Ten Mistakes to Avoid When Building a Data Warehouse Team
(Originally published by TDWI) By Maureen Clarry
(This article is available for download, as a PDF document.)
What does it mean to build a data warehousing team? Does it mean defining, structuring, or hiring? Does it mean increasing the size or impact of the team? Or, does it mean developing relationships among a group of people?
And who is responsible for building the data warehousing team? Managers are certainly responsible, but each person on the team can and should make a contribution to the team’s productivity and cohesion.
Regardless of your position or title, if you want to become a positive force in building a successful data warehouse team, this booklet is for YOU. Each of the 10 mistakes below is represented by a phrase or notion that is symptomatic of a current or future problem that your team may encounter. Listen for these “red flag comments” on your team and learn what YOU can do to prevent them.
MISTAKE # 1: Hiring yourself
“This person is great—they see things exactly as I do…”
It’s an easy trap to fall into—surrounding yourself with people who think just like you do. In reality, you need people who are different than you are. If you are a big picture person, hire someone who’s more detail oriented and leverage each other’s strengths. If you are action oriented, hire someone that will slow you down enough to analyze alternatives. If you don’t know yourself well, people who are your opposite may just drive you crazy. If you know yourself and are aware of your strengths, hiring someone who is your opposite can be a great asset.
For example, data warehousing organizations dominated exclusively by detailed, action-oriented types tend to deliver results quickly but may struggle with momentum, integration, and long-term implications. Conversely, data warehousing organizations dominated exclusively by big-picture, analytic types tend to consider meaningful alternatives but may struggle with taking too long, being too conceptual, and quickly delivering tangible results.
Data warehousing organizations that incorporate different personality types and perspectives quickly deliver meaningful results within a context that provides momentum and integration. Understanding different perspectives generally leads to better solutions than a single viewpoint shared by you and all the yous you hire.
MISTAKE # 2: Squelching disagreement
“We get along great—we never disagree…”
Sounds positive, right? Wrong! It could be a symptom of groupthink—a condition where people are so uncomfortable with tension that they do not assert their opinions or risk being different.
Surveys show that conflict is glossed over and avoided in more than half of the cases in which it appears. Another 30 percent of the time, conflict provokes heated clashes with no productive result. Only one out of five times is conflict surfaced, debated and authentically solved.
Are the benefits of speaking up worth the risk? Is doing the right thing for the business the right thing to do? The best data warehousing teams learn to manage conflict within “The Zone.” “The Zone” describes the appropriate level of interpersonal tension that exists in the team. If the tension is too high, conflicts become personal, comments become destructive, and the culture becomes toxic. If the tension is too low, people shut down, are afraid to risk offering their opinions, and the culture becomes complacent.
“The Zone” is managing the tension so that it’s not too high or too low—creating a culture where people contribute their opinions, debate the issues, listen and have constructive disagreements without obliterating everyone in their path. This is not always easy since we, as adults, are not taught conflict management skills but rather rely on learned patterns from growing up.
Operating in the “The Zone” requires us to become aware of our learned conflict style, determine if it is the most productive way to handle a given situation, and discover ways to create healthy tension in our team. It is possible to disagree without being disagreeable.
MISTAKE # 3: Confusing titles with roles and responsibilities
“It’s not my job…”
Titles, roles, and responsibilities. Three common words that are tossed around in status meetings, depicted on organization charts and included in project plans. We all have assumptions about what roles and responsibilities come along with our title. In fact, it is not uncommon for people from different companies or departments to have the same title but completely different roles.
In a recent survey by The Data Warehousing Institute, 77 percent of data warehousing professionals reported that they fulfill more than one role on their team—on average, they reported they fulfill 4.8 roles! No wonder it’s so confusing to understand who is responsible for what. And as hard as we might try to “get it straight” at the beginning of the data warehouse project, titles, roles, and responsibilities change over the life of the project.
Successful data warehousing teams recognize that titles are important for status or career paths and that roles are important as a general description of a person’s function. But when it comes down to getting work done, the best data warehouse teams focus on clarifying responsibilities. If each team member knows what he or she is responsible for, or what they are “on the hook for,” their roles will become a product of their responsibilities, the confusion will subside and the work will get done. And based on the changing nature of most data warehousing projects, this frequent clarification is not necessarily a conversation that has to be initiated or resolved by management.
