Agile? Waterfall? - Pick Carefully or Pay the Price!

Managing projects across an enterprise is an art as well as a science. There are several well known paradigms used by organizations across the world - waterfall to agile with variations in-between. I have always told those that have asked that there is not one silver bullet to magically solve the project management puzzle. In fact, my recommendations to people have always been to create a project management paradigm that fits the culture and needs of the organization.

My teams work in an industry that is heavily regulated by state and federal agencies and also must conform to several industry governing entities that set standards within the payments space. Many of the projects that we work on are implementing standard interfaces between the various payment networks, acquirer, issuers and other players within the financial industry space. Implementing these interfaces is a technical exercise with very little to no user interface. You either get the spec right or the money doesn't move. These initiatives really are not setup to be run utilizing agile techniques.

Now, let me shift over to some of the other initiatives we run within the organization. Our mobile and internet applications. As you can probably guess, these applications have significant user interfaces. These project efforts can utilize and benefit from agile techniques that can give our 'user representatives' much quicker access to potential solutions as well as drive increased cooperation between the development, user and quality assurance teams. This is a place where our teams are experimenting with agile techniques as a way to accelerate delivery through the development channel.

One organization utilizing different project management paradigms based on the unique needs of the initiative!

I'm running at this from two different directions at the same time:

1. I'm working with our project management organization to review our overall defined lifecycle to recommend changes that will allow projects that might benefit from being run utilizing agile techniques to take advantage of a slimmed down process.  We still need to figure out how to satisfy the documentation requirements of the various state/federal agencies and industry governing bodies, but overall, they are supportive of our desire to create multiple lanes that can be managed differently based on the overall risk of the effort.
2. I'm encouraging our project managers and team members to challenge the process. When working any individual initiative within the overall portfolio to make recommendations to skip various project artifacts, project steps or entire slices of the overall project lifecycle. The key is to document the decision as to why decisions are being made to skip certain pieces of the lifecycle based on risk and impacts to the overall effort.

This is what works in the enterprise that I currently work for. 

On a regular basis, I have to sit across the table from auditors and prove to them that all of the appropriate documentation has been generated. That there is traceability though the project effort from requirements, through design and testing. I have to show them our implementation and fallback plans as well as prove that those plans were followed during actual implementations. Our auditors will randomly pick projects and then go through all of the documentation and measure it against the documented lifecycle to ensure that it's all there. If something is missing, there has to be proof - either through project meeting notes or via change requests - that show the decision was made to skip the document/project step(s). This proof also needs to discuss the risk to the product, the enterprise, our financial institutions, acquirers and/or processors. It is the goal of our project management organization to manage these risk issues and ensure that we are safeguarding our ability to process payment transactions and move money within the payments space.

In past lives with other organizations I've run the gamut from running lifecycles that are even more formal than the one my teams currently use to running very informal lifecycles similar to the agile paradigm. The key is understanding the culture of the organization, the risk tolerance within the organization and the  time to market pressures within the industry.  As leaders within your organization, it's your job to discuss these issues and build processes that match the particular needs of your organization, manage the risk to the organization and your customers and build a project lifecycle paradigm that can deliver.

If you'd like more information on my background: LinkedIn Profile

Don’t Hide Behind the Process

Well, it’s been a while since I’ve discussed process issues.  Time to dig in and have some fun!

Most of you that have followed my previous posts know that I’m a process kind of guy!  I’ve introduced formal lifecycle management into several organizations.  The teams that I’ve led have successfully utilized process enhancements to drive improvements in productivity, overall capacity and quality within various organizations large and small.  Utilization of a known process allows you to begin to measure various pieces of the lifecycle – looking for inefficiencies and issues that impact your overall ability to deliver for the organization.

Now all that said, each organization needs to find the lifecycle process and development paradigm that will work with their culture.  I’m not going to try and tell you what development paradigm you need to follow because I don’t know your organization.  Within the organization I currently work for we largely follow a waterfall process with some agile techniques utilized within various teams.  This works for us and the style of applications that we create/support – mostly technical interfaces within the payments industry.  Our overall lifecycle has various checkpoints/gating that we use to validate we are ready to move to the next stage.  This works for us!  We’ve significantly increased our productivity and capacity over the last decade and have improved the overall quality of applications that we are delivering to our production environment.

