p>

Avoid Analysis Paralysis Piloting also helps with the second roadblock – designing and designing and designing until it’s perfect – analysis paralysis. You don’t have to get it perfect right out of the gate. If you’re like most organizations, any step in the right direction will make an impact. Plan to pilot your process with a small group, and then make revisions before rolling it out further. Not only does that approach deal with this common roadblock, it also helps to reinforce a continuous improvement culture.

Roll-out Finally, the last common roadblock in implementation is making sure the team members know what they are supposed to be doing. It’s important to have documentation on the process, but it’s also very important to provide initial training as well as ongoing coaching as the change is made. People learn in different ways, and often need the opportunity to ask clarifying questions. Formal training is key to a successful process roll out. Another technique we’ve found beneficial is having a process coach. As the team members start to use the new process, they’ll run into situations and may need help figuring out how to handle them. In truth, these situations may not have been considered when designing the process, and the process may need to be tweaked. There’s nothing wrong with that, but only someone with the big picture in view can tell that. Providing ongoing support will make the team members much more successful.

Why Bother? I acknowledge that this stuff is not always easy, especially for those of us that are much more comfortable with computers than we are with people, but it is important. We can’t continue to go on with the status quo. We have the tools we need to create an environment where we know what’s going on with our projects and we can consistently delive

Current State Anyone that has worked in the technology industry for any length of time has probably worked on a software project that was not completely successful. It may have been late, over budget, poor quality, not what the customers wanted, or some combination. We all also know that it can be done better. There is a wealth of literature on Software Processes.

The issue is not that companies do not know what to do in principle. The issue is that it is not easy to do in practice. It’s simple, but not easy. The difference between principle and practice may explain why so many organizations stay in their “dysfunctional” state, which might look something like this:

• Starting fires before you can put them out. One company I worked with had no testing or release process and as a result, created many problems for themselves by releasing software updates that inevitably broke other pieces of their very complex system. As you might imagine, it was almost impossible to make forward progress with all the fires they were creating for themselves.
• Burning the midnight oil. Another organization I know of has an ongoing project that is over a year late and millions over budget. The team has been working 60-80 hour weeks for about 14 months.
• Can’t keep up. A third organization I worked with was in an industry that required changes to keep up with compliance issues. Their process was so big and unwieldy that they could not even keep up with the mandatory changes, much less add any value added features for their clients.

So how do we get out of these situations? A software process improvement initiative can be a challenging undertaking. The path can be littered with roadblocks. In this paper, I will share some of the techniques we at Advanced Decision have used to navigate this path and plot a course to successful software development.

Buy-In The first and perhaps most significant roadblock is a lack of buy-in. The tricky part about buy-in is that to be successful in software process improvement, executive management, middle management, the development team, and in many cases, customers must all support the initiative, and they all buy in for different reasons. I’ve seen companies try to change their process by decree. I’ve never seen this work.

So what to do? The first overarching principal to gaining buy-in is over communicating. I’ve read that if you communicate something seven times, half of the people will have heard it. I don’t know if this is true, but I do know you need to communicate the changes, and the benefits multiple times in multiple ways.

Additionally, each group will have different reservations and will be motivated by different benefits, so tailoring the message is critical. When communicating with each stakeholder group, make sure to address the reservations, and highlight the benefits that are relevant to them.

Group

Reservations

Appealing Benefits

Developers

• Loss of control
• Lots of paperwork
• Loss of creativity

• Fewer fires to fight
• No more “death marches”
• Less rework
• Fewer bugs

Middle Management

• Loss of control
• Lots of paperwork
• Slow things down

• Better planning
• Fewer fires
• More satisfied customers

Executive Management

• Less flexibility
• Slow things down

• More visibility
• More satisfied customers

Customers

• Slow things down
• Less flexibility
• Less “direct access”

• Higher quality
• More visibility
• Getting more of what they want

Bite-sized pieces The second common roadblock is trying to change too much. I worked with an organization that had engaged one of the large consulting firms to help them with their process improvement. The result was a 100+ page PowerPoint with so many recommendations that the client organization was completely overwhelmed and did nothing. Granted, all of these recommendations were valid, and would have helped this organization, but would have required wholesale changes in virtually every aspect of their product development. The method that ultimately worked for this organization was one of incremental change. We spent some time interviewing key players in each group, and analyzing past projects and came up with the first couple of areas we wanted to address, based on the biggest bang for the buck. Note that we resisted the temptation to chart the whole process improvement path, and just took the first few steps, with the intent of assessing the situation again after evaluating the initial progress.