Teach team members how to negotiate overlaps and gaps in responsibilities with each other so that discussions about roles and titles don’t become an obstacle to progress.
MISTAKE #4: Talking the walk
“There’s no trust…”
Distrust of management is a common theme related to turnover. But trust is not just reserved for management—it is each team member’s most important asset. Do you walk the talk or talk the walk? Can people believe what you say? Do you keep your promises? Do you share what you know?
Trust is an essential component for a data warehousing team. When trust is present, you will be able to build a team. When trust is missing, you won’t. People who have spent their careers in a single department (IT or business unit) are often taught loyalty to their immediate group and distrust the motives of others, even if they are in the same company. Lack of communication, past history, and lack of clear, common goals contribute to the problem.
So how do you tackle the problem of lack of trust on a data warehousing team? It starts with you! It’s not about saying the right words or platitudes. You earn trust by demonstrating honesty, integrity, reliability, and openness in your actions. No matter what your position on the team, a working relationship can crumble by not delivering on a commitment or not sharing information. You make the choice to be trustworthy, trust others, work together, grow and succeed, or take the adversarial path and fail.
No matter what your position on the team, be sure that your attitudes and actions are consistent with your words. Walk your talk.
MISTAKE #5: Thinking one size fits all
“If we could just get the right organization chart…”
The question: “How should we organize our data warehousing effort?” is one of the most frequently asked among data warehousing managers. Yet, successful data warehousing organizations come in a variety of shapes and sizes. There are literally one-person, covert, start-up efforts as well as large, fully-funded, permanent organizations.
In the TDWI Best Practices Award competition, 15 very successful organizations were depicted with variations of hierarchy charts, network diagrams, matrices, circles, squares, arrows and stick figures! Those searching for the Holy Grail of the perfect org chart might be disappointed to discover that one size, one org chart, does not fit all. However, what is consistent among successful data warehousing organizations is their ability to create adaptive structures that incorporate business participation and continually focus on delivering business value.
Adaptive structures support rapid change yet provide stability for individuals and the data warehousing effort. Adaptive organizations build competencies in key skills and then structure specific task-focused projects by deploying those skills as needed. Adaptive organizations have clear, but changing, roles and responsibilities. They plan for knowledge transfer and career development. Adaptive organizations rely on formal communication that follows organizational lines but cultivate informal communication laterally in support of their goals. Adaptive organizational structures accommodate long-term planning and incremental results.
Although you frequently hear the words “business sponsorship” or “business commitment,” perhaps a more important word is participation. Participation means including business people as part of the data warehousing organization. It might be a temporary, task-focused data warehousing structure and their involvement might be full-time or part-time, but they are part of the organization.
Beyond “sponsorship” or “commitment”—words that frequently imply money or intellectual support—participation means that business people step from their functional boxes on the org chart to become an integral part of the data warehousing team. Their names are on your org chart. Business sponsorship, commitment AND their participation provide the initial and ongoing focus for delivering business value.
MISTAKE #6: Pointing fingers
“If THEY would just…”
You hear it at all levels of the organization. “If THEY would just deliver.” “If THEY would just give better direction.” “If THEY would just step up to the challenge.” The biggest problem with pointing fingers at THEM is that we hand over responsibility for ourselves, the project, and the organization to THEM.
If problems are always out there and we have no part in creating them, we become poor helpless suffering victims of THEM. But if we don’t see ourselves as victims of these problems but as co-creators of them, we have to admit we are playing a part in the perpetuation of the problems. We see that these problems persist because of things we do or don’t do, and that if we changed what we did, we might be able to make the problems go away.
So the choice becomes: Do I point the finger at someone else or do I point the finger at myself? Do I count on someone else to get it fixed or do whatever it takes to get it fixed? Do I waste my energy on blaming others, or do I spend my energy figuring out a strategy to create a different outcome?
