Wednesday, January 23, 2008

3) Activity Control Versus Outcome Control

Now we are getting down to the nitty-gritty of why the majority of CRM initiatives fail. For most CRM initiatives to succeed, users must be willing to record their activities with customers. The benefits in doing so cannot be denied however in most cases, users strongly resist entering their customer activities into CRM, forgoing the benefits of doing so and despite their assurances that they will use the tool. From my research, I now believe this "Trust Dilemma" is articulated very well in a quote by Douglas Hartle

"It is a rare dog that will carry the stick with which it is to be beaten."

Perfect Versus Imperfect Knowledge (Outcome control versus activity control)
Typically, evaluative processes in business are focused on outcome controls such as sales numbers and product mix because sophisticated ERP systems should provide perfect knowledge of these results.
Perfect knowledge in activity control is rare in that it involves knowing exactly what steps were done by individuals that led to the (overall) documented outcomes, as well as the results of each individual activity.

Current examples of perfect knowledge in activity control include:

  • Professional televised sports where the activities of each individual player and position are recorded during play using video and manual documentation by others. Combined with the final game result and individual player statistics, this information exists as perfect knowledge of individual activities and overall team performance.

  • High level chess where player moves and counter moves are documented and published along with results. 2,138 games played by chess master Garry Kasparov, against all opponents, have been meticulously recorded showing each move and countermove by each player, and the final result of each game.
    The benefits of perfect knowledge of past activities (and results) by coaches and upcoming opponents are obvious. Such knowledge allows for decisions to be made and strategies to be created that can dramatically affect future outcomes. An important note
    here is since, in these two examples, every activity and result is always recorded without input from the players themselves, players cannot choose to “opt out” of activity controls, if they wish to continue to receive the benefits of being in the game. This study will attempt to demonstrate the length that players will go through to “opt out” of activity-based control systems that do not provide a disproportionately larger reward to perceived risk ratio.

Hypothetical Suppositions
If an active, professional basketball player could, through self-effort, remove their image from all previous game recordings (Activity controls), so only their comprehensive individual and team outcomes (Stats) remained
-Would they choose this option?
-If yes, would they encourage others on their team to do the same?
-What if they were given the option of only removing themselves from select games?
It is conceivable that players would prefer outcome controls over activity controls, particularly during “Slumps” or if they received any negative feedback based on activities, rather than outcomes. This decision would be much easier if other players also opted out of activity controls to spread the blame or justify the decision.
If, when Garry Kasparov was actively playing chess, he could choose, through self-effort, to remove his previous individual moves from public record, leaving only win/lose outcomes
-Would he have chosen this option?

“CRM has to be easy for users to input their activities into or they won’t use it”
How easy does CRM have to be for users to input their activities?
Before one can answer this question, it is important to look at user resistance to automatic activity control systems in use, that require no effort by users to provide the information about their activities.

Opting out of a non-user-inputted, automated activity control

For users that have been assigned an automated activity control system, any opportunity or decision, to disengage or “mutiny against” the control system, will be weighed against perceived value or penalty by the user in doing so. One such anecdotal example came out of a conversation with a friend that has served with the Canadian military in Afghanistan. All Canadian military personnel are equipped with GPS transmitters that provide their information and location, to officers remotely guiding their actions. These GPS transmitters are used both in non-combat maneuvers here in Canada as well as in combat situations in Afghanistan. The purpose of GPS in training situations, is primarily to evaluate the activities and movements of troops as they perform their maneuvers to documented results (Low Cards).This information provides perfect knowledge of user activities and subsequent results, without any effort required by users to provide the information. Unlike peacetime training, in combat situations air cover is provided and GPS information is used to keep bombs from being dropped on troops and to generally keep them out of harms way (High Card), as well as activity controls. While activity control in peacetime serves to prepare troops for combat situations and thus protect them from harm, the perceived value by troops is far less and “Big Brother” is the common term used to describe GPS-based activity controls in training situations. My friend confided that a course of action sometimes taken by troops during training, is to turn off the GPS transmitters whenever possible. This collaborative, “Mutiny” approach against the “Low Card” activity controls, are often endorsed by junior officers that are held accountable for activity control information gathered about troops under their command. The same users, I was told, would never consider “opting out” in combat by disabling their GPS transmitters because the perceived value of higher safety outweighs activity controls.