The benefit of an incremental approach to process improvement is that you can you can get improvements much more quickly, and you don’t bring the whole development group to a grinding halt. No organization can tolerate a major disruption. Culturally, incremental refinement also creates the opportunity for continuous improvement to become the norm. In one organization I worked with, the first couple of rounds of improvements were initiated and led by middle management, but after that a cross functional group of individual contributors (consisting of development, product marketing, QA, and customer support) took the reigns and continued on the path.

Broad Participation That brings me to another potential obstacle – lack of engagement by the team. By this, I mean a feeling that it is someone else’s process – no sense of ownership. The easiest way to deal with this obstacle is to avoid it entirely. I once hired a consultant to help me with process improvements, and although he and I had a pretty good idea of what we thought the end result would look like, we knew we had to get the team involved, even if it meant that the initial process design would go slower. Well, we did, and not only was the team on board, the final process included modifications suggested by team members that made it much better. This was a case of doing the right thing for not entirely the right reason, and it worked out great. One important point here, it’s important to signal the importance of this by involving “A players.” Everyone knows who they are, and if a process improvement initiative is staffed entirely with lesser players, it will not be viewed as important. Additionally, these “A players” are often leaders in their community and influence others, so their adoption and support is critical.

Pilot the process The next few roadblocks have to do with implementation. The first is “how do I roll this out to the whole organization,” and the answer is, you don’t. Piloting the process with one team or on one project is the way to go. You’ll get to iron out any kinks and show a win before you roll it out to the rest of the organization.

Avoid Analysis Paralysis Piloting also helps with the second roadblock – designing and designing and designing until it’s perfect – analysis paralysis. You don’t have to get it perfect right out of the gate. If you’re like most organizations, any step in the right direction will make an impact. Plan to pilot your process with a small group, and then make revisions before rolling it out further. Not only does that approach deal with this common roadblock, it also helps to reinforce a continuous improvement culture.

Roll-out Finally, the last common roadblock in implementation is making sure the team members know what they are supposed to be doing. It’s important to have documentation on the process, but it’s also very important to provide initial training as well as ongoing coaching as the change is made. People learn in different ways, and often need the opportunity to ask clarifying questions. Formal training is key to a successful process roll out. Another technique we’ve found beneficial is having a process coach. As the team members start to use the new process, they’ll run into situations and may need help figuring out how to handle them. In truth, these situations may not have been considered when designing the process, and the process may need to be tweaked. There’s nothing wrong with that, but only someone with the big picture in view can tell that. Providing ongoing support will make the team members much more successful.

Why Bother? I acknowledge that this stuff is not always easy, especially for those of us that are much more comfortable with computers than we are with people, but it is important. We can’t continue to go on with the status quo. We have the tools we need to create an environment where we know what’s going on with our projects and we can consistently deliver

littered with roadblocks. In this paper, I will share some of the techniques we at Advanced Decision have used to navigate this path and plot a course to successful software development.

Buy-In The first and perhaps most significant roadblock is a lack of buy-in. The tricky part about buy-in is that to be successful in software process improvement, executive management, middle management, the development team, and in many cases, customers must all support the initiative, and they all buy in for different reasons. I’ve seen companies try to change their process by decree. I’ve never seen this work.

So what to do? The first overarching principal to gaining buy-in is over communicating. I’ve read that if you communicate something seven times, half of the people will have heard it. I don’t know if this is true, but I do know you need to communicate the changes, and the benefits multiple times in multiple ways.

Additionally, each group will have different reservations and will be motivated by different benefits, so tailoring the message is critical. When communicating with each stakeholder group, make sure to address the reservations, and highlight the benefits that are relevant to them.