While all this has been good, I occasionally notice when people are hiding behind the process versus just getting the work done!  Not every project needs every step of the lifecycle.  Some are small enough and straight forward enough that we can skip various parts of the process or make the decision not to create some of the standard artifacts/documentation.

Decisions need to be made up front when the project is defined.  Those in the know, need to direct the teams when a project needs the full pomp and circumstance of the lifecycle and when the teams have the option to skinny up the process and skip parts of the lifecycle.  That means at the very earliest stages of Discovery, the Project Sponsor should be giving direction to the team on what the risks are associated with the project which can then drive decisions on how much of the process can or should be used on the project.

Additionally, once a project has been initiated, I think individual team members need to be cognizant of what is happening and make recommendations to the Project Manager if they feel certain pieces of the lifecycle could be skipped or maybe some documentation might not need to be created.  I’m not saying that we’ll always make the choice to skip over pieces of the lifecycle or to not create various pieces of documentation, but as leaders, we need to listen to the recommendations of the people closest to the work and if it makes sense, say yes!  When we decide that we can skinny up the process used within an individual project or that we can eliminate some of the standard documentation, those decisions should be documented.  That way, there is a record of the decision and the reason is documented and shared with the Project Sponsor.

Overall, we still need to deliver for the organization!  Delivering the requested solutions with the necessary features/functionality; delivering a solution at or below the expected budget; delivering a solution with the highest levels of quality.  Our teams should be held accountable to ensure we are delivering successful solutions at the end of the project.

If you'd like more information on my background: LinkedIn Profile

Plan for Success ... Be Aware of Reality!

I run hard! My teams run hard!  We’ve got a lot on our plates and the pipeline is overflowing.  As soon as something is done, there are 5 more things waiting in the wings that need attention.  That’s a good thing – it tells me that the business values the services that my teams deliver on and that we can continue to be a difference maker within the organization.

As we go about our business – we plan for success.  Delivering product and services that our Financial Institutions can use to succeed in the market and make a difference with their customers.  Over the last several years, we have built a process around activity that ensures that we know why we are building something, we know the success factors around what it is we are building and that we then can design, build and QA the stuff we are moving into our production environments.

The teams have done a great job of reducing the number of iterations built before code can move into production.  Translation, they’ve reduce the number of critical errors found during testing that requires additional iterations prior to implementation!  They’ve also significantly reduced the number of issues that require after hours support.  Our teams have worked together to reduce the overall amount of time needed to regression test our products.  We are doing automated nightly builds on our code and are well on our way to having a full suite of regression tests run along with the automated builds.

So why am I talking about all of this?  Well, to put it all in perspective, it doesn’t seem to matter how well we refine the processes and manage the activity, we still have things go wrong!

1. Requirements identified after development is under way. 
2. Designs between teams that sometimes conflict. 
3. Estimates of project activity that end up being woefully incorrect. 
4. Resource constraints across teams just based on the sheer volume of activity. 
5. Data issues within our test environments. 
6. Validation exercises that end up taking longer than planned.

You know the drill.  You’ve all seen similar stuff happen to your projects.  So what’s a team to do?  Well as much as we want to keep our rose colored glasses in place and see sunshine and rainbows, it’s our job as leaders and technical experts to measure and apply a dose of reality within our projects.  The organization is looking to us to deliver new features/functions into the production environment.  They don’t really care to hear about how we make the sausage or why it might be taking longer.  What they do want is a semi-reasonable date as to when the features/functions will be available and what it means to our Financial Institutions and our Acquirers.

So, translating back to our internal teams, that means we need to make reasonable estimates along the way and refine as we move through the process.  As technicians, looking at and identifying the tasks needed to complete the activity – you can’t assume everything will go the way you want and your estimates need to include the breathing room you’ll need when something unexpected pops its head up.  As project managers, you can’t believe everything you’re being told – you need to validate what is being said and ensure people are on track.  Depending on the complexity of the project, you may need to put in some buffer time between the various activities to ensure that the teams have time to stay on track and recover from unknown issues.  This may not be needed on smaller less complex projects, but as the complexity gets bigger, you’d better start thinking seriously about buffer space.

