How to put IT Service Quality into your team's DNA

Editor's Notes

• #4: If there’s any doubt of the value of service quality, this should be enough evidence. Even if you’re not close to that point, it could happen to you.
• #5: Anyone heard about this?
• #6: To audience: Do you recall hearing about any of these? Have they affected you? Any other recent ones? Google alerts
• #7: If you’re going to impact it, you have to understand it
• #8: One way to think about it, is to take a scenario. The following scenario is adapted from a real world example, at a place I previously worked at.
• #9: Here’s Steve. At the board meeting he’s about to report to the other leaders in the org about how IT has met its goals
• #10: He’s proud to report that the goal was 3 nines, and his team hit 99.91%. Goodness. Right?
• #11: Here’s Meg, and she says, Steve, I’m confused about your number with all its nines. It sounds impressive, but here’s what my experience was this quarter…
• #12: This quarter we were the top sponsor for Australia’s got talent, and we spent a lot of money on that….
• #13: But when customers flocked to our web site during the advertisement, it wasn’t available!
• #14: Ah, whilst 99.91 ~= only 40 min downtime/mo, you’re telling me you really want to know about the business impact of downtime. OK, let me think about that one.
• #15: Just as Steve thinks the spotlight is off him, Sylvia speaks up.
• #16: Steve, even when the apps are up. They are very slow, especially during peak hour.
• #17: OK, response times matter to you. Got it. Next quarter I’ll report on that too.
• #18: And finally Mitch speaks up, and says
• #19: Steve, my teams are having to manually touch a lot of customer orders because the systems error out a lot. I mean, from my perspective, availability and performance are OK. But all the manual interventions are costing me money.
• #20: Ok, very well, next quarter I will come with metrics on that.
• #21: In the meantime, the business continues to grow. Everybody’s happy ya?
• #22: No! Even before steve can get back to the board, he makes the news headlines.
• #23: So he also understands that Scalability is important
• #24: In this scenario, we just covered four important measures of service quality. DRES is a quick way to remember the minimum four: Downtime, Response times Error rates, and Scalability. This is just one perspective on IT Service Quality. One that we detail further in the book.
• #25: Here’s another set of attributes I’ve used to define IT Service Quality, using a handy acronym: ASSURED (Availability, Scalability, Security, Usability, Response times, Error rates, Data quality) as an example of an extended framework
• #26: This is all good, but what does the customer think? Some research from Avant corporation has shown that 80% of IT Service providers think they’re doing a good job, but only 8% of their customers agree
• #27: The measures we’ve discussed so far are critical, but they have two limitations: Are an inside out view They are measuring technology quality, not the quality of business outcomes
• #28: You also need an outside in view. We’ll cover that in more detail a little later in the presentation.
• #29: But first, I want to elaborate on the danger of using an IT-centric measurement of performance. Lets back to Steve’s Availability metric. Many IT Shops focus mostly on that as a metric. The problem with availability expressed in terms of nines, is that it gives you Tunnel vision. If you are so focused on Availability (Downtime) that you lose sight of other critical IT Service Quality areas, such as Response Times, Error Rates and Scalability (and perhaps others), you will get burned. Your internal and external customers are counting on you to implement a holistic quality plan.
• #30: Availability is deceptive, perhaps on purpose. It can make a bad result sound good. 98% was a great test score in school, but it’s lousy availability. I often refer to availability as a metric IT invented to make system quality seem better than it actually is. Business people tend to respond favorably to that.
• #31: Here’s an important number that’s made the news recently but that’s also difficult to make sense of. The authors of this sign have tried.
• #32: Engage audience on what point this picture is making. In these two pictures, we see that a fire raged at the top level of two different structures. In both cases the fire lasted 5 hours. If the damage caused by these incidents was measured in terms of how long the incident lasted, would it be accurately representing the extent of the damage and impact? Yet in IT when we report availability we do just that. What people care about is not how long it lasted, but what the impact was. Was anyone harmed? What’s the cost of reconstruction? What will the inconvenience be to a family vs. a business?
• #33: Alternatives: +Measure money lost. But hard to actually implement, get people to agree on $ values etc +Use Business Impact Index
• #34: There is another way.
• #37: The graph on the left is a traditional Availability graph. The first problem with this graph is that it is artificially flattened by the Y axis going from 50 to 100. We’ve seen many occurrences of such graphs even showing a Y axis going from 0 to 100 while availability is always in the nineties. So this graph gives the impression availability is generally good. The graph on the right tells a different, more accurate story.
