We recently updated the Secure Funding goal diagram, shown below. The change was the addition of the Choose Funding Scope decision point, to recognize that the type of team you’re funding will affect how you fund said team. We have also published details about each decision point and its associated options, including the tradeoffs of each, at the Secure Funding overview page. This is information that we previously had only published in the first Disciplined Agile Delivery book but we’ve decided to start publishing updated versions of this information to the web to make it more accessible to you.
IT Portfolio Management addresses how an IT organization goes about identifying, prioritizing, organizing, and governing their various IT endeavors. Disciplined Agile Portfolio Management seeks to do this in a lightweight and streamlined manner that maximizes the creation of business value in a long-term sustainable manner. IT endeavors typically include solution delivery initiatives/projects, stable product development teams, business experiments (along the lines of a lean startup strategy), and the operation of existing IT-based solutions.
Being agile, having an agile mindset, is foundational to working in an agile manner. The Disciplined Agile Manifesto and the principles of lean software development provide an important start at this mindset. In this blog we explore similar agile philosophies that are specific to successful portfolio management. These philosophies are:
- Keep it simple. Keep your portfolio management activities as streamlined and lightweight as can be. Your goal should be to focus on making good decisions and providing guidance to people, not on maintaining extensive documentation or reviewing documentation. You still may choose to maintain artifacts such as a portfolio roadmap and portfolio budget, but those too should be as lightweight and concise as possible.
- Focus on value over cost. Shifting your mindset from “what is this going to cost?” to “what value will this generate?” is critical to your success because it helps you to focus making better IT investments. You want to invest in IT endeavors that enhance your organization’s ability to produce value for your customers. This in turn provides the profits required for further investment.
- Reduce the cost of delay. One of the strategies for maximizing stakeholder value is to invest in developing functionality that will provide value to the organization soonest. For example, if you delay developing functionality that will generate annual revenue of $10 million for six months you have an effective cost of delay of $5 million because you missed out on half a year of revenue. Disciplined agile portfolio managers consider the cost of developing a solution, the cost of delay that results from putting off said development, and the revenue generated (or cost savings) when calculating the overall value of a solution.
- Invest in streamlining the creation of value. Not only do we want to produce value for our customers, we also want to be good at doing so. The implication is that we sometimes need to invest in ourselves through process or environment improvements.
- Prefer stable teams over project teams. Although portfolio management is often believed to be the oversight of project teams, it really is more about the coordination and oversight of teams in general. In Disciplined Agile we recognize that an initiative is seldom finished at the end of a “project”. There are usually subsequent changes required over time requiring future releases of the solution. The agile community has discovered that long-lived stable teams, whose membership evolves over time, have significant advantages over short-lived project teams. A significant productivity improvement occurs when IT organizations shift away from the project mindset of bringing people to the work and instead decide to bring work to the people (the stable teams).
- Align teams to value streams. These stable teams should be aligned long-term to value streams or lines of business (LOBs). High performing agile teams reliably deliver value frequently to their business stakeholders. As a result the business learns to trust the teams aligned to their areas. This positive feedback cycle maximizes the effectiveness of your agile teams. Additionally, over time they learn more about the business adding to their effectiveness.
- Enable diversity. Every person, every team, and every organization is unique. Every team faces a unique situation that evolves over time. The implication is that teams must be allowed to organize themselves and to tailor their process to meet the context of the situation that they face. This is why the Disciplined Agile framework focuses on providing, and comparing and contrasting, a wealth of process choices. The implication for portfolio managers is that they need to be flexible in their approach. They will work differently with each team because each team works differently, yet they must still provide good guidance to these teams and monitor the effectiveness of each team appropriately.
- Trust but verify. Effective governance is based on enabling and then trusting your teams to do the right thing. However, effective teams monitor themselves through the use of automated dashboard technology and close collaboration, and the metrics that teams collect should be made visible to senior management and other stakeholders so that they may monitor what is happening.