I am not advocating that you needlessly pad time into your projects to make a 6 week effort end up taking 10 weeks.  What I am advocating for is that project resources accurately evaluate their tasks in a realistic fashion and feed those numbers back up to the project manager – that includes identifying the riskiest tasks and ensuring that you’re not being overly optimistic on the estimates.  Additionally, the project manager needs to review those estimates coming in and ensure that the numbers are ‘good’ and make judgements on where additional time may be needed to address parts of the process that look to be difficult.

I’ll give you a perfect example within my own organization.  With our large enterprise wide projects, we perform an integration test within development prior to moving the code into our formal QA and Parallel environments.  We began this formal ‘integration’ test effort probably a year and a half ago.  This testing required us to ensure mock data was setup across various platforms and application silos that would allow us to take transactions across the entire ecosystem.  This was initiated to ensure that by the time the code hit QA that we had successfully tested input/outputs and application handoffs between the various silos.

When we initiated this new step in our overall lifecycle – we were extremely aggressive in how long we thought it would take us to build the environment, populate the environment with the correct data and then execute the integration test plan.  I kept pushing the team to get all this activity done within a short time window so that we could then get into the ‘real’ testing and move the project forward.  Whether planned or not – the time associated with this integration testing took longer than planned and we eventually had to account for that within our planning.  Now that we’ve been through several iterations, we are beginning to automate various pieces of the environment and data setup/configuration – this should help us draw the time back down

Our goals were right, this helped us implement code into production that had a higher level of quality, however, I created a set of false expectations within the teams on the durations associated with this activity and they needed to give me feedback to reset my expectations and identify the correct timeline within the overall project activity. 

If you'd like more information on my background: LinkedIn Profile

Logic and Project Management!

As a child, I was glued to the television whenever there was an opportunity to watch Star Trek!  I loved the show - campy by the standards of today, I couldn't help myself as I wanted to be on the Starship Enterprise as she traveled to the stars.  And, yes, my favorite character was Spock - followed closely by Scotty.  Spock in may ways spoke to many of us - using logic instead of emotion to solve problems.  And when he did let his emotional side out of the box - showed that emotion could and would frequently lead to painful mistakes.

So, why am I using space within a blog focused on project management and applications development to discuss Spock?  Logic - use it to succeed within your role.  Strip aside the emotion that you feel when confronted with issues and follow the facts!  Stay calm, track down the details and use those details to identify the move forward solution.

No project of any realistic size moves from idea to implementation without some sort of issue rising up, ready to take the whole thing off the rails and into the ditch.  Some problems are small and others tend decloak themselves like a Klingon battle cruiser to surprise and fire upon your Starship Enterprise!  These incoming salvos threaten to overwhelm you, the team and the project.  At this point, you can either act emotionally, running around without a plan and make the problem worse.  Or, you can act like Spock - use logic to create a plan, execute the plan and overcome the issue.

Planning:

Sometimes with a small issue, you can 'wing it' and successfully manage the issue.  That said, the larger the issue, the more likely you will need some type of plan and process to deal with the issue.  Leadership in the organization is going to expect you to communicate the issue, progress in finding a solution and then progress against whatever plans are ultimately put in motion to resolve the issue.

There will be wide variety of additional items you will need to manage within the overall effort:

1. You may need to setup a research team to investigate how the issue was missed or what the cause of the issue was so that this information can be communicated to the management team.
2. Additional personnel resources beyond the original scope of effort associated with the project to identify the overall solution as well as assisting in any work needed to implement the solution.
3. Additional budget resources beyond the original scope of the effort to cover any additional internal or external resources needed to provide the overall solution.
4. Communication plans may need to be put in place to update internal resources, communicate with and manage customer expectations and manage any risk to the overall reputation of the organization.

Each of the above items will require management effort from the Project Manager to ensure that future milestones within the overall effort are adjusted and manged properly; and to coordinate activity between various team members, departments, and outside consultants.  All of this will take careful planning, using organizational skills to minimize the impact to the project.

Methodical:

Many times as a Project Manager you are dealing with choices - making decisions about where short-cuts can be taken; making decisions on whether to go with Plan A or Plan B when there is no consensus from the team on which direction to take; sometimes sifting thru agonizing amounts of detail to understand an issue so that it con be communicated to the management team.