• #39: OK, so we saw how we can turn a not so good IT metric into a better one. But as mentioned earlier, you also need an outside in view of IT performance.
• #40: Tells you not just what the customers think of you but, more importantly, why! (as opposed to most surveys that only gather satisfaction data)
• #42: In all but one dimension, we are failing to meet the customer’s minimum level of service and that is Tangibles. Clearly a lot of work to be done!
• #46: Problem management is perhaps the most important process to get right to improve service quality. And that starts with having the right problem manager on your team.
• #47: It’s not enough to have a good problem mgt process. You need someone who will really think like Columbo. -- The most target rich environment you have for figuring out what might hurt you next is looking at what’s already hurt you. Your previous incidents. It pays to pore through last year’s incidents looking for trends, to understand what is really going on. You might be able to knock out 6 future incidents with a single project. Of course to do this you need to try to get to actual root cause as often as possible. If you don’t get to root cause, that’s like letting a killer go free. He’s gonna kill again… and maybe become a serial killer, unless you do something about it. Why is that? Consider this example…
• #48: Some of you may have seen this show that aired on ABC a couple months ago – Recipe for murder. Anybody here seen that piece? It’s a fascinating story about Sydney in the 1950s, when a bunch of people were dying of mysterious causes, and nobody knew why. http://www.abc.net.au/tv/programs/recipeformurder.htm
• #49: In those days, if you went to the doctor exhibiting strange symptoms, and they had no idea what your problem was they would simply conclude that you had what was “The Nerves”. And what was the remedy? Nerve Syrup! Did they really expect that to solve the problem??? Optional: It’s human nature to want to find a quick and simple solution.
• #50: Fortunately, in 1952 a couple of detectives in Sydney started applying methodical investigation techniques to try to understand why so many people were dropping like flies. They quickly started noticing some commonalities between cases, and suspected foul play.
• #51: In fact, they suspected that poisoning could be the cause, and started sampling the cakes and tea that some women were serving their "loved ones" they found that the samples tested positive for Thallium, look it up, it’s the perfect murder weapon (the active ingredient in rat poison. Colorless, tasteless, and with no smell)
• #52: And thanks to their great detective work, they put these lovely women behind bars. So what does that mean for Problem Management in IT? Other than a Problem can take the form of a nice old lady serving tea? Well, that it’s human nature to try to find quick fixes and magic remedies to solve problems… http://www.abc.net.au/tv/programs/recipeformurder.htm
• #53: … but the reality is, it pays to think like a detective to really solve complex problems. And that means, applying this methodical approach that has worked for years in high performance law enforcement organisations, and has worked for me when I was a problem manager.
• #55: So what about the DNA? After all, the core topic of our discussion today is getting IT Service Quality into the DNA of the entire organization. So how do you do that? So far we’ve discussed elements contributing to good IT Service Quality. We talked about 4 cornerstones including BII, good problem management. These all lead to better IT Service Quality. … But that’s still not enough. To answer that question, I’m going to ask you to think like an economist for a minute. There’s a well known economist, who in the 18 th century postulated - that people undertake economic activity because it’s in their own interest, not because they think it’s for the greater good of the public. Any guesses as to the name of the economist or the theory?
• #56: Adam Smith, who lived in the 18 th century, was a Scottish moral philosopher and a pioneer of political economics.  
• #57: And his metaphor was the invisible hand. Adam Smith used the metaphor of the invisible hand in The Wealth of Nations. The free market, while appearing chaotic and unrestrained, is actually guided to produce the right amount and variety of goods by a so-called "invisible hand".
• #58: Okay so back to our original question. What about the DNA? In IT there is an invisible hand too. I call it the invisible hand of IT.