- Govern by risk not by artifacts. Traditional governance often focuses on the review of common artifacts such as requirement documents, architecture models, and test results. Because it is relatively easy for teams to create the documentation that you want to see, in practice there is very little governance value in reviewing these artifacts. Agile governance instead focuses on addressing common risks such as ensuring there is an agree to vision for what the team should accomplish, that the architectural strategy has been proven to be viable early in the lifecycle, and that the team has produced sufficient business value for their stakeholders.
- Rolling wave over annual planning. Annual planning often begins in earnest mid-year which means that prioritized initiatives may not be actually delivered for up to 18 months in the following year. This is not business agility. Make your planning a continuous “rolling wave” activity year round with more detail devoted to planning initiatives no longer than 6 months out. Initiatives planned beyond 6 months should be described at a very high level.
- Prefer small initiatives over large initiatives. It is a proven fact that the larger the initiative the greater the chance of failure. Smaller initiatives are easier to plan and are lower risk to execute. Keep your initiatives small to allow for more frequent delivery of value with less investment in work in progress (WIP).
- Invest in quality. Ensuring that IT delivery teams produce new business value for your organization is clearly important. But so is ensuring that you will still be able to continue doing so a few years from now. To ensure the long-term sustainability of your IT teams you must allow them to make investments in quality. These investments include building high quality assets in the first place but also fixing low quality assets, what agilists refer to as paying down technical debt, as well.
- Optimize the whole. Disciplined agilists are enterprise aware. We choose to optimize the whole instead of locally optimizing a single activity. The implication is that you cannot consider portfolio management on its own but instead must consider it in the context of the other parts of your organization that it affects. A strategy that may make things easier to manage your portfolio, such as having a single way to fund IT delivery teams, may make it easier for your portfolio management efforts but could inflict undue costs and bureaucracy on the teams that you’re funding. For example, does it really make sense for a team asking for $50,000 in funding to go through the same level of rigor as a team asking for $5,000,000? Likely not. Context counts.
Our experience is that the philosophies describe above enable portfolio managers to be more effective in practice. We hope you have found this blog of value and we welcome your feedback.
The basic idea with rolling wave planning is that you plan things that are near in time to you in detail and things that are distant in time at a higher level. The thinking is that the longer away in time that something is the greater the chance that it will change during that time, therefore any investment in thinking through the details is likely wasted. You still want to plan at a high level to both guide your current decisions and to set people’s expectations as to what is likely to come.
Rolling wave planning is implemented in several places of the DA framework. First, as you can see in Figure 1 below, it is an option of the Level of Detail decision point of the Develop Initial Release Plan process goal. A rolling wave approach to release planning has the advantages of more accurate and flexible planning although can be a bit disconcerting to traditional managers who are used to annual planning strategies.
Figure 1. The Develop Initial Release Plan goal diagram.
The Portfolio Management process blade supports rolling wave budgeting as an option for its Manage the Budget decision point. This is depicted in Figure 2. The advantages are greater flexibility and greater likelihood of investing your IT funding more effectively, albeit at the loss of the false predictability provided by an annual budgeting strategy.
Figure 2. The goal diagram for the Portfolio Management process blade.
The Program Management process blade supports rolling wave planning of a program itself, as you seen in Figure 3. Planning and coordination are critical on a large program, and rolling wave planning offers the advantages greater flexibility, the ability to think important cross-team issues through, and the ability to react to changing stakeholder needs. The primary disadvantage is that it can be disconcerting for traditionalists who are used to thinking every thing through from the beginning.
Figure 3. The goal diagram for the Program Management process blade.
As you can see in Figure 4, rolling wave strategies can be applied in Product Management to evolve the business vision/roadmap. A continuous, rolling wave approach is critical to your success because the market place changes so quickly – these days, few organizations can tolerate an annual approach to business planning and in the case of companies with external customers an ad-hoc approach can prove to be too unpredictable for them.