As the complexity of a problem increases, the number of details being tracked or managed associated with the issue will grow beyond your minds ability to track them all.  You are going to need processes in place that allow you to align the issue resolution within the overall tracking mechanisms in place to manage the project.

1. You will need to track new data points within and aligned to the overall project - new milestone dates, budget information, new internal and/or external resources, new equipment that may need to be purchased.
2. You will need to realign your current schedule to accommodate any new milestones associated with the new effort.  Especially those new milestones that will require current or future activity within your plan to be delayed or altered.
3. You will need to ensure that communication plans are executed properly not only to manage internal expectations, but just as important to ensure that customers, vendors, the media and other 3rd parties are kept in the loop.

Your role as Project Manager requires that you have a plan, that you execute the plan and that you manage the expectations of your team members.  You need to be the calm within the storm.

Detail Oriented:

Team members are going to be hitting you with lots of information - your job as Project Manager is to sort through the minutiae and ensure that the relevant data is recorded, tracked and used to make decisions throughout the overall project lifecycle.  As Project Manager, you will be expected to identify the important pieces of data, communicate that to the team and ensure that the information is available within meetings as decisions are made on the forward progress of the project effort.  

Additionally, management will expect that you understand the relevant pieces of information within the flood of details and know when to present that information up the ladder.  Many a Project Manager or Manager have failed when they have forgotten to keep the leadership team informed on decision points and the underlying information used to make those decisions.  I'm not saying that you should have the expectation that your being micro managed - but when there is an unforeseen issue that crops up, over communication is necessary to ensure that everyone understands the nature of the issue, the solution being implemented and the overall impacts to the organization.

Logical:

You can be angry and frustrated - that however is not going to help you move the ball forward and find a solution.  At this point, you need to step away from your emotions and use logic to understand where your at, what data points you have available to make a decision and then make short and long term decisions based on that information.  Knowing that you can adjust future activity and decisions based upon additional data points that surface.

Spock would 'follow the facts' to get to the answer - wherever those facts happened to lead him.  That's your role!  You will be tempted as angry and frustrated team members, management, customers or 3rd parties call you or come over to talk with you.  At times, you will want to 'fire back' at the individuals - however, it is better to step back, take a breath and then approach the conversation.  Focus on the fact, just the facts!  Use logic - it is the best tool you have at your disposal to deescalate the situation.

Mr. Nimoy, you are and were an inspiration to many of us ... may you live long and prosper among the stars!

If you'd like more information on my background: LinkedIn Profile

Off-Shore Development - Plan it right or Fail!

Off-shore development continues to be a discussion point within many organizations.  I've managed teams across the US and had the occasion where I've used off-shore resources.  Let me share some of the experiences in the hopes that it helps others that struggle with the model or are questioning whether to proceed with the use of off-shore resources.  I won't claim this represents the model to run off-shore development, ultimately, you need to make the decisions that will work within your organization.  Hopefully, this will shine a different light on the topic and allow you to gain insight that may help you make the decision to use off-shore development resources or to keep the development on-shore.

First, it is not easy.  If you think this will solve issues within your current development lifecycle - don't bet on it.  If you're going to seriously begin to use off-shore resources, you better plan appropriately, or you'll pay the price with development timelines that get extended and budgets that creep beyond your expectations.  In the worst case scenario, you'll end up rolling product into production that has not been properly tested and vetted.

Some of the breakdowns that my teams encountered in using off-shore resources are as follows:

1. Communication - eliminating the natural lag in communications as teams work shifts that do not align.
2. Design Aberrations - ensuring the product developed fully conforms to the original design/requirements.
3. Coding Standards - putting processes in place to ensure that the off-shore development team followed the development coding standards of our internal teams.
4. Quality of Code - ensuring that the code had been properly tested all the way through integration testing with other application silos/subsystems.
5. Quality of the Team - ensuring the consistency of development resources applied to projects.
6. Product Knowledge - ensuring that knowledge of team was retained from project to project.

The above list covers the major challenges that we encountered.  Let me go in to some level of detail so that you can get the flavor of what was experienced and how we handled the longer term needs of the organizations use of off-shore resources.

