, which will lead to better executions of the “potential dropout” program the more you test it.

(more…)

Acting on Buyer Engagement

Thursday, January 21st, 2010

Over the years I’ve argued that there is a single, easy to track metric for buyer engagement – Recency.  Though you can develop really complex models for purchase likelihood, just knowing “weeks since last purchase” gets you a long way to understanding how to optimize Marketing and Service programs for profit.

Which brings me to the latest Marketing Science article I have reviewed for the Web Analytics Association, Dynamic Customer Management and the Value of One-to-One Marketing, where the researchers find “customized promotions yield large increases in revenue and profits relative to uniform promotion policies”.  And what variable is most effective when customizing promotions?

The researchers took 56 weeks of purchase behavior from an online store, and used the first 50 weeks to construct a predictive model of purchase behavior.   Inputs to the model included Price, presence of Banner Ads, 3 types of promotions, order sizes, number of orders, merchandise category, demographics, and weeks since last purchase (Recency).

The last 6 weeks of data were used to test the predictive power of the model, and the answer to which variable is most predictive of purchase is displayed in the chart below, click to enlarge:

Weeks since last purchase dominated the predictive power of the model, controlling not only the Natural purchase rate (labeled Baseline in chart above, people who received no promotions) but the response to all three different types of promotion.

(more…)

Choosing the Size of Control Groups

Tuesday, December 29th, 2009

The following is from the December 2009 Drilling Down Newsletter.  Got a question about Customer Measurement, Management, Valuation, Retention, Loyalty, Defection?  Just ask your question.  Also, feel free to leave a comment and I’ll reply.

Want to see the answers to previous questions?  Here’s the blog archive; the pre-blog newsletter archives are here.

 Q:  I am a big fan of your web site and read your Drilling Down book. Great work!

A:  Thanks for the kind words!

Q:  I was wondering if you could help me picking the right control group size for a project of ours?  The population is 25 million telco customers that for which we want to do a long term impact analysis (month by month) in regards to revenue increase versus control group.  The marketing initiatives are mix of retention, lifecycle and tactical/seasonal activities.  We want to measure revenue increase through any of the marketing activities compared to control group.

A:   Great project, this is the kind of idea that can really improve margins if you can find out which specific tactics drop the most profit to the bottom line.

Q:   I have searched the web for some help and found calculators that say: On 25 million and smallest expected uplift of 0.1% and highest likely rate of > 5% the calculator gives 250k (1%).  Is that sufficient to calculate the net impact on the remaining base?  Would be very grateful if you could give me your thoughts.

A:  Well, it could be and might not be…

(more…)

Customer Value in the Freemium Model

Friday, December 4th, 2009

The following is from the November 2009 Drilling Down Newsletter.  Got a question about Customer Measurement, Management, Valuation, Retention, Loyalty, Defection?  Just ask your question.  Also, feel free to leave a comment and I’ll reply.

Want to see the answers to previous questions?  Here’s the blog archive; the pre-blog newsletter archives are here.

Q: You kindly clarified a few issues when I was reading Drilling Down earlier this year – so I hope you don’t mind the direct email.

A: Yes, I remember!

I am working for www.XYZ.com, a social networking / virtual world site based abroad but visitors are 85% US.

Our growth up to now has been mainly viral and in the summer we hit 1.2M UVs operating on the Freemium model with only 5% of our registered users converting to paying customers and a significant portion of our revenue coming from ads.  On average our customers are active on the site for something like 4 months making their first purchase around day 28. 

But to take us to the next stage we are embarking on some marketing for the first time using AdWords and various revenue share campaigns, and of course to do this sensibly we need to arrive at a reasonable estimate of LTV.

A: Makes sense!

Q: To calculate an adjusted LTV I removed all customers with a lifetime of less than 4 months but this gives a low estimate as this calculation ignores the bumper summer months and the extra paid for features put in place earlier this year.  Calculating LTV using ARPU and monthly churn (not sure how to calculate this in our environment) gives another different estimate.  Is there any help or advice you could perhaps give us?  If not in the US then perhaps you could recommend somebody abroad – can’t find anything in the literature relevant for start-up like us.

A:  It sounds to me like you’re trying to make this too complicated, at least for the place you are at this time.  Monthly churn and the “28 day” threshold are nice to know on a tactical level, but LTV is more of a Strategic idea that does not necessarily benefit from analysis at that level.  And you may not really want LTV, but a derivative that might be more helpful.

(more…)

“X Month” Value

Friday, November 20th, 2009

The basic concept of LifeTime Value (LTV) was ably outlined by Seth Godin in a great post here.  If you know the average net value of a customer is $2500 over their “Life”, why would you not spend  $50 (or $200, really) to acquire each one?  As long as you stuck to the model, your company would be insanely profitable over time.

Their are 2 primary challenges to implementing this idea.

1.  “Over time” is a concept many management folks have a hard time embracing; what matters are the profits this year, or this quarter, or this month.  Unless the whole company embraces an “over time” measurement approach it is difficult for Marketers and Analysts to drive towards programs and practices supporting the LTV outcome.

2.  The $2500 is an average figure.  Most customers are worth less; 10% or 20% are worth much more.

Most people I talk to embrace the general idea of LTV models intuitively.  It’s really a cash flow concept, isn’t it?

So Financial people get it right away, and if Marketers could align with it, there would be no conflicts and the Marketing budget becomes virtually unlimited.

In fact, many folks in the PPC world follow just this model – they have unlimited budget as long as each conversion costs no more than “X”.  Because the company knows if it spends no more than X on a conversion, it always makes money.   Marketers and Analysts involved with these “Cost < X” PPC programs love them, because Management loves them. 