Group

Reservations

Appealing Benefits

Developers

• Loss of control
• Lots of paperwork
• Loss of creativity

• Fewer fires to fight
• No more “death marches”
• Less rework
• Fewer bugs

Middle Management

• Loss of control
• Lots of paperwork
• Slow things down

• Better planning
• Fewer fires
• More satisfied customers

Executive Management

• Less flexibility
• Slow things down

• More visibility
• More satisfied customers

Customers

• Slow things down
• Less flexibility
• Less “direct access”

• Higher quality
• More visibility
• Getting more of what they want

Bite-sized pieces The second common roadblock is trying to change too much. I worked with an organization that had engaged one of the large consulting firms to help them with their process improvement. The result was a 100+ page PowerPoint with so many recommendations that the client organization was completely overwhelmed and did nothing. Granted, all of these recommendations were valid, and would have helped this organization, but would have required wholesale changes in virtually every aspect of their product development. The method that ultimately worked for this organization was one of incremental change. We spent some time interviewing key players in each group, and analyzing past projects and came up with the first couple of areas we wanted to address, based on the biggest bang for the buck. Note that we resisted the temptation to chart the whole process improvement path, and just took the first few steps, with the intent of assessing the situation again after evaluating the initial progress.

The benefit of an incremental approach to process improvement is that you can you can get improvements much more quickly, and you don’t bring the whole development group to a grinding halt. No organization can tolerate a major disruption. Culturally, incremental refinement also creates the opportunity for continuous improvement to become the norm. In one organization I worked with, the first couple of rounds of improvements were initiated and led by middle management, but after that a cross functional group of individual contributors (consisting of development, product marketing, QA, and customer support) took the reigns and continued on the path.

Broad Participation That brings me to another potential obstacle – lack of engagement by the team. By this, I mean a feeling that it is someone else’s process – no sense of ownership. The easiest way to deal with this obstacle is to avoid it entirely. I once hired a consultant to help me with process improvements, and although he and I had a pretty good idea of what we thought the end result would look like, we knew we had to get the team involved, even if it meant that the initial process design would go slower. Well, we did, and not only was the team on board, the final process included modifications suggested by team members that made it much better. This was a case of doing the right thing for not entirely the right reason, and it worked out great. One important point here, it’s important to signal the importance of this by involving “A players.” Everyone knows who they are, and if a process improvement initiative is staffed entirely with lesser players, it will not be viewed as important. Additionally, these “A players” are often leaders in their community and influence others, so their adoption and support is critical.

Pilot the process The next few roadblocks have to do with implementation. The first is “how do I roll this out to the whole organization,” and the answer is, you don’t. Piloting the process with one team or on one project is the way to go. You’ll get to iron out any kinks and show a win before you roll it out to the rest of the organization.

Avoid Analysis Paralysis Piloting also helps with the second roadblock – designing and designing and designing until it’s perfect – analysis paralysis. You don’t have to get it perfect right out of the gate. If you’re like most organizations, any step in the right direction will make an impact. Plan to pilot your process with a small group, and then make revisions before rolling it out further. Not only does that approach deal with this common roadblock, it also helps to reinforce a continuous improvement culture.

Roll-out Finally, the last common roadblock in implementation is making sure the team members know what they are supposed to be doing. It’s important to have documentation on the process, but it’s also very important to provide initial training as well as ongoing coaching as the change is made. People learn in different ways, and often need the opportunity to ask clarifying questions. Formal training is key to a successful process roll out. Another technique we’ve found beneficial is having a process coach. As the team members start to use the new process, they’ll run into situations and may need help figuring out how to handle them. In truth, these situations may not have been considered when designing the process, and the process may need to be tweaked. There’s nothing wrong with that, but only someone with the big picture in view can tell that. Providing ongoing support will make the team members much more successful.