Communication: Each and every day, without your knowledge dozens and dozens of project decisions are made as developers lean across the walls and talk to each other or see each other walking around the office.  My teams have been no different, requirements and designs could change based on a hurdle that needed to be solved and a quick conversation in the hall.  Suddenly you have a team separated by multiple time zones - in my case, they had left the office for the day by the time we arrived in the office.  If the remote team had a question, or hit a roadblock and needed direction, there was an automatic lag of at least 1 work day.  If it was a serious issue, the team might loose several days as communication threads worked their way through email.  In extreme instances, we might have some of the remote team stay late or some of the local team come in early to establish conference calls and work out the issues.

Design Aberrations: The standard methodology we used to develop project requirements within the organization wasn't sufficient to support the use of outside resources.  Didn't matter if those resources were on-shore or off-shore, our teams were comfortable dealing with each other and could abbreviate how the requirements were documented.  There were a lot of things assumed within the requirements that didn't need further explanation because the teams had years of experience working together.  This continued to feed through the process as design documents were generated by the team.  Suddenly we introduced a team of developers that did not have that tribal knowledge.  The results were obvious in hindsight, the off-shore resources built the product to match the documented requirements and designs.  The only problem - those requirements and design documents didn't contain the unspoken elements known by the internal teams.

Coding Standards: Some of the products that we chose to develop off-shore ended up needing to be supported by local teams.  We quickly found out that the coding standards we had in place were not documented thoroughly and left many items up for consideration.  Translation, the remote team was producing code that looked like none of the other code that was used internally.

Quality of Code: The internal teams had many years of experience working together, touching each others code and making the necessary revisions to the code base.  Suddenly we had bugs that had been solved being reintroduced into the code base later in the development cycle.  Worse yet, we would find chunks of code that were never being executed, yet were loading up the code base.

Quality of the Team: Just as we were getting used to working with one set of developers, suddenly they would be reassigned to a different project and we would end up having to reacquaint ourselves with a new team.  To say that this slowed down development and introduced inconsistencies into the code base would be an understatement.

Product Knowledge: As we completed development with one project and started up the next project, we were not assured that the team we had gotten used to would continue on and do development on the next project.

None of this is serious and can be resolved, but you need to think through the engagement up front and take the necessary measures to protect your organization.  Here are some of the tactics we began to utilize to minimize the risk in the use of off-shore development resources:

Communication: We required the off-shore company to place one of their resources on-site with our local teams.  They became the primary communication point between our teams and the off-shore resources.  They sat in on the meetings locally and could catch the small things that needed to be communicated to the off-shore team.  They could do the impromptu meetings in the hallway and get that information promptly to the remote teams.  Our project timelines included buffers to accommodate the communication lag between the teams.

Design Aberrations: We ended up beefing up our Business Analyst and Architecture teams to create more thorough documentation.  Our documentation became much more thorough and attempted to accommodate the things that people took for granted.  The architectural diagrams became much  more sophisticated outlining all servers touched, all databases touched and were color coded to identify new, changed or removed elements.  Design documentation began to reference not only the objects that would be touched, but could go down to the individual methods and parameters when needed.

Coding Standards: We ended up going through our coding standards and documenting much more thoroughly how code was to be structured, how variables names were to be created, what security precautions were to be taken when creating the code, etc. 

Quality of the Code: Code created by off-shore resources was required to go through the same code review process that we used for internal teams.  This ensured that our senior developers were reviewing the code - exactly as would be done for internally written code and that they could catch issues and feed them back to the remote team.

Quality of the Team: After some experience with team members shifting, we ended up negotiating with the off-shore company to ensure that we had a core set of developers that would not be reassigned off of the project.  This allowed us to maintain some level of consistency within the off-shore team and addressed quality of code, coding standards and communication issues.

Product Knowledge: Similar to the Quality of the Team issue - by keeping a core team of developers assigned to our projects, the off-shore teams knowledge base of our products began to increase and they could begin to anticipate some of the issues ahead of time.

I want to state again, these were experiences that my team had in working with off-shore resources.  Other organizations may have different experiences.  Hopefully, some of these issues and the ways in which we resolved them will assist you with determining whether you can support the use of off-shore development resources.