Customer benefit is often weighed against the self-interests in activity controls
No where is this statement more appropriate than in the bitter battle between Yellow Cab drivers in New York City (NYC) and the Taxi and Limousine Commission (TLC) regarding the “Low card” activity controls of drivers using Global Positioning System (GPS) transmitters. Historically, NYC Taxis were not dispatched but were “flagged” by passengers and only accepted cash as payment. Manual “Trip sheets” were maintained by drivers listing pickup and drop-off points and the fare amount. This documentation formed the basis of declared income by drivers. In 2005 the TLC mandated that as of October, 2007, all NYC Taxis must have GPS transmitters installed with passenger terminals that also allow passengers view their location and pay by credit card. Passenger pickup and drop-off points with fare paid are recorded automatically recorded by the GPS units and transmitted real-time to the TLC. This system eliminates manual trip sheets and provides the TLC with an accurate and unbiased view of NYC Taxi driver activities and income. Many punitive controls have been put into place by the TLC to ensure compliance by drivers and continuous system usage is mandatory for employment
NYC taxi drivers have rallied heavily against these new “Low Card” activity controls, complaining about the cost of the units as well as the loss of privacy and autonomy. Interestingly, no statements by drivers admit the benefit to passengers of paying by credit card, nor address the (unsubstantiated) possibility that drivers rely on undeclared income made easier in a manual trip sheet system. TLC statements have focused on benefits to
passengers and for drivers, the end to manual trip sheets. To my knowledge, driver income accuracy has not been addressed in statements by the TLC except that the automated trip reports will only be available to the IRS by subpoena. The TLC has also addressed driver concerns regarding the issuance of tickets for traffic by stating “There are currently no plans to issue traffic citations based on GPS data.”

Hypothetical Suppositions
When the IRS is aware that automated, accurate records of taxi-driver income are available, will subpoenas for these records increase over those for the manual trip sheets?
If a taxi-driver is involved in a traffic accident and there will be a GPS record of how fast the cab was travelling, will the claimant demand the record?
If the TLC receives traffic complaints about drivers, will they be forced to look at GPS driver data?
Can activity control (Low card) data such as this be left unused once its existence is known?

Automating the decision to “Do the right thing”

If toll booths were completely run on the “Honor System” without automated enforcement, would drivers stop to pay the toll for the good of all or would they justify to themselves why they do not need to pay?

If the TLC allowed cab drivers to choose whether or not to install GPS and rely on “Market forces” instead, would drivers install the units to the benefit of passengers or justify to themselves why they do not need such activity controls?

When police departments can decide whether or not to have forward facing video cameras in their patrol cars, why does it usually only come about after racial-profiling lawsuit settlements?

A study on medical morbidity published in 2003 by Folkman, McPhee and Lo found that of training physicians who made serious errors causing death or injury to patients:
  • Only 54% shared the error with a colleague
  • Only 24% told the family of the patient

The bottom line is that people are highly resistant to providing any information on their activities that can be used against them.

My next post will explore how CRM can actually succeed if activity control elements are replaced by P.A.C.T.

Tuesday, January 22, 2008

2) "A Beautiful Mind" that started this journey

My research into the real reason behind CRM failures started after watching the movie "A Beautiful Mind".

In the movie, Russell Crowe plays the role of John Forbes Nash, the Nobel Laureate who suffered from schizophrenia. After watching a movie purporting to be based on a true story, it is my habit to do online research to find out what the "Real" true story is. While this often sucks the life out of the pure enjoyment of watching a movie, it feeds my inquisitive nature and my love of learning.
In reality, Nash was contacted by aliens, not the CIA, and his ever supportive wife divorced him but remarried him years later. Nash shared his Nobel Prize and no Laureate gets to make a speech at the ceremony. The key point here is I discovered the thesis Nash wrote at Princeton.

In his Princeton University doctoral thesis "Non-Cooperative Games", Nash expanded the concept of Game Theory by looking at a simple poker game. In all Game Theory, players will always do that which will give themselves the biggest payoff, with no consideration to the payoff to others.

I had no intention of entering into a research project at this point. I came up with an idea to create a game that would demonstrate to CRM users the benefits to themselves and to each other in using CRM to record their activities. This is the actual game paper I created:

A Card Game Analogy

When looking at any business report or situation (Poker Hand), there exist “High Cards” and “Low Cards”. High cards are those pieces of information a player feels will impact their hand the most .
These high cards often change and fluctuate depending on job role, culture and corporate direction. Because high cards can be subjective, incentive and corporate programs are often created to reflect (and drive) high card values and behaviors.
If you are left with high cards at the end of a hand, they count as negative.
Low cards are those pieces of information that in and of themselves do not command attention but in a cumulative way have an impact on high card values.
If a particular low card value is impacting, or needs to impact a high card value in a disproportionate way, it may be deemed a high card value for a period of time. Low cards are one to one issues
Often people leave the company and stacks of cards are found that should have been played, but were not.
If players never know if the other players are blindfolded or not, they cannot trust the game and will look for alternatives to ensure cards played are being acted upon.

