What is the role of a manager in Agile project? Do we need Manager at all?
 

A switch from waterfall to agile project and suddenly you would start seeing whole new world. Is this shift complicated? How can I be successful as manager in agile project? How does it defer from traditional methodology?  And there are many more questions which every manager who starts with agile project or managers already part of agile project come up with.  Most importantly what is my ROLE?  Lets explore

As a manager, when you get into a project you want to manage the day to day deliverables and processes but wait that’s a job of a scrum master, you move on to prioritize the work but that’s again product owner owns. Wow, then you jump on to assign work and get involved in day to day activities but wait again, isn’t your teams are self-organizing?

By heart, I am a software engineer. Can I code? Off course you can if you don’t have any other work and moreover you are expected to become a team member.  Frustrating, well yes and no. If you are quite adamant and not ready to unlearn some of the traditional practices as manager, you would be biggest bottleneck in getting your project successful.  You can’t do many things in agile while you can control everything by keeping some distance while taking complete ownership.

As a manager, I would suggest you to do the following

  • Motivate team – Constantly encourage your team members. Agile focus on every individual and a motivated team member can do wonders. If you are thinking of spoon feeding agile member then either the member has too many challenges or you are not a right manager.
  • Celebrate victories – Find an opportunity to celebrate your victories. Even if that are small as every celebration would enhance your team motivation and bring in lots of energy to system.
  • Focus on business priorities – Constantly focus on your business priorities, contribute in vision, drive vision by ensuring work gets done, keep abreast with business demand. Learn domain and make your team gain expertise over that. Build a model to train new joiners quickly. Encourage collaboration and encourage teams to discuss the business.
  • Be open – This is one of the biggest challenge I see in this industry. You don’t want to hurt anybody and with that you always go with giving positive feedback. Remember the areas of improvement and constructive feedback helps an individual to grow. You are supporting short term growth by killing his/her long term growth as well as organization priorities. You need to constantly strive to build your team and focus on every individual to be successful. More you focus on people, better off your results are. Rate your people as they are. You need to motivate your high performers and bring you low performers to get better. Always remember that you need to get better constantly.
  • Generate Metrics and constantly monitor them – The world is not perfect place. Teams which are doing amazing, may go down. Constantly monitor productivity, prod issues, defect rate and whatever is important in your organization.
  • Trust team and encourage the culture of trusting people – In my experience, if you have an average team, and that team trust each other and ready to support each other, they would end up doing much more and better compare to your star team where you see conflicts. I have seen multiple examples in past where the whole team gel extremely well and the result you see are outstanding. Not just project did the best