I typically do not use off-shore resources unless there is a need to temporarily augment our internal resources to accomplish specific development/project goals.  That's a choice that I feel comfortable with based on the particular business needs within the organization.  In previous roles, I've had the opportunity to work extensively with off-shore resources.  If you're going to use off-shore resources, I highly recommend you pick a couple of small projects and trial a couple of different vendors to establish which one will work better with your current development paradigm.  Don't be afraid to switch  vendors if you don't feel you're being taken care of - ultimately, it's you that will be held accountable for the delivered product.

If you'd like more information on my background: LinkedIn Profile

Test Driven Development - Think about it!

Over the last few years I've been moving my teams towards Test Driven Development.  If you're not there, you need to be looking at this paradigm.  In essence, the following is what you are shooting for:

1. Write your test.
2. Run your test - the test should fail because the code hasn't been written yet.
3. Create the bare minimum of code that will allow the test to pass.
4. Continue to refactor the code until you are satisfied - is it simple, have you removed any duplication?
5. Repeat - accumulating additional tests until full functionality matches the agreed upon requirements and design criteria.

I've been on this push for a couple of different reasons: 1) The QA Team is spending an inordinate amount of time performing regression tests that should be in the build process; 2) The QA Teams and Development Teams should spend most of their time on new feature/new function testing - not regression testing; 3)  Within the old paradigm that we were following, by the time the development team member became aware of a defect in the code, they had long since moved on to other parts of the code and it was not fresh on the mind.

I've recently spoken with several groups of students - some high school and some at the collegiate level - and the one common theme that I have been impressing on all of these groups is that they need to think about testing first.  That means while doing the design work and prior to any code being written, they should understand how they will test the code and then execute against the plan.

In a separate session. when meeting with some of the instructors - another individual recommended that the instructors should have a base set of tests created that the students will need to execute against their assignments.  The students could do this at any time to get immediate feedback.  If an instructor were to pursue this paradigm within their classrooms, I would encourage them to create a portion of the tests, but force the students to create their own set of tests.

Unless our teams are working on a completely new system - there should exist somewhere in the organization a base set of tests that cover the current functionality.  It may not be extensive, but it's there somewhere.  That's the baseline!  No build of the system should happen without improvements to that baseline with your development team increasing the code coverage with additional unit tests that are then included in the set of regression tests the next time that code is touched.

Let's face it, immediately after the code has been written and the developers focus shifts on to the next task, their ability to maintain the code begins to drift.  I'm not saying they can't maintain it, but that they will need to spend time re familiarizing themselves with the code and then planning how to make the change.  Additionally, you can't guarantee that the next time the code needs to be touched it will be touched by the same developer.  By having these tests complete and in the pool of regression tests, you set an expectation with the developer that they can't claim they are complete with the code until all regression tests are complete and they can prove that they have tested against the new feature/functionality of the application.

Really, when it comes down to it, it's about accountability within the development team.  They are just as responsible for the quality of the code as are the people working the front end of the project - discovery and requirements, design - and those working the back end of the project quality assurance and the user representatives.

If you'd like more information on my background: LinkedIn Profile

Estimates - The Buck Stops with You!

Why are people so uncomfortable making a decision?  I admit I scratch my head frequently at some of the issues that land on my plate because someone was uncomfortable making a decision – or because they feel they haven’t accumulated enough information.

This is not a phenomenon unique to my current organization.  I’ve seen it across most organizations that I’ve worked in.  Managers have engrained into people that they are not allowed to make decisions and that they must review everything prior to anything getting done.  YIKES!  And then we wonder why people fail!  Better yet, we wonder why people leave us and move on to a new job – leaving us to spread the workload around to others on the team.

Look, I won’t claim my teams are exempt from this pattern.  But I do try to encourage people to think through problems they are facing and if they are unsure of a direction, at least bring options into the conversation when we sit down to have a discussion.  This at least allows me to see that they have thought through the problem and are attempting to address the issue.  Usually through conversation they’ll come to some type of a conclusion, thank me for the discussion and head back out.

I think where we struggle the most within my current teams is getting people to make estimates on project activity.  This is a personal decision by team members measuring how much work they will be responsible for against requested project activity.  When we first introduced this concept, there was a lot of push-back.  And to be frank, there were a lot of errors in the initial estimates.  But over time, people have grown more confident and are estimates are aligning better with the actuals that we measure over the life of a given project.