Figure 4. The goal diagram for the Product Management process blade.
Previously we saw that rolling wave strategies can be applied to evolve your technology roadmap, as indicated in the goal diagram for Enterprise Architecture in Figure 5. The advantages of this approach are that your roadmap evolved in sync with both changes in technology and with your organization’s rate of experimentation and learning. The main disadvantage is that your technology roadmap is effectively a moving target.
Figure 5. The goal diagram for the Enterprise Architecture process blade.
As you can see, rolling wave strategies are an integral part of the Disciplined Agile (DA) framework. In fact, in most situations they prove to be the most effective and flexible strategies available to you. The advantages of rolling wave planning tend to greatly outweigh the disadvantages. More on this next time.
On February 23, 2016 I gave a webinar entitled (In Agile) Where do all the Managers go? A recording of the webinar is posted on Youtube and a PDF of the slides on Slideshare. This blog overviews the webinar and provides answers to the numerous questions that were asked during it.
The webinar began with a discussion of four trends that are reducing the need for people in management positions:
- Technical management tasks are performed by the team. As a result there is much less work for managers to do.
- Leadership is addressed by new roles. Team leadership responsibilities are in the hands of non-managers.
- Experienced organizations are moving towards stable teams. Important side effects of this are that much less “resource management” is required and team budgeting is greatly simplified.
- Status reporting is being automated away. Once again, less work for managers to do.
We then discussed the options that existing managers have in an agile environment. In Disciplined Agile there are four roles that existing managers are likely to transition to: Team Lead, Product Owner, Team Member, and Specialist. Specialist roles – such as Data Manager, Portfolio Manager, Program Manager, and Operations Manager – occur at scale and the corresponding positions are few and far between. Read the article Disciplined Agile Roles at Scale for more details.
We end with words of advice for existing managers: Observe what is actually happening; be flexible; and choose to evolve.
Questions and Answers
We’ve organized the questions into the following topics:
- Evolving to new roles
- Addressing “management activities”
- Potential management roles
- People management
- Management reporting
- During your agile transformation
- Stable teams
- Training and certification
Evolving to New Roles
Will not the existing technical managers be disappointed with only people management work?
That depends on the person. Some will be very happy to do this, some will not.
How will managers fit into a leader role?
It depends on the person again. Some managers are very good leaders right now, some have the potential to be good leaders, and some don’t. They will need training and coaching to fit into their new role(s).
Addressing “Management Activities”
If there are no PMs in Agile, who handles communication with clients (meeting deadlines, priorities, etc.)
The Product Owner.
How does individual performance to be taken up in Agile team? I think that is more crucial and challenging for Agile Leader / Manager.
It is always difficult to address performance-related activities. There are many lines of thought on how to do this. The most progressive is for the Team Lead to provide feedback to team members on a just-in-time basis. If the Team Lead seems behaviour, either desirable or undesirable, but a team member then they should comment on it right away so as to reinforce or dissuade it as soon as possible. Many organizations still have an annual review process, a strategy that many organizations have abandoned due to it’s ineffectiveness, where functional managers get involved with the review process.
I have seen that you have selected the Team Lead as the responsible of assess team members and budgeting the project. In Scrum the Product Owner is compare to a CEO that’s the reason I would say the Product Owner is responsible for bugdeting and about assessing I prefer a more democratic form which involve all the members. So what do you think about PO managing the budget and a democratic assessing vs one single vision assess?
Yes, I misspoke during the webinar. The Product Owner is often responsible for the team’s budget and is responsible for reporting the current financial information to the stakeholders. The Team Lead is often responsible for similar reporting to their management team.
Having multiple people involved with reviews/feedback is usually a pretty good idea. The People Management process blade captures several potential strategies. However, it is still a good idea for the Team Lead to provide feedback as well, see my earlier answer.