And Management loves them, why?  Because the CFO loves these programs  Why?  Because they are based on Cash Flow analysis, which CFO’s understand very, very well.

So then, what will it take to get more acquisition budgets like these Cost < X  PPC programs?  We have to address the two challenges above:

(more…)

Awareness versus Persuasion

Wednesday, September 23rd, 2009

In the early days of Home Shopping Network (live TV, not online), we were doing some ethnographic research and started to find “physical clusters” of customers – neighbors or people who worked together.  For example, one of these groups was nurses at hospitals,  especially nurses  who worked the night shift.

We looked for the most active member of the cluster (our “thought leader”) and asked them if they would help us with a “member get a member” program.  Would they be willing to distribute discount coupons to their friends, especially ones who were not already customers?  Time after time, the answer was:

“Honey, all my friends are already customers of yours”.

We launched the program anyway, because it was a pet project from upstairs  – I was a junior marketer at that point so I couldn’t kill it ;)  The program never, ever worked, no matter how hard we tried.  It generated very few new customers while giving lots of discounts to people who were already active buyers.  Basically,  the cost of those discounts overwhelmed the value of the new customers generated.

Apparently a similar thing happens online with Social marketing.

As part of a WAA program that reviews academic research for WAA members, I was able to take a look at a paper titled:  Firm-Created Word-of-Mouth Communication: Evidence from a Field Test by David Godes and Dina Mayzlin.

(more…)

Adoption and Abandonment

Friday, August 7th, 2009

Out of the Wharton School we have a nice piece of behavioral research on the effect speed of Adoption has on longer-term commitment.  The article, The Long-term Downside of Overnight Success, describes research finding “the adoption velocity has a negative effect on the cumulative number of adopters”. 

This research dovetails nicely with a lot of the topics discussed here on the blog lately, so I thought I’d use it (with a nod to Godin’s post on Strategy vs. Tactics today) to provide some fodder for thought.

First, the importance of Psychology in Marketing.  So many of the “discoveries” arrived at through  brute force testing of Online Advertising are already well known in the greater discipline of Marketing through Psychology.  For more on this read “The Other 3P’s” and if you’d like to do something about lack of knowledge in this area, make sure to read this comment on source books.

Second, this research is a great example of isolating the true drivers of behavior.  The idea of looking at baby names to isolate the real behavior from “technology and other commercial effects” while including “symbolic meaning about identity” results in a broad, Strategic-level answer to the question, not a Tactical one. 

Why is this important?  It means the results can be applied across a host of different Marketing situations, rather than only a specific one. 

(more…)

Norms of Reciprocity

Friday, June 26th, 2009

Social Marketing Doesn’t Rely on Social Media

Do you believe human beings share certain fundamental traits that define “being human”?

If so, do you believe that human beings tend to behave in certain ways under certain circumstances?

If so, do you then believe since human behavior has these tendencies, it can often be predicted?

If so, then do you think perhaps the study of Psychology and Sociology might provide you some clues to creating successful businesses, campaigns, products, and services?  While your friends and competitors are all iterating their way into oblivion?

On the web, time and time again, we see the same themes repeating.  Yet with each introduction of a new technology, these themes tend to be treated like a new discovery, even though the theme has been well established in the past.

Norms of Reciprocity is a constant human theme.  You may know the expression of these norms as ”Sharing”.  Web old timers will probably recognize this idea as “Give, then Take” from the I-Sales discussion list as early as 1995.  In various forms, this theme goes back to the beginning of human history, all the way back to the handshake and other greeting gestures.  This same theme is embedded in countless Religions all over the world: “Do onto others as you would wish them do onto you”.  At least a couple centuries old, this idea.

Norms of Reciprocity simply means this: When you do something nice for a human being, help them in some way, this human tends to feel Gratitude towards ”the doer” and tends to do something nice back.  Gratitude drives the desire to Reciprocate, because it’s just what humans do, it’s normal, a “norm”.

Norms of Reciprocity.

(more…)

Analyze, Not Justify

Friday, June 19th, 2009

Does this issue affect the Web Analytics Maturity Model?

A conference call with a Potential Client last week jogged my memory on a couple of events that happened during the flurry of Web Analytics conferences this Spring.  Here’s a portion of the call…

PC: “We’ve tried proving the profitability of our Marketing efforts and can’t seem to get the numbers working correctly.  So Jim, what we’d like you to do is take all this data we have, and justify the Marketing decisions we’ve made by proving out the ROI.”

Jim: “I’m sorry, did you say justify?  To me, justify means “find a way to prove it works”.  Is that what you are asking me to do?  Wouldn’t it be more beneficial to analyze the results, and then optimize your Marketing based on these results?”

PC: “Jim, around here we’re pretty clear our Marketing works, and Management knows this.  But Finance is asking for some backup, some numbers to justify the spend, not to analyze it.  We don’t need analysis, we need your ‘expert credibility’ to help us out with this.”

Jim: “I see,” thinking this is not a job I’m going to enjoy.  It’s the old ‘buy an outside expert’ routine, which I detest.

PC: “Jim, the team is united behind this mission, are you on board?”

Jim: “Well, perhaps I could be on board, as long as what you want is an analysis, which may also justify the decisions you have made.  But it might not, so I just want to be clear on what…”

PC: “You  know what Jim?  I don’t feel we’re going to have a fit here, I’m getting you’re not a team player.  Thanks for your time”.  CLICK

Sigh.  I’m actually grateful they hung up, I really dislike explaining to people why I won’t work with them.

(more…)