custom IT research

now browsing by category


Pick the Right Customer

pick the right customer

Pick the right customer with your data?

In the March 2014 issue of the Harvard Business Review (HBR), Robert Simons,  the Charles M. Williams Professor of Business Administration at Harvard Business School, writes that organizations need to identify their primary customer type, and focus their business delivery there. With the right subscription you can read the article here. The author’s thesis is that organizations won’t dominate their markets by trying to be all things to all classes of customer. Winners know who they are targeting, and why.

However, if organizations rely on their own data to classify customers and determine, for example, their profitability, I question the value of their results. Our “Poor Data Quality – Negative Business Outcomes” survey shows customer data quality results that should make you weep. Please keep the following two charts in mind as you read on. Note that in both cases, more than 80% of respondents cited poor customer data as a cause of difficulty or a limit on success. Would you make smart decisions and pick the right customer with data in this condition? Click either graph to enlarge it.

[table id=4 /]

Four Steps to Pick the Right Customer

pick he right customerRobert Simons discusses four critical steps to pick the right customer for your organization. Let’s take a quick look at how poor customer data quality will shoot down Step 1 before you ever get started.

Step 1: Identify Your Primary Customer – Robert Simons identifies three decision elements to help you pick the right customer as your primary target. These elements are perspective, capabilities, and profit potential.

Perspective refers to your business vision, culture, and the folklore of your organization.

Your reputation in social media and survey results will give you insight into whether your various customers and prospects share your concept of perspective. Unfortunately, your poor customer data is not likely to help you get a clear picture of customer buying habits, returns, use of incentives, and other insights that might confirm your concept of perspective, or cause you to vary it.

Capabilities, according to Simon,  “refers to the embedded resources of the firm” and determines your organization’s critical differentiation for various classes of customer. You are probably positioned to serve certain types of customer better than others, but with poor quality customer data, you will have a hard time proving it. The same limitations you will have in validating your assumptions about perspective will challenge you in customer analysis.

Profit Potential segments your customers by their current and likely future profitability. Customers who can easily switch suppliers, buy only with incentives, have high rates of return, and need high levels of customer service are probably not your big profit generators. Does your poor quality customer data give you the profitability insights you need? Are you cross-selling and up-selling based on revenue instead of profitability? Customer data that is inaccurate, ambiguously defined, disorganized, or stale, is more likely to do your decision-making process harm.

Here are the other steps Robert Simons addresses.

  • Step 2: Understand What Your Primary Customer Values
  • Step 3: Allocate Resources to Win
  • Step 4: Make the Control Process Interactive

I will look at the impacts of poor customer data on decision-making in steps 2-4 with future posts. However, I think you get the idea. You won’t have confidence to make good decisions that are informed by bad data. Chances are, you won’t pick the right customer. Poor customer data quality will lead you to a negative business outcome when attempting to segment your customer base and pick the right customer for your organization.

 The Bottom Line

As Cal Braunstein of the Robert Frances Group and I speak with survey respondents we hear that many organizations are in denial about their poor data quality. One good way to send the wake-up call is to have us conduct a poor data quality – negative business outcomes survey especially for your firm. Your executives will receive eye-opening insights that will inspire the person in deepest denial to data quality improvement action.

One thing we hear again and again is that good things begin in earnest when high-level executives understand the data quality issue and get behind a data quality initiative. Why not get started down the road to data quality success today? Contact us, and let’s get your data quality improvement initiative underway.



Bad Data is a Social Disease of Business – Part 1

bad data is a social disease

Your organization’s bad data is a social disease easily passed to your business partners and stakeholders

With 200 completed responses in our “Poor Data Quality – Negative Business Outcomes” survey, run in conjunction with The Robert Frances Group, the IBM InfoGovernance Community, and Chaordix, it is safe to say that bad data is a social disease that can spread easily and quickly.

Merriam-Webster defines a social disease as

a disease (as tuberculosis) whose incidence is directly related to social and economic factors

OK, that definition works for the bad data social disease. In this case, the social and economic factors enabling and potentiating this disease include

  • Business management failing to fund and support data governance initiatives
  • IT management failing to sell the value of data quality to their business colleagues,
  • Business partners failing to challenge and push-back when bad data is exchanged
  • Financial analysts not downgrading firms that repeatedly refile 10-Ks due to bad data
  • Customers not abandoning firms that err due to bad data quality and management

No doubt you can think of other reasons why the bad data social disease spreads.