Potential Management Roles
I think there is still a need a bridge manager role between Finance, Teams, and PMO type orgs to ensure Product owners have budget… views. Thoughts?
In smaller organizations this likely isn’t an issue. In larger organizations there is often a Portfolio Management effort that is responsible for such issues.
What might be potential responsibilities of an Operations Manager, Data Manager, …?
Please read the article Disciplined Agile Roles at Scale for descriptions of these roles.
Okay, the data management team needs a team manager/leader. Are large organizations using various resource managers? (Although would be less necessary with stable teams I would think)
Exactly. Large organizations still tend to have people in resource manager roles, although sometimes they have different titles such as CoE Lead or HR Manager, but with stable teams they need far fewer of them.
If the team has a Team Lead/Scrum Master that is only the servant leader for 1-2 teams, is it suitable to have people managers?
What value would a “people manager” bring to the team? This is the fundamental dilemma for managers, for everyone for that matter, when an organization moves to agile ways of working. If they’re not bringing real value to the team then they either need to find ways to do so, which likely isn’t whatever management activities they’re trying to cling to, or they need to go elsewhere and try to add value there.
Do you intend to update the DA 2.0 interative pic on the DAD site to talk about “Potential Management Roles at Scale” as mentioned in page 18 of this presentation?
Yes. We actually have something in beta that we haven’t released yet. We’re just about to release an update to the main picture, which in turn requires an update to the role version of the interactive pic.
What is the most basic difference between Project/Program/Portfolio Managers in Agile?
Quick answer is that there isn’t Project Managers in Disciplined Agile nor in methods such as Scrum, XP, and so on. At the program level (a large team of teams) you likely need someone in a Program Manager (or more accurately Program Coordinator) role to coordinate activities (see the Program Management process blade for details). A Portfolio Manager is focused on the IT level and should be concerned about pre-development activities, development/delivery teams that are currently in flight, as well as operational activities.
Also, please read the article Disciplined Agile Roles at Scale for descriptions of these roles.
How does one manage the career path of the Team Leads? Is there career progression beyond a TL to be a specialist or does s/he continue being a TL throughout his career?
Everyone is different, so there isn’t one exact answer. It depends on what the person wants to do and what positions are available to them. If their desire is to move into management then there are fewer IT management positions available to them. If they want to become an AO or PO then they need to work towards getting the skills and experience to fulfill those sorts of roles. The People Management process blade includes career management strategies.
How do you evaluate what roles are/will be necessary?
It depends on the needs of the team in the situation that they face. The primary delivery roles typically exist on all delivery teams and the secondary roles start to appear at scale.
How do you see the role of a BA in agile?
Most existing BAs, like most existing project managers, will need to transition to other roles. However, at scale there is a need for some people in the specialist BA role. I recently has a user group presentation recorded on this very topic. See Disciplined Agile Business Analysis: Lessons from the Trenches.
How do we approach a situation where management wants weekly status reports from a Program Manager who can combine both Team Lead & Product Owner roles, as well as manage multiple projects that may be similar in nature or not.
A few thoughts on this:
- It’s an incredibly bad idea to combine the TL and PO roles because it puts too much responsibility in the hands of one person. Furthermore, putting it into the hands of a former manager, someone who may have a command-and-control mindset instead of the collaborative mindset required of agile, can exacerbate the problem.
- You may need someone that Team Leads should work with to coordinate activities between teams (such as a Program Manager or Portfolio Manager) and someone that Product Owners should work with (a Chief Product Owner) to coordinate requirements activities. See The Product Owner Team.
- I do see stuff like this happen when organizations are transitioning to agile. They are still learning how to make agile work within their environment, they have a lot of people who haven’t yet made the transition, and they have a lot of middle management staff whom they want to treat fairly by finding them other work. Sadly that other work is often overhead that can be done away with given a bit of thinking.