We go through regular training with people that are new on our teams.  It really doesn’t matter if they’ve been in development or whether they are new to the ropes.  Most developers have never been trained on how to produce reliable estimates.

So how do you pull together an estimate – well, for one thing, it’s different depending on what stage of the project you’re in: 

1. Discovery Phase: Usually estimates at this stage are done by someone within the management chain along with senior level technology resources.  How?  Well, the truth of the matter is the project scope is looked at in comparison to other projects that have been completed that smell like or look like they have similar impacts.
2. Requirements Phase: Typically this is a minor refinement of the estimate that was given in the Discovery Phase.  Any changes to scope are accounted for and typically there are additional details highlighted that point to previously unconsidered impacts.
3. Technical Design Phase: Now we are putting some meat on the bone.  At this point, technical resources have engaged and identified impacts across current systems/application silos and have also identified where new functionality must be created.  Along with the technical estimates business teams have assessed impacts to current and new processes.  Between the two paths, the full project estimate is considerably more accurate than anything that was put together in the first 2 phases of the lifecycle.
4. Development Phase: Some organizations perform the final detail design in this phase.  If so, this is the last chance to update those estimates both from a technical and business perspective.  This should allow the resources that are going to do the actual work to review the technical designs, identify the specific chunks of code or the business processes that are going to change the final say in what amount of time is necessary to complete project activity.

Rules of the road when it comes to performing estimates:

1. If you’re implementing new technology into your current technology stack – maybe you also need to set aside time to install and integrate the technology.
2. Don’t forget process changes.  Sometimes these changes can be just as critical to the project timeline as the technology changes.  Don’t underestimate the impacts these can have across your teams.
3. Don’t forget testing – and I’m not just talking about QA.  I’m talking about Unit Testing, Integration Testing, Parallel Testing, Performance Testing, Regression Testing, New Feature/Function Testing, and UAT Testing.
4. Don’t forget implementation planning and actual implementation activity.  While the universe may have been created with a big bang – in most instances you are going to see implementation activity spread out over a period of time.  Some implementation activity begins before the code touches production, and some activity happens months after the code is in production – you need to account for all of the activity.
5. Don’t forget to provide a buffer for 3^rd party activity.  I have yet to see a 3^rd party engagement go off without some type of change that has to happen.  The initial timeline provided for the engagement never usually is achieved.

Items for consideration when building the estimates:

• How many screen modifications need to be made?
• How many new screens are there in the project?
• How many web services need to be modified?
• How many new web services are there in the project?
• How many reports/data extracts need to be modified?
• How many new reports/data extracts need to be created.
• How many tables/stored procedures need to be modified?
• How many new tables/stored procedures need to be created?
• How many 3^rd party apps need to be touched?
• How many new 3^rd party integration activities are needed?
• How many processes need to be modified?
• How many new processes need to be created?
• What infrastructure changes are needed – servers, server types, routers, switches, security devices?
• What new infrastructure pieces need to be installed – servers, routers, switches, security devices?

Realize that you’re first efforts are going to be wildly off base.  Don’t be afraid to tell your team that you’re giving them time to get better at creating estimates.  This means when they’re wrong – it’s a conversation not a point against them in their annual review.  Use each estimate as a coaching opportunity when the final numbers come in – what they did right and what didn’t look so good and how do we make it better.

This by no means covers it all, but points you in the right direction.  Now I’ve heard some people say that in an Agile process you don’t need to worry about this.  Yeah, right.  Trust me, someone in the organization is telling the CFO what this thing your building is going to cost – how many 2 week iterations it’s going to take to get full functionality.  If you’re not responsible for making those conversations happen today – you will be at some point.  Better to start figuring this out know, before you’re asked to provide that first estimate.

Realize that you’re first efforts are going to be wildly off base.  Don’t be afraid to tell your team that you’re giving them time to get better at creating estimates.  This means when they’re wrong – it’s a conversation not a point against them in their annual review.  Use each estimate as a coaching opportunity when the final numbers come in – what they did right and what didn’t look so good and how do we make it better.

If you'd like more information on my background: LinkedIn Profile