• #59: One important component of building a culture of quality into your organization’s DNAs is what we call the Invisible Hand of IT. Your 10 or 100 or 1000 people are independently making many decisions that affect your quality every day – many of them that you have no exposure to whatsoever. So how do you have all these decisions contribute to high IT service quality? It’s not enough to just communicate about the importance of IT Service Quality. You need to make it such that your people understand why it’s in their best interest to do so. Adam smith Cross functional goals - DRES Publishing Diagrams
• #60: Just like profit motive is the invisible driver in the economy to get people to innovate, do the right thing, what’s the invisible hand of IT service Quality? What’s the motive? What unites people? What makes Jill not purposely do something that will negate what jack is doing to further IT’s cause? The invisible hand of IT is the combination of: Having good DRES goals Making sure those goals are shared by multiple parts of the organization (it’s not just a production or operations thing) Repeatedly broadcasting those goals. All hands meetings (no pun intended), posters etc Linking pay to performance. Drive accountability at the individual level. Use BII goals across the board for at least part of the bonuses. Evolving to a culture of performance. Good performers will like goals. Underachievers will try to avoid them. (Rejected by organization due to DNA incompatibility). Also, don’t only reward heroism
• #61: In 1961, President Kennedy challenged the nation to send a man to the moon and return him safely to the earth by the end of the decade. It was a bold and lofty aspiration, and it was also clearly and concisely communicated: only a sentence was needed. It was immediately comprehendible, measurable, and highly relevant—not to mention inspiring Both quantifiable and measurable, Kennedy’s throwing down the gauntlet made it unequivocally clear how we would know if we had succeeded. Not only was it understandable, quantifiable, and measurable, it was also relevant and meaningful to the country We believe that one of the best things we have going for us when managing large IT systems is that they are easily quantifiable and measurable. This allows us to leverage the power of managing to tangible, objective goals. A goal is an objective goal if there is a straightforward way to determine whether you are making progress or not and there is no talking your way out of it—you either succeeded or you didn’t.
• #62: As a result of this remarkable challenge, NASA created the Apollo program
• #63: The Apollo program used a series of clearly defined and measured incremental objectives (Apollo missions 1 to 10)….
• #64: … which led to that momentous day on July 20, 1969, when Neil Armstrong, as part of Apollo 11, took that first step on the moon, more than 226,000 miles away from the earth.
• #65: Returning to earth
• #66: Before it became a reality, this staggering achievement had to first be a goal. A clearly understood objective that thousands of people in the space program, the associated private sector contractors, Congress, and the entire American public could rally behind and use to focus their collective energies. If so many people did it, surely in IT we can too.
• #68: We discussed earlier that evolving to a culture of goals was important to the organizational DNA transformation. Here we’re going to discuss how transparency and information sharing can also contribute to taking the organization to the next level. Starting with Transparency. Once more we’ll use an analogy, this time in the financial industry.
• #69: Today we take for granted that as investors we can get access to data about mutual funds and other assets. But it wasn’t always that way. In the 80s the founder of Morningstar – his name was Joe Mansueto – had a vision of making fund data available to the average investor. What he sold to investors was transparency. Was it bad for the mutual fund industry? Not at all! In fact, it drove funds to seek better performance and the fund industry thrived. And Morningstar did well too, whether the data they were selling was good or bad.
• #70: Here are some examples of what Morningstar’s reports look like. We’ve all seen this. Over time Morningstar has developed some innovative ways of representing data and has gotten effective at capturing summary facts in a one page sheet, with details in following pages. We’ve see a similar approach used successfully in IT. In fact, we advocate having a Morningstar within your IT organization. This has the benefit of getting everybody in the org to understand what’s working what’s not, where to focus And it’s good for credibility with the business.
• #71: The second aspect of Information Promotion is making use of diagrams. Many IT organizations understand the concepts of Business Intelligence, and how to best visualize business data. But ironically not as many organizations do as good a at visualizing their own business. Many IT organizations struggle with the problem of tribal knowledge, and how to institutionalize knowledge. But yet we often times rely on what’s in people’s heads to make many decisions. We have found that it is worth the investment in time to diagram IT. Here we show a few examples, and there are may others in the book, such as this one, which we call this the Application Dependency Diagram. Much information can be conveyed in diagrams. Many dimensions are represented here.
• #72: Situational Problem Solving diagram. IT is generally good about documenting existing infrastructure. But diagrams can be very useful for documenting a particular and complex problem requiring communication among technical and business audiences. This is better than having to waive your arms lots over and over again.
• #73: Data Flow diagram
• #75: This presentation has been about how to get IT Service Quality in the DNA of your organisation. However, it’s important to not lose sight of the fact that we don’t want quality for the sake of quality. We need to be focusing on what matters to the business.
• #76: Everything we talked about so far, is very hard to accomplish if your organisation is technically focussed. As we disucssed, you need to have your teams The first step is to know where you sit on the continuum. (Take a 2 minute break to fill out the continuum survey.)
• #77: We’ve discussed some of the things you need to do, but how do you get there? Being focused on business outcomes is critical. At ProActive we view it as a transformation journey. Destination: - Integrated frameworks? - Business Value? And Business Focus - Right org? - …
• #78: You can check the box for a meeting re: the continuum