- If you institute automated dashboards, what we originally referred to as Development Intelligence in Disciplined Agile, then a lot of your status reporting goes away.
In a typical organization, where to team lead(s) report into?
It depends. We’ve seen them report into a Program Manager or a Portfolio Manager. During the transition effort a Project Management Office (PMO) may still exist so Team Leads might report into there, although we often find that there’s a serious cultural and mindset difference that can be very frustrating for everyone involved.
During Your Agile Transformation
What about managers being responsible to support an agile transformation journey in a large organization?
Yes, they would very likely be working as part of an Agile Center of Excellence (CoE), although that would be mostly staffed by experienced agile coaches. There is a need for one or more senior execs to sponsor your agile transformation.
How to deal with “Project Manager” role renamed as “Agile Project Manager” but expected to do the same responsibilities as traditional PM?
We see this sort of stuff all the time unfortunately. First thing to do is to get these people educated in how agile actually works in practice, we’d suggest DA 101: The Disciplined Agile Experience or DA 104: Introduction to Disciplined Agile as your best option to get the whole picture. Next, work through with them how they would actually add real value on the team (see the discussions earlier). Very likely many of the activities that they think need to be are being handled by someone else or have been automated away. Third, get them some coaching to help them to truly transition to agile.
On my project, I am the Team Lead and there is a Project Manager. So far, I have observed that there are several conflicts in responsibilities. How do we come to an agreement of who handles which responsibilities? For my next project, would you suggest I work on a project with no project manager?
We often have to run facilitated workshops in organizations where we work through the roles and responsibilities that are needed in practice. We do this with a wide range of people and we do so in a collaborative and public manner. You need to come to an agreement as to who does what. Doesn’t sound like that’s happened in your case. When you work through this sort of an exercise you quickly discover that you don’t need a project manager, although there may be some project control officer (PCO) responsibilities that would be assigned to either the team lead or some sort of administrative role (such as PCO).
Do you have any advice on how to deal with the removal of the traditional hierarchy – in a flattening of responsibilities, ‘reporting-lines’ and salaries (or having a vast range of skills and pay-scales all with a job title of ‘team member’?)
This is what an agile transformation will accomplish for your organization. It takes time and investment in your people to implement. I highly suggest that you get some experienced coaches to help you do this.
Management is severely, negatively, personally affected by Agile, and will not look fondly upon it in many cases. Any tips to reduce this? Do you recommend mass management reduction, or multiple smaller rounds?
The first step is to recognize that your organization doesn’t exist to create jobs for managers, regardless of what the managers may think. Agile is about focusing on value, so why wouldn’t a good manager be interested in being actively involved with doing so? My recommendation is always to get training and coaching for everyone, including managers. As I described in the webinar there are many options for existing managers in the agile world if they’re willing to be flexible and evolve. Your organization should choose to help people make these transitions to new roles. However, if people are not willing to make the transition then they shouldn’t be surprised if the find themselves being asked to seek employment elsewhere.
It sounds like the person asking about “who’s responsible for delivery” might have used “responsible” when they meant “accountable” – many managers are the single wringable neck for something in their job description. Do you feel Agile draws the same distinction between the two like ITSM, for instance, does?
Agile is based on a collaborative, teamwork-based mindset. Having said that, it does make sense to have someone ultimately responsible for certain things. For example, the Product Owner is responsible for prioritizing the work on an agile team. Similarly, you may have someone in the Release Manager role who is responsible for overall Release Management within your organization. This is particularly important for regulatory environments where by law you need to have someone not involved with development who makes the final decision as to whether the solution is released or not.
From your observations and experience, what is the average timeframe for the managers number to decrease? How long does the process of the shift take?
It depends. We’ve seen this happen over timeframes as short as six months to several years. With solid coaching this process will go a lot faster and smoother.
How to motivate and enable senior leaders to give up control?