If you’re a customer wishing THEY would just deliver, what can you do to enable the data warehouse team to provide you with the quality system and service you want? If you’re a team member wishing THEY would give better direction, what can you do to get involved, understand their situation, and propose solutions that will move the data warehouse forward? If you’re a manager wishing THEY would step up to the challenge, what can you do to create an environment where you involve them, ask for help, share information, and make it possible for them to take more responsibility for the project?
Pointing fingers can create interesting workplace drama but it is a poor problem-solving strategy.
MISTAKE #7: Interviewing only for technical skills and experience
“Tell me about your data warehousing experience…”
Most hiring managers are adept at asking questions related to facts about a candidate’s technical background or opinions about themselves. In fact, most interviewers spend approximately 70 percent of the time on those types of questions: “What methods/techniques have you used to gather and analyze requirements from business users?” “How would you rank your skill levels on this database on a scale of 1–10?”
Even though these are valid questions to ask, only about 30 percent of the interview should consist of these type of questions. If you want to get to the core of the candidate’s relevant background, you should spend 30 percent of the interview on skills/experience and self-opinion questions and 70 percent on questions that measure the candidate’s fit to be successful in your environment. Define a success profile for your data warehouse team that identifies the INTRApersonal skills, INTERpersonal skills, stress management skills, adaptability skills, and attitudes that are relevant to the position within your organization.
For example, for a person to be successful in this position in your organization, they must:
- Possess integrity, take initiative, and be creative and self confident (examples of intrapersonal skills)
- Persuade, motivate, collaborate, network (examples of interpersonal skills)
- Tolerate ambiguity, embrace change, develop contingencies, prioritize (examples of adaptability skills)
- Cope with deadlines, manage conflict, mediate conflict, be resourceful (examples of stress management skills)
- Focus on customers, pay attention to detail, take responsibility, be optimistic, comply (examples of attitudes)
Since the best predictor of a candidate’s future success is to look at detailed examples of his/her behavior in past situations under similar circumstances, develop questions that relate to each item in your description.
The candidate’s response to the question: “Describe a time when you undertook a project above and beyond your normal responsibilities,” will tell you about their initiative (an intrapersonal skill). The candidate’s response to the question: “Describe a challenge you had in overcoming a co-worker’s or customer’s objections. How did it work out?” will tell you about their ability to persuade (an interpersonal skill).
The candidate’s response to the question: “Tell me about a time that you had to work with conflicting, delayed or unclear information; what did you do to make the most of the situation?” will tell you about their ability to tolerate ambiguity (an adaptability skill). The candidate’s response to the question: “Give me an example of a time when a co-worker really tested your patience. How did you handle the situation?” will tell you about their ability to manage conflict (a stress management skill).
The candidate’s response to the question: “Describe a situation in which your team succeeded because you exhibited a positive approach in guiding them,” will tell you about their ability to be optimistic (an attitude).
For every question, you’ll need to interpret whether the response demonstrates evidence of the experience, skills, attitudes, and behaviors you are looking for. Did the candidate identify the desired approach? How effective was the outcome? Was the response a “cookbook answer?” How similar were the past experiences to the situations the candidate will be faced with in your environment?
MISTAKE #8: Limiting leadership
“That’s management’s job…”
In successful data warehousing organizations, not only are we seeing new forms of organizational structures emerge to cope with faster-moving and more competitive environments, but also a new kind of employee. Specifically, we are seeing employees who are not content to passively follow.
For those raised on traditional notions of leadership, the idea of “too many chiefs” may come to mind. The reality is that only a handful of people rise to top management positions. But leadership is a behavior—not always a role. The most effective data warehousing organizations rely on a “community of leaders” and not just on the people with management titles and positions.
Formal and informal leaders assume responsibility for the success of the data warehousing initiative and do what it takes to make it successful. Leaders step outside of their “box,” they communicate, they contribute, they risk. Leaders believe that “if you’re not part of the solution, you’re part of the problem.” Leaders abhor complacency. Leaders “shake things up” in the positive sense of the phrase; they disturb the status quo and create a sense of urgency.