Was the source of this disease a doorknob? Not according to our survey respondents, who said the chief sources of  bad data social disease are inaccurate, ambiguously defined, and unreliable data, as shown in the graphic following.

As you can see, in the chart immediately previous, those are not the only causes of the disease.

bad data quality is a social disease

Social diseases negatively affect the sufferer, their partners, and the community around them.  According to our respondents:

bad data is a social disease

  • 95% of those suffering supply chain issues noted reduced or lost savings that might have been attained from better supply chain integration.
  • 72% reported customer data problems, and 71% of those respondents lost business because the didn’t know their customer
  • 71% of those suffering financial reporting problems said poor data quality cause them to reach and act upon erroneous conclusions based upon materially faulty revenues, expenses, and/or liabilities
  • 66% missed the chance to accelerate receivables collection
  • 49% reported operations problems from bad data, and 87% of those respondents suffered excess costs for business operations
  • 27% reported strategic planning problems, with 75% of those indicating challenges with financial records, profits and losses of units, taxes paid, capital, true customer profiles, overhead allocations, marginal costs, shareholders, etc.)

This response from our survey respondents highlights the truly dismal state of data quality across a spectrum of organizations.

What’s in Part 2?

In next week’s post, we’ll examine some of our survey results specific to bad data and the supply chain. A successful supply chain requires sound internal data integration and equally sound data exchange and integration across chain participants.A network of willing participants exchanging data is fertile ground for spreading the social disease of business.

Expect a thought experiment about wringing the bad data quality costs out of supply chain management, and see what some supply chain experts think about the dependency of effective supply chains on high quality data.

The Bottom Line

Believe that bad data is a social disease and take a stand on wiping it out. The simplest first step is to make your experiences known to us by visiting the IBM InfoGovernance site and taking our “Poor Data Quality – Negative Business Outcomes” survey.

When you get to the question about participating in an interview, answer “YES”and give us real examples of problems, solutions attempted, success attained, and failures sustained. Only by publicizing the magnitude and pervasiveness of this social disease will we collectively stand a chance of achieving cure and prevention.

As a follow-up next step, work with us to survey your organization in a private study that parallels our public InfoGovernance study. The public study forms an excellent baseline for us to compare the specific data quality issues within your organization. You will not attain and sustain data quality until your management understands the depth and breadth of the problem and its cost to your organization’s bottom line.

Bad Data is a needless and costly social disease of business. Let’s move forward swiftly and decisively to wipe it out! 

High Confidence in your Data Demands High Quality Data

poor business outcomes

High Confidence in Your Data – IBM Gets Serious

IBM is serious about you having high confidence in your data. Last week, on September 10th, IBM let us in on many aspects of its Big Data strategy. I thought the best part was their focus on Confidence in Big Data.

Big Confidence in Big Data

That was IBM’s message, clearly delivered by Brian Vile in his presentation last Friday at the InfoGov Community call. Readers of this blog know that along with The Robert Frances Group, Chaordix, and the InfoGov community, we are running a Poor Data Quality – Negative Business Outcomes survey. Thus far, we have learned of real and substantial costs to poor data quality, and that business executives see higher costs than do their IT executive counterparts. Not very confidence-inspiring. Garbage in means garbage out. That hasn’t changed since I wrote my first FORTRAN program back in 1968, for an IBM 1620. So what is my old friend IBM planning to do about eliminating poor data quality?

 It’s the Governance, stupid…

high confidence in your data quality IBM is right! Confidence in data is essential. Confidence comes from basing decisions on high quality data and getting great results. Governance is key to achieving and maintaining the high data quality for confidence in data, analysis, strategy, and tactics  based on that data analysis.

For example, without governance it would be difficult to discuss data sources, lineage, security, and archiving. These are key characteristics of data management and custodianship that give you confidence in your data analytics. Big data really does need quality, OK, sometimes use whatever data you have, but if you have big decisions to make, with big dollars at stake, you will want high confidence in your data.

Six Big Ones…

IBM presented six innovations that should build high confidence in your data. high confidence in your dataFrom my perspective, the most important is the the Information Governance Dashboard. This was depicted as a set of “Speedos” that give real-time insight into key performance indicators for high quality data. Will this be the device that appears in weekly data governance review calls? Naturally, it is too soon to tell, but I am hopeful. One of the IBM business partners focused on the Information Governance Dashboard is InfoTrellis. I’ll tell you more about their offerings in a future post.