In agile, particularly in Disciplined Agile, senior leaders have greater visibility and opportunities to steer than what they had in the traditional world. What they need to do is give up their false sense of control that traditional strategies provide. The real issue usually isn’t senior leaders but instead is middle management. They are the people who are currently performing many of the management tasks that are implemented in a more streamlined manner following agile approaches.
How can middle management start the agility journey when top leaders are not yet on board?
Agile typically begins following a stealth adoption strategy where senior leaders are unaware that it’s happening. The point is that anyone, including middle management, can start adopting agile strategies long before senior leadership gets involved. Strategies such as working collaboratively, enabling your team(s) to plan and organize their own work, adopting dashboard technology, and streamlining the bureaucracy whenever possible is very possible to accomplish on your own.
Thanks for being frank about the role(s) for managers in an evolving Agile culture: Agree, traditional project management organization’s aren’t highlighting these trends (and positive outcomes.)
You’re welcome. Traditional project management organizations often go at it from the point of view of how to continue justifying management activities. We go at it from the point of view of how to improve your overall organizational effectiveness and as a result come to a different conclusion.
Will the idea for stable team become stale after some years? People tend to get frustrated doing same work. What’s the solution in that case?
Stable teams evolve over time. You’ll get people joining the team every so often and similarly leaving the team every so often. It’s natural for people to want to move on and try something new every few years. As a result your organization will still need People Management activities in place that motivate and enable people to manage their careers.
As far as stable teams go, commonly Valve, Inc. is referred to a place where teams are formed around projects that the team members find the most interesting. Project leaders try to sell their project to get developers. Your thoughts?
This is great technique that other organizations may be able to adopt. Allowing teams to form themselves is likely the most effective way to do so. However, like all strategies, there are some potential disadvantages. Team culture may become ingrained and they will not attract people with a different culture who would have the potential to add some real value to the team otherwise.
Is there a method to build the stable teams? Domain, Product, line of business?
There are several strategies for doing this. The most common is to form feature teams that do all of the work to implement a feature as a vertical slice through your entire infrastructure. Another approach is to form component teams that work on a technical or domain component/framework/LoB. A third approach is internal open source. We’ve discussed these strategies in greater detail at Strategies for Organizing Large Agile Teams.
Do you think stable teams concept will work in service-based organisation?
Yes. It’s a bit more difficult because you’d be bringing entire customer projects to the team at once instead of a flow of smaller features. Of course you can break each large project up into smaller features and feed them to teams in an interleaved manner, requiring a sophisticated approach to requirements management.
Training and Certification
What baseline training do you recommend for agile managers?
A good place to start is training on agile thinking, often referred to as how to be agile. Then I would recommend training that describes the full delivery lifecycle from end-to-end, something like DA 101: The Disciplined Agile Experience or DA 104: Introduction to Disciplined Agile. You want to understand all aspects of the agile delivery process, not just the management ones. Scrum training is popular but far too narrow. SAFe training isn’t for beginners.
I would like to participate in a certification workshop/further training. There doesn’t seem to be many offerings in the US. Are there plans to expand training opportunities in the states?
Yes. In fact we have training coming up in the Baltimore area in March and Philadelphia in April. We will have more open enrollment workshops scheduled soon. Please visit the homepage of the Disciplined Agile Consortium for a listing of upcoming public workshops.
What should we be telling folks that have PMP’s – are they still valid? is PMP training moving toward Agile software development.
Yes, the PMI is moving towards agile but they have a very large ship to turn. Unfortunately the PMI training tends to suffer from the challenges that I described earlier – it seems to promote a rather unrealistic vision of how managers can potentially fit into agile.