High Card Dilemma: If a high card is played in CRM, assuming it will be read (and acted upon) and it is not, the player could be in a worse situation with the customer. If players assume the high card may or may not be read and acted upon, there is no point in playing in CRM because sporadic action will result in conflicting communication that may exasperate the issue even more (Blindfolded players). Players cannot win the game unless they play the high cards.
Low Card Dilemma: Over time, as all the hands I play are recorded and analyzed, others will know exactly what patterns I use in playing. I will lose my ability to bluff and others will be able to critique my playing. But if many players are recording all hands played, I will be able to view playing patterns of successful and unsuccessful players and allow experts to coach me to being a better player.
The coaches must also show all low cards played. Discarded low cards may be picked up and used by other players
THE DEALER IS ALWAYS THE CUSTOMER although wild cards are decided by players at the beginning of each round.
Discards are low cards deemed by others to be high cards
Endless deck with an equal number of high and low cards
Players may opt out of the game at the beginning.

What happens if players leave the game but don’t tell anyone?
High cards are recorded in your hand and then also sent to other players (Partners) that you think can also use them. Whoever receives the card either plays it or sends it to another player. Unplayed high cards at the end of the round, count against the holder and those that sent the cards.

You are all playing against the “HOUSE”

Game Theory Simplified
Game theory seeks to understand the rational choices that people will make to receive maximum personal payoff in a given situation. The most famous game theory is "The Prisoner's Dilemma".

The Prisoner’s Dilemma
In 1950, Rand Corp. scientists Merill Flood and Melvin Dresher, researching game theory in terms of its possible applicability to global nuclear strategy, came up with a series of non-zero-sum puzzles. From this evolved the Prisoner’s Dilemma:
Smith and Jones are arrested on suspicion of a crime. Their attorney tells them the evidence is flimsy, so if they both stay silent, their sentence will likely be a year at most on minor charges. The suspects are put in separate cells and each visited by the district attorney with the following deal:
If you cooperate and confess to the crime but your accomplice remains silent, you will go free because you cooperated, and we will jail your partner for 20 years.
If you confess and your partner does, then he will go free and you will get 2o years.
If you both confess, you’ll both get 10 years

A strategy of mutual silence results in the best collective outcome, but it requires the partners to trust each other because it places the silent player at risk of being exploited for the other’s gain. The dominant strategy therefore, is to confess (Defect).

John Nash took this theory one step further by mathematically proving that individual players could benefit by understanding that the other players will consistently do that which will give themselves the biggest payoff.

What this all has to do with CRM

CRM is sold to users under the premise that recording their activities in CRM will be beneficial to them.
Despite what is said in the room, users will inevitably make the decision whether or not to support CRM by recording their activities in a centralized database.
We know from many studies that CRM most often fails, because users choose not to use the system rather than support it. Although I strongly believe that consistent use of CRM will help employees be more successful, users will inevitably weigh any benefits in recording their activities against any percieved threats in doing so, when deciding whether or not to use CRM.

My study took a comprehensive look at what really goes on in the minds of CRM users, between the stated intentions to use CRM, and the inevitable failure of the program due to the lack of input by users, or distrust in the data provided.
My study does not support the conventional view of what CRM can or will achieve, in terms of allowing companies to have users submit to activity controls, regardless of how benignly these controls are presented to users.
Examples of common statements include:
  • “The company has neither the time or inclination to watch your activities”
  • “Information sharing is important with so many people calling on the same customers”
  • “The information will not be used against you but we do need reports in order to measure the success of the ____________ program”
  • “As long as you are getting results, we have no interest in monitoring how you are getting those results”

If the dominant strategy in a Trust/Don’t Trust Dilemma is “Don’t Trust”, by either the company or CRM users, the factors causing the distrust must be removed or CRM will fail.

Because this study and its conclusions may seem to cast doubt on the overall value in even attempting a CRM implementation, recommendations are made on how companies can increase the odds of a successful implementation substantially. The results of this study have been validated by sales reps and sales managers from several companies.

**The term “Activity Controls” does not refer to activities or information currently communicated between customer facing employees, via emails and/or phone calls, to achieve a specific outcome desired by the originating employee. Activity controls simply mean the ability to observe and record current activities and through analysis, the ability to change future activities and behaviors.
On my next entry, I will address the topic of perfect versus imperfect knowlege of activities and "High Card" versus "Low Card" activities.
Best Regards

Monday, January 21, 2008

1) The true killer of what should be a great tool for companies and their customers.