Why Bother? I acknowledge that this stuff is not always easy, especially for those of us that are much more comfortable with computers than we are with people, but it is important. We can’t continue to go on with the status quo. We have the tools we need to create an environment where we know what’s going on with our projects and we can consistently delive

ed customers Executive Management

• Less flexibility
• Slow things down

• More visibility
• More satisfied customers

Customers

• Slow things down
• Less flexibility
• Less “direct access”

• Higher quality
• More visibility
• Getting more of what they want

Bite-sized pieces The second common roadblock is trying to change too much. I worked with an organization that had engaged one of the large consulting firms to help them with their process improvement. The result was a 100+ page PowerPoint with so many recommendations that the client organization was completely overwhelmed and did nothing. Granted, all of these recommendations were valid, and would have helped this organization, but would have required wholesale changes in virtually every aspect of their product development. The method that ultimately worked for this organization was one of incremental change. We spent some time interviewing key players in each group, and analyzing past projects and came up with the first couple of areas we wanted to address, based on the biggest bang for the buck. Note that we resisted the temptation to chart the whole process improvement path, and just took the first few steps, with the intent of assessing the situation again after evaluating the initial progress.

The benefit of an incremental approach to process improvement is that you can you can get improvements much more quickly, and you don’t bring the whole development group to a grinding halt. No organization can tolerate a major disruption. Culturally, incremental refinement also creates the opportunity for continuous improvement to become the norm. In one organization I worked with, the first couple of rounds of improvements were initiated and led by middle management, but after that a cross functional group of individual contributors (consisting of development, product marketing, QA, and customer support) took the reigns and continued on the path.

Broad Participation That brings me to another potential obstacle – lack of engagement by the team. By this, I mean a feeling that it is someone else’s process – no sense of ownership. The easiest way to deal with this obstacle is to avoid it entirely. I once hired a consultant to help me with process improvements, and although he and I had a pretty good idea of what we thought the end result would look like, we knew we had to get the team involved, even if it meant that the initial process design would go slower. Well, we did, and not only was the team on board, the final process included modifications suggested by team members that made it much better. This was a case of doing the right thing for not entirely the right reason, and it worked out great. One important point here, it’s important to signal the importance of this by involving “A players.” Everyone knows who they are, and if a process improvement initiative is staffed entirely with lesser players, it will not be viewed as important. Additionally, these “A players” are often leaders in their community and influence others, so their adoption and support is critical.

Pilot the process The next few roadblocks have to do with implementation. The first is “how do I roll this out to the whole organization,” and the answer is, you don’t. Piloting the process with one team or on one project is the way to go. You’ll get to iron out any kinks and show a win before you roll it out to the rest of the organization.

Avoid Analysis Paralysis Piloting also helps with the second roadblock – designing and designing and designing until it’s perfect – analysis paralysis. You don’t have to get it perfect right out of the gate. If you’re like most organizations, any step in the right direction will make an impact. Plan to pilot your process with a small group, and then make revisions before rolling it out further. Not only does that approach deal with this common roadblock, it also helps to reinforce a continuous improvement culture.

Roll-out Finally, the last common roadblock in implementation is making sure the team members know what they are supposed to be doing. It’s important to have documentation on the process, but it’s also very important to provide initial training as well as ongoing coaching as the change is made. People learn in different ways, and often need the opportunity to ask clarifying questions. Formal training is key to a successful process roll out. Another technique we’ve found beneficial is having a process coach. As the team members start to use the new process, they’ll run into situations and may need help figuring out how to handle them. In truth, these situations may not have been considered when designing the process, and the process may need to be tweaked. There’s nothing wrong with that, but only someone with the big picture in view can tell that. Providing ongoing support will make the team members much more successful.

Why Bother? I acknowledge that this stuff is not always easy, especially for those of us that are much more comfortable with computers than we are with people, but it is important. We can’t continue to go on with the status quo. We have the tools we need to create an environment where we know what’s going on with our projects and we can consistently delive

ovement to become the norm. In one organization I worked with, the first couple of rounds of improvements were initiated and led by middle management, but after that a cross functional group of individual contributors (consisting of development, product marketing, QA, and customer support) took the reigns and continued on the path.