In IT we are often asked to estimate the expected time/schedule or cost of software development. Sadly, the desire of stakeholders to have “predictable” schedules or costs results in significant dysfunction within a software development team. When a software team is forced by their stakeholders to commit to a schedule/cost they must then ensure that the schedule/cost doesn’t slip. For example, to protect themselves from increased time and cost due to scope creep, software development teams will make it difficult for stakeholders to change their requirements during Construction and even go so far as to drop promised scope late in a project. The desire of stakeholders to reduce their financial risk often results in behaviors by the software development team that ensure that stakeholders don’t get what they actually want. Naturally IT gets blamed for this.
We need to do better. In this blog we summarize the things that we know to be true about software development estimation. In no particular order, they are:
- Estimates are guesses. Look up the word in the dictionary – An estimate is a rough approximation or calculation, in other words a guess. Unfortunately, too many people think that estimates are promises, or worse yet guarantees. In our opinion “guesstimate” is a far more appropriate word that “estimate”.
- Scope on IT projects is a moving target. Our stakeholders struggle to tell us what they want and even when they do they change their minds anyway. Any guesstimate based on varying scope must also vary in kind.
- Guesstimates are probability distributions. Although your stakeholders may ask for a fixed amount guesstimate, for example “this will cost $1 million”, the reality is that there’s a chance the cost will be less than $1 million and a very good chance that it will be more. There is ample evidence that the initial estimate for a software development project should be given in a range of -25% to +75%, so your million dollar project should be quoted as a range of $750,000 and $1,750,000. This is shown in the diagram above by the green distribution curve. In many organizations this can be politically difficult to do, and strangely enough in many cases stakeholders prefer to be lied to (it’s going to be $1,000,000) rather than be told the truth.
- Guesstimates must reflect the quality of the inputs. A guesstimate needs to reflect the quality of the information going into it – if your scope is fuzzy your guesstimate based on that scope needs to be equally fuzzy. Sometimes stakeholders want a guesstimate with a tight range, perhaps +/- 10% (the red curve in the diagram), early in a project. To provide a tight range such as this you need to have a very good understanding of the requirements and the design. Early in the software development process this can only be done through more detailed modeling, an expensive and risky proposition which often proves to be a wasted effort because the requirements will evolve over time anyway.
- Guesstimates anchor perception. The primary danger of providing guesstimates to people is that they believe them. Tell someone that it’s going to be $1,000,000 and they fixate on that cost even while they are changing their minds. Tell them that it’s going to be between $750,000 and $1,750,000 and most people will fixate on the cost of $750,000. Some people will focus on the average cost of $1,250,000 even though the median was $1,000,000 (guesstimates are in effect Weibull probability distributions).
- It’s easier to guesstimate small things. ‘Nuff said.
- It’s easier to guesstimate work you’re just about to do instead of work in the distant future. It is much easier to identify the details of work to be done right now, and thus turn a large piece of work into a collection of smaller pieces that are easier to guesstimate. In part this is because you have a much better understanding of the current situation you are working in and in part because you are more focused on the here and now.
- The people doing the work will likely give a better guesstimate. They are more motivated to get the guesstimate right, particularly when they must commit to it, and have a much better idea of their abilities. Granted, someone may need to coach people through the guesstimation effort. In Disciplined Agile this is a responsibility of the team lead.
- Someone who has done the work before will give a better guesstimate than someone who hasn’t. Experience counts.
- Guesstimates reflect the situation that you face. Which organizational situation do you think will result in a short schedule and lower cost: Five people co-located in a single room or the same five people working from different locations in difficult time zones? Or how about a team working under regulatory constraints versus the same team without those constraints? Context counts.
- Multiple guesstimates are better than a single guesstimate. Getting guesstimates from several people provides insights from several points of view, hopefully prompting an intelligent conversation that enables you to develop a guesstimate with better confidence. Similarly, the same person producing guesstimates for the same piece of work using different guesstimation strategies will also provide a range of answers that you can combine.
- Guesstimates should be updated over time. As your understanding of what stakeholders want improves, and your understanding of how well your team works together, you should update your guesstimates. As your understanding of the fundamental inputs into your estimate improves you are able to produce a better estimate, thus enabling the stakeholders of that guesstimate to make better decisions.