Leaders know their strengths and seek out others who mitigate their weaknesses. Leaders learn, ask questions, and are perpetually curious about how to make things better. Leaders invest more in developing their personal power than their political power. Leaders build relationships up, down, and across the organization to get things done. Leaders take initiative and are comfortable with the tension that their actions may create for them.
So, if you’re building a data warehouse team in a fast-changing, increasingly competitive business environment, which do you prefer: (a) limiting these leadership behaviors to the few formal roles on the org chart, or (b) creating an environment where these leadership behaviors are prevalent throughout the team? If management is defined as doing things right and leadership is doing the right things, a successful data warehousing team does the right things right.
MISTAKE #9: Becoming too task focused
“We don’t have time to invest in relationships…”
A good data warehousing team spends 70 percent of their energy on tasks and 30 percent on building relationships. In many teams, with deadlines looming, the tendency is to spend 90 percent on tasks and 10 percent on relationships.
The investment in the relational items such as communication, participation, appreciation, partnership, and morale are seen as a luxury until such issues as turnover, poor buy-in, or ineffective processes prove costly.
That doesn’t mean that you carve 30 percent of your project plan into off-site retreats, picnics, happy hours, and group activities. Those are social activities that may or may not build good working relationships. The task and relational issues can be incorporated simultaneously within the day-to-day project to make sure there’s a healthy balance.
For example, at a project kick-off meeting, are you just discussing the project charter, schedule, deadlines, and responsibilities? Or, are you accelerating the development of your team by including agenda items that help team members know and trust each other. Without getting into personal details that might be uncomfortable, it is possible to discover what makes your team members tick. What do they want to learn on the project? What are their hot buttons in working with people? How do they think? How do they process information? What do they consider their strengths and weaknesses?
Teams that focus exclusively on task issues will ultimately stall out and fail. It’s not a sustainable position for human beings over time. However, if you simultaneously focus on both tasks and relationships, trust will develop among team members. In fact, in our experience, loyalty to the team is a better predictor of retention than loyalty to the company. If you invest time in communication, participation, and appreciation, people will not only be committed to the success of the project, they will also be committed to the success of each other.
MISTAKE #10: All decisions are created equal
“How was THAT decision made…?”
You’ve been there—the big decision is announced and you’re left wondering why and how anyone could come to such a conclusion. Conversely, you make a decision and wonder why your team isn’t committed to the process, task or issue. Or, you try to include your team in decision making and you’re met with blank stares, crossed arms, and the occasional nod.
There is no single right way for making all decisions. But the process for each decision should be determined based on ownership, time, importance, expertise, experience, and cohesion. Who should be involved so that the decision will be supported? How much time can be committed to making the decision? How important is the issue? Who has the expertise required to contribute effectively to the decision? How experienced are the individuals in group decision making? Will the decision-making opportunity increase the cohesion of the team? And, most importantly, is it clear who will ultimately make the final decision? Are you asking for input so you can make the decision, or are you asking the team to make the decision?
In general, to make the best decisions, seek the maximum appropriate involvement. Be deliberate in determining what type of decision model is best for the circumstances. Sometimes that means you make the decision solo because it is of relatively low importance and time constrained. Sometimes that means you have time to ask for input based on expertise but you retain decision-making responsibility. Sometimes that means you take the time necessary to build consensus because the issue is important and you want to create ownership.
Regardless of the decision-making model, make sure you communicate what model you are using. Confusion about the decision-making process is the initial source for dissatisfaction with the decision. Participation in decision making takes time initially but it creates more ownership for the outcome and ultimately saves time. Slow down to speed up.
Maureen Clarry has been on the faculty of The Data Warehousing Institute since 1998 and teaches regularly on organizational and leadership issues related to business intelligence. Maureen is the CEO / President of CONNECT: The Knowledge Network, a consulting firm that specializes in IT people and organizations. CONNECT was recognized as the 2000 South Metro Denver Small Business of the Year, and has been listed in the Top 100 Women-Owned Businesses and the Top 250 Privately-Owned Businesses in Colorado. Maureen also participates on the Data Warehousing Advisory Board for The Daniels College of Business at the University of Denver and was recognized by the Denver Business Journal as one of Denver’s Top Women Business Leaders in 2004.