Broad Participation That brings me to another potential obstacle – lack of engagement by the team. By this, I mean a feeling that it is someone else’s process – no sense of ownership. The easiest way to deal with this obstacle is to avoid it entirely. I once hired a consultant to help me with process improvements, and although he and I had a pretty good idea of what we thought the end result would look like, we knew we had to get the team involved, even if it meant that the initial process design would go slower. Well, we did, and not only was the team on board, the final process included modifications suggested by team members that made it much better. This was a case of doing the right thing for not entirely the right reason, and it worked out great. One important point here, it’s important to signal the importance of this by involving “A players.” Everyone knows who they are, and if a process improvement initiative is staffed entirely with lesser players, it will not be viewed as important. Additionally, these “A players” are often leaders in their community and influence others, so their adoption and support is critical.

Pilot the process The next few roadblocks have to do with implementation. The first is “how do I roll this out to the whole organization,” and the answer is, you don’t. Piloting the process with one team or on one project is the way to go. You’ll get to iron out any kinks and show a win before you roll it out to the rest of the organization.

Avoid Analysis Paralysis Piloting also helps with the second roadblock – designing and designing and designing until it’s perfect – analysis paralysis. You don’t have to get it perfect right out of the gate. If you’re like most organizations, any step in the right direction will make an impact. Plan to pilot your process with a small group, and then make revisions before rolling it out further. Not only does that approach deal with this common roadblock, it also helps to reinforce a continuous improvement culture.

Roll-out Finally, the last common roadblock in implementation is making sure the team members know what they are supposed to be doing. It’s important to have documentation on the process, but it’s also very important to provide initial training as well as ongoing coaching as the change is made. People learn in different ways, and often need the opportunity to ask clarifying questions. Formal training is key to a successful process roll out. Another technique we’ve found beneficial is having a process coach. As the team members start to use the new process, they’ll run into situations and may need help figuring out how to handle them. In truth, these situations may not have been considered when designing the process, and the process may need to be tweaked. There’s nothing wrong with that, but only someone with the big picture in view can tell that. Providing ongoing support will make the team members much more successful.

Why Bother? I acknowledge that this stuff is not always easy, especially for those of us that are much more comfortable with computers than we are with people, but it is important. We can’t continue to go on with the status quo. We have the tools we need to create an environment where we know what’s going on with our projects and we can consistently delive

p>

Avoid Analysis Paralysis Piloting also helps with the second roadblock – designing and designing and designing until it’s perfect – analysis paralysis. You don’t have to get it perfect right out of the gate. If you’re like most organizations, any step in the right direction will make an impact. Plan to pilot your process with a small group, and then make revisions before rolling it out further. Not only does that approach deal with this common roadblock, it also helps to reinforce a continuous improvement culture.

Roll-out Finally, the last common roadblock in implementation is making sure the team members know what they are supposed to be doing. It’s important to have documentation on the process, but it’s also very important to provide initial training as well as ongoing coaching as the change is made. People learn in different ways, and often need the opportunity to ask clarifying questions. Formal training is key to a successful process roll out. Another technique we’ve found beneficial is having a process coach. As the team members start to use the new process, they’ll run into situations and may need help figuring out how to handle them. In truth, these situations may not have been considered when designing the process, and the process may need to be tweaked. There’s nothing wrong with that, but only someone with the big picture in view can tell that. Providing ongoing support will make the team members much more successful.

Why Bother? I acknowledge that this stuff is not always easy, especially for those of us that are much more comfortable with computers than we are with people, but it is important. We can’t continue to go on with the status quo. We have the tools we need to create an environment where we know what’s going on with our projects and we can consistently deliver on our commitments. By implementing some structure in the development process, we can allow our teams to use their creativity to solve the tough problems and drive innovation. Additionally we can create a sane environment where our people can do good work and still maintain balance between work and the rest of their lives.

Companies are under increasing pressure to deliver more with less. Whether the pressure comes from overseas or from upstarts right here at home does not matter. To survive in today’s fast-paced world, we need to be able to deliver quality products quickly and consistently. If you use these techniques to overcome these common roadblocks to process implementation, you are much more likely to create a sustainable, positive change in your organization.