 The Bottom Line

This whole Big Data Confidence announcement has convinced me that the sleeping giant has awoken to the sickening state of poor data quality. Perhaps now, something different will happen… the quality of your data will improve. Perhaps you will develop high confidence in your data, after a very long wait. Stay tuned, and I’ll keep you posted.

Causes of Poor Data Quality – Business and IT Perspectives

custom IT research

Did Business and IT Executives see the causes of poor data quality the same way?

Our survey investigating the causes of poor data quality leading to negative business outcomes has been in flight for four weeks now. You can read about our first survey insights here, and the survey launch here. I’m running this survey in conjunction with the IBM InfoGov Community, the Robert Frances Group, and Chaordix. At the time of this writing, we have 141 completed surveys and another 48 in progress. The analyses presented in this post are based on that survey response set. It should be interesting to see how business (non-IT respondents) and IT executives perceived the causes of poor data quality, both in aggregate, and by functional business area.

Who is an Executive?

We requested that survey respondents indicate their title, and used the following breakdown, as shown in the graphic following.

causes of poor data quality

Figure 1 – Respondent’s Titles

For the purposes of these analyses, respondents identified as “Executive Management C-Level or VP” and “Director” are Executives. The same title structure was applied to both IT and Business respondents.

On the business side, I’m scoring 19 respondents, comprising 39.6% of the total, as executives. On the IT side, I’m scoring 29 respondents, or 20.6% of the total respondents as executives.

Executives and Company Size

To ensure that I’m comparing apples to apples, I looked at the respondents reported company size. Insights into causes of poor data quality would probably differ between the CIO of a 50 person company, and one from a 5,000 person company. More business executive respondents come from larger companies than their IT counterparts, but there is reasonable symmetry for comparison.

Most individuals scored as Business Executives (~ 79%) came from organizations with more than 1000 employees, as shown following.

causes of poor data quality

Figure 2 – Business Executives by Company Size

As for IT Executives responding, 50% came from organizations with more than 1000 employees.

causes of poor data quality

Figure 3 – IT Executives by Company Size

Causes of Poor Data Quality across All Functional Areas

Near the end of the survey, and regardless of which functional areas respondents reported finding data quality problems, they were asked to identify the top three causes of poor data quality. Business and IT Executives agreed that the number one cause of poor data quality is Inaccurate Data. Business Executives scored Ambiguously Defined data as the next greatest cause of poor data quality, followed by Unreliable data.  IT Executives split the second greatest cause of poor data quality between Disorganized data and Unreliable data, with Ambiguously Defined data as a fourth cause. So, whether our respondents are on the business side of the house or the IT side, data that is

  • Inaccurate,
  • Ambiguously Defined,
  • Disorganized, and
  • Unreliable

Is costing their organizations time and money.

Graphical survey results of this information is shown in Figure 4, following.

 causes of poor data quality causes of poor data quality

Figure 4 – Business (left) and IT Executive Scoring on Top Three Causes of Poor Data Quality

The takeaway here is that Business Executives are aware of poor data quality, and believe it has much the same causes as their IT Executive colleagues. If both groups see the same problems cause poor data quality, why hasn’t business or IT stepped up to the plate and insisted these problems be fixed?

Customer-related Costs of Poor Data Quality

Our survey offers respondents the opportunity to monetize problems due to poor data quality. I think it is interesting to compare the perceptions of Business and IT Executive about costs or lost revenue opportunities when customer-related data is of poor quality, regardless of the cause. See Figure 5, following. Click the graphics to see full size images.


Customer Problem

Business Executive Response

IT Executive Response


Lost Customer  causes of poor data quality  causes of poor data quality Business sees worse costs than IT for losing a customer
High Cost to Acquire Customer  causes of poor data quality  causes of poor data quality Business sees higher costs to acquire and lower “don’t know” %
Lost Opp. for Additional Sales to Customer  causes of poor data quality  causes of poor data quality Business sees greater lost opportunity costs for add’l sales to customer than IT

Figure 5 – Comparison of Business and IT Executive Perceptions of Cost of Poor Customer Data

Across these three customer data cost or loss areas, Business Executives see consistently higher costs or missed profits than IT Executives.