- It costs money to produce a guesstimate. The precision of an estimate is driven by the detail and stability of the information going into it. Want a tighter range on your estimate? Then you’re going to have to have a better handle on the requirements, design, and capabilities of the team doing the work. This greater precision requires greater cost. The fundamental question posed by the #NoEstimates community is effectively “Is the value of improved decision making capability from having the guesstimate greater than the cost of creating the guesstimate?” The implication is that you must ensure the cost is much less than the benefit, hence their focus on finding ways to streamline and even eliminate the guesstimation effort.
- Guesstimation is far more art than science. See point #1 about estimates being guesses. The best guesstimates are done by the people doing the work, just before they need to do the work, for small pieces of work.
- Formal software guesstimation schemes are little more than a scientific façade. Function point counting, feature point counting, and COCOMO II are all examples of formal strategies. They boil down to generating numeric guesses from your detailed requirements and design, plugging these guesses into an algorithm which then produces a guesstimate. These are all expensive strategies (they require detailed requirements and design work to be performed) that prove to be risky (because they often force you into a waterfall approach) in practice. Yes, they do in fact work to some extent, but in practice there are much less expensive and less risky strategies to choose from. People like these type of guesstimation strategies because they provide a false sense of security due to their complexity and cost.
- Past history isn’t as valuable as people hope. Some formal guesstimation strategies are based on past history, but this proves to be a false foundation from which to build upon for several reasons. First, people have different levels of capability which change over time as they learn. Capers Jones has shown that developers have productivity ranges of 1 to 25, the implication being that if you don’t know exactly who is on a team and how well they work together your corporate history will be questionable. Second, technologies evolve quickly so past history from working with older versions of technologies or completely different technologies becomes questionable at best. Third, people and teams change (hopefully for the better) over time, implying that an input into your guesstimate is fuzzy at best. Fourth, because every team is unique and faces a unique situation basing estimates on past history from other teams in different situations proves questionable.
- Beware professional guesstimators. They tend to break many of the rules we’ve described above.
To summarize, when you are required to provide estimates for your software development efforts that you should take a pragmatic, light-weight approach to doing so. This blog posting has provided many practical insights that should help guide your decisions. These insights and many more, are built right into the Disciplined Agile framework.
We recently published an article describing how the Disciplined Agile Delivery (DAD) framework is being extended to cover Portfolio Management activities. The following diagram depicts the DAD goal diagram for Portfolio Management, and as you would expect your organizations has a range of options to choose from. In this diagram we use the term endeavor to refer to a project, product, or experiment.
The article describes each one of these process factors – Identify Endeavors, Explore Potential Endeavors, and so on – in greater detail. A future version of the article will describe the strategies/practices associated with each factor. We have also included a high-level workflow diagram, see below, to overview from the point of view of the Portfolio Management process blade how it fits into the rest of the Disciplined Agile Delivery framework.
An interesting aspect of the flow diagram is that it shows the relationship of Portfolio Management to other IT Plan activities. For example, Enterprise Architecture provides a Technology Roadmap and Product Management provides a Business Roadmap and an indication of stakeholder priorities – all critical information that are inputs into the efforts around prioritizing potential endeavors. The point is that this diagram reflects workflow, not organizational design. For example, in some companies there is no Product Management team, instead responsibility for the Product Management activities are spread out amongst several teams. A common strategy is to have the Portfolio Management team responsible for understanding stakeholder priorities and the Enterprise Architecture team responsible for the business roadmap. Another approach would be to have the Portfolio Management team subsume both of those activities. A third approach would be to have three teams, one for each of Enterprise Architecture, Product Management, and Portfolio Management. Different organizations will make different organizational design decisions, and of course these decisions will evolve over time. As always, context counts.
I hope that you find the more detailed Portfolio Management article to be of interest.