About Me
I have been a sales rep for various companies for over twenty years. Since I got my first 486 laptop, and took my first (And only) computer course called "How To Turn It On" (I am not joking), I have been using CRM. I have (of my own accord and expense), purchased and used almost every contact management program, owned six Palm Pilots, two Pocket PC's and a Blackberry. All of this has been in search of the best way to manage my customer relationships.
In this I have been very successful, and have always maintained an exceptional sales record. I am very much a "Go to guy" for computer and software related issues, despite my lack of formal training in IT.

Three years ago, I was asked to lead the Canadian CRM implementation for my company, as well as continue my role as sales rep. After completion of the roll out, I was asked to lead the North American CRM project full time.

I have conducted two full research studies on CRM failures. The first study was in university, the second was on my own, in an attempt to understand why the same employees that told me CRM was going to be a "Great Tool" for them to use, weren't using it at all.

My research paper "The CRM Dilemma" has been validated by many people in sales and sales management from companies where CRM has been implemented, and has failed. My presentation of this research to my company that had not yet fully implemented CRM, ended my career with them. I continue to be passionate about providing a tool that employees will actually use, rather than neglecting and conducting mutiny against something that will help them because of one element of CRM.

The CRM Dilemma

Let me caution that if you have not yet implemented CRM at your company, you should probably not continue reading this blog. If like me, you implemented CRM to great fanfare, using all best practices available, only to find no one is using it, you may be very interested in this research.

My wife speaks of the day I discovered the CRM Dilemma and how shocked and depressed I was. I really believed that I had chosen the wrong career path in moving from sales to CRM. Although my findings have been validated time and time again, they are hard to accept.

After presenting my research, the IT Project Manager said "If I was considering CRM and heard your research findings, I would probably recommend we don't even try implementing, and save the money." Usually, the problem is easy, but the solution is hard. In this case, the problem is much harder to accept than the solution.

To follow I will list the six most commonly held best practices for lessening the odds of failure in your CRM implementation. These are directly from my group university research project "Avoiding the Pitfalls - A framework for Successful CRM Implementation" (We got an A+)

  1. Have a clear CRM vision and communicate it often from the executive level
  2. Focus on your customer strategy
  3. Focus on collaboration with all stakeholders
  4. Evaluate all business processes (Don't automate a bad process)
  5. Use proper training and follow up with users
  6. Ensure data quality

Your list may vary slightly, but it has been said that your odds of successful CRM increase substantially if you achieve all these steps. Because these steps are so difficult to achieve, it is easy to believe your CRM failure was because you didn't have enough_____(Fill in blank from above list)

On this blog, I will begin to lay out my research that proves traditional CRM will fail, even if all these steps are followed, because of The CRM Dilemma. With the exception of compliance-regulated industries (Financial, Legal, etc) which I will address separately, the definition of the CRM that will almost always fail can be taken from the "What Is CRM?" line from any CRM company. An example from Sales

"The simplest, broadest definition can be found in the name: CRM is a comprehensive way to manage the relationship with your customers — including potential customers — for long-lasting and mutual benefit. More specifically, modern CRM systems enable you to capture information surrounding customer interactions and integrate it with every customer-related function and data point. The resulting information mosaic is then used to create and automate a variety of processes that identify, and describe, valuable customers. Most important, these processes help you personalize new and ongoing interactions to cost-effectively acquire, stay close to, and retain these "good" customers."

This sounds like something every company needs and should be striving for. It would quite frankly be career suicide to stand up at a meeting, and say that what is described above would be bad for your company. Nor can an employee stand up and say they do not want CRM and they will do everything they can to sabotage it, and make it go away.

On this blog, I will conduct a staged argument of my research that involves a very diverse group of subjects, from Chess Masters to New York City Taxi Drivers. I have no doubt that many will disagree with my research findings. I welcome the opportunity for someone to prove my research wrong.

If I am wrong, I can become a successful CRM Implementation Consultant. If I am right, I will be able to help extract value for companies that have CRM systems in place that employees aren't using, by changing what CRM is used for. This change by the way, has nothing to do with words or promises to CRM users about what CRM information is to be used for.

CRM needs to removed from our vernacular. There needs to be a fundamental shift in our understanding of how users view these programs, and most importantly, what lengths they will go to as a group to get rid of the threat of CRM.


There are true CRM success stories (To a point). CRM can work when the intent is to record strictly what the customer is saying. Typically, call centers and web forms work in this fashion. As mentioned before, in highly regulated industries, a minimum standard of record keeping is required for oversight compliance. In some companies, commission sales reps are told "If it isn't in CRM, it didn't happen" meaning that commission payment is contingent upon a minimum standard of record keeping. None of these examples really fits the definition of "True CRM" but they are held up by the CRM industry because for the most part that is all they have.

On my next entry, I will address Game Theory in relation to CRM.

Best Regards