The Bottom Line

I think we have uncovered an interesting pattern here, in which Business Executives see the cost of poor data quality as being higher than IT Executives see it. While there are too few data points to make a statistically relevant statement, that is how the trend appears to me. I would encourage many more InfoGov members to respond to our survey, to solidify these findings.

I’ve given you a glimpse of Executive level response in the Customer section of our survey. There are many more categories of respondent and many more functional areas for analysis. I urge the reader to stay tuned. If you have not joined the InfoGov community and taken the survey, please visit the site, and spend some time letting us know about your poor data quality experiences.





Customer-Related Data Quality Issues Cause Pain

Poor Customer Data Quality Hurts!

Customer-Related Data Quality Pain in Early Survey Results

We have an early yet strong indication of customer-related data quality issues across the full size range of organizations responding to our Poor Data Quality Survey. While there are many other areas of inquiry within our survey, I’ve selected customer data quality for this review because of its universality. Every business needs, and hopefully values their customers. However, despite all the talk about “knowing your customer” and the ascendancy and near-universality of customer relationship management (CRM), our respondents say that their customer-related data quality is sub-par and that has real dollar costs.

Strong and Rapid Response Suggests we Struck a Nerve

At the time of this writing, the survey, developed by the consortium of Principal Consulting, Robert Frances Group, Chaordix, and the IBM InfoGov Community has been in flight for two weeks, and has 88 completed responses with another 40 in the pipeline. Given the early August launch, with summer and vacation on people’s minds, the extent of this survey response suggests that Poor Data Quality leading to Negative Business Outcomes has struck a raw nerve.  You can read about the survey launch, and many prior posts on the survey, starting here.

Customer-Related Data Quality Issues Caused Lost Business

Our survey respondents were unanimous about the existence of poor customer-related data quality and told us about the problems they experiences, the likely causes, and the monetized cost of those problems. I’ve prepared several graphs that I’ll present following, with minimal commentary. The results largely speak for themselves. Click on the graphs to see a full size version.

poor customer data qualityAs shown in the analysis of survey question 9, poor customer data qualitymore than 70% of the respondents either customer-related data problems or knew of such problems in their organization. Question 10 responses show more than 68% of respondents believed their organization lost client business because of poor customer-related data quality.

poor customer data quality

Let’s look at question 10 responses broken down by business size. Bad news… nearly 89% of respondents working for businesses in the 5,000 -10,000 employee size are reporting losses due poor customer-related data quality. Only for the smallest business in the survey did fewer than 50% of respondents report customer data-related losses.




Poor Customer-Related Data Quality Made Customer Acquisition & Retention More Costly

As shown in the Question 13 graph, the 5,000 – 10,000 employee-sized businesses lead poor customer data qualitythe way again, with 100% of respondents indicating that poor customer-related data added to the cost of acquiring and/or retaining their customers. In no business size category did less than 50% of respondents affirm additional customer acquisition and retention costs.




Similar Causes Cited for Poor Customer-Related Data Quality and Poor Data Quality Overall

poor customer data qualityInaccurate, unreliable, ambiguously poor data quality reasonsdefined, and missing data lead the causes cited by respondents when asked about poor customer-related data quality (left graphic), and when asked about the top three causes of poor data quality across all survey problem reporting areas.



These are fumbles in basic data quality blocking and tackling.

Respondents Estimated the Cost of Poor Customer-Related Data Quality

poor customer data quality has real costsA key driver of our survey effort is putting a monetized value on the cost of poor data quality. Therefore, we asked our respondents to estimate the cost, loss in value, or similar concept that resulted from poor data quality, or the increase in profit or value that would result if poor data quality were avoided.

More than half our respondents offered a monetized estimate to our question about additional revenue that might have been realized if poor customer-related data quality had been eliminated. Over 15% of respondents believed that up to 5% more revenue might have been attained if poor customer data quality issues were eliminated. An additional 20% believed that between 6% and 10% more revenue might have been attained.

What could 6% more revenue mean in real dollars?

So, let’s say a 1 billion dollar company fixed its customer-related data quality issues and attained a 6% revenue increase. That would mean $60,000,000 in additional revenue. It might cost that company about $3,000,000 to obtain and implement data quality tooling and services to make that fix. That’s a 1900% Return on Investment.

Given numbers like these, is it really possible that IT organizations haven’t been able to sell their business colleagues on data quality investments? That sounds improbable, yet it appears to be true. Remember, this is just the customer-related data. There are 12 other functional business area on our survey, each with additional opportunities to increase revenue and/or decrease cost. 

The Bottom Line

We are at a very early point in the life of this poor data quality survey, which will run through the end of this calendar year, and then repeat annually. The interim results I’ve presented here are not surprising, but they will prove disappointing to data quality professionals. I hope they will motivate organizations to improve their data quality situation.

One takeaway at this stage is that InfoGov community professionals are eager to be heard about the data quality problems they experience and their associated costs. They want the world to know that data quality problems are still rife, and they are costing organizations and their stakeholders real dollars.

I will report on survey progress, aggregated responses, and additional insights as we enter Fall, and the relaxation of summer begins to fade.


Poor Data Quality Survey Now In Flight!

very hot content

Now, be part of the solution instead of the poor data quality problem!

Take our Poor Data Quality Survey at the InfoGov Community site. If you are a member of the InfoGov Community, logging in will trigger the survey. If you are not yet a member, visit the site, join, and take the survey. There is no cost to join the InfoGov community, and many benefits.

Our poor data quality survey is designed to make it easy for respondents to identify data related problems by business functional areas (e.g. Sales, Marketing, Finance, HR), and to select causes (e.g. missing data, stale data, naming issues, silo restrictions) of the poor business outcomes from poor data quality. Most importantly, our survey respondents will be able to monetize the effects of data quality problems.

What is in the Poor Data Quality survey?

This is a two-part survey. In part one, we cover the functional areas of business, asking respondents to identify and monetize problems in any of these areas for which they have experience and insight.

Did you experience or are you aware of these problems due to poor data quality?

1.Customer – Know and retain your customer
2.Marketing issue– Up-sell and cross-sell
3.Business Operations – Accounts Receivable, Payable, etc
4.Compliance – Pay fines, fees, go to jail, etc.
5.Financial – manual reconciliation problems, material faulty revenues, expenses, and/or liabilities
6.Human resources – internal and/or outsourced
7.Supply Chain – supplier quality or timeliness problems
8.Intellectual Property (IP) – impaired or inaccurate tracking, loss, etc.
9.Procurement, Logistics and Just In Time (JIT) – late/wrong inventory, stopped production lines
10.Plant and Equipment – missing spares, delayed maintenance, safety issues
11.Strategic planning – investment priorities, reorganization, M&A, divestitures, etc.
12.Ah Ha Moments
Survey respondents will be asked to categorize poor data quality problems, and to monetize the problems, as shown following.
poor data quality survey
poor data quality survey
In part 2we will interview respondents willing to speak with us. We expect to interview 10-15% of survey respondents. We will respect any interviewee requests for anonymity.

What about Demographics?

 Naturally, we want to analyze the survey results by respondent role, reporting relationship, industry, geographic area, and company size. Respondents answer with the click of a radio button as a clip from the survey shows, following.

poor data quality survey

What will we do with all the data?

Analyze it, and publish our findings. By the way, we plan to close the survey at the end of this calendar year, and then run it annually and publish additional reports.

Survey results in different flavors will be available for purchase after we publish. Individual purchasers will receive a .pdf or an online view of results. Corporate clients will receive comprehensive analyses of results, graphical presentations, and interviews. Technology vendors may opt to receive the detailed data tabulations in addition to our standard corporate deliverable package.

We will also offer custom client surveys. We’ll run this survey specifically for your organization and compile the results for your organization’s consumption only. The general survey results will provide a baseline from which your organization can measure its successes and challenges.

Please do contact us here, for specifics on pricing and deliverables.

Success has many fathers…

I have written in this blog, here, here, here, and here, about the poor data quality survey while it was under development, in conjunction with The Robert Frances Group, Chaordix, and the InfoGov Community. Well, launch has occurred this morning and our survey is now in flight.  Let me thank Cal Braunstein, CEO of The Robert Francis Group, Todd Courtnage, Director of Cloud Computing for Chaordix, Joe Royer of Principal Financial, a motivated InfoGov member, and Steven Adler of IBM, head of the Information Governance Community, for all the hard work that made our survey launch possible today.

The Bottom Line

The information we gather, analyze, and publish as a result of this survey will help organizations justify investing in improving their data quality. Better data quality means better business outcomes for all of us. Please do help us make a difference by taking the poor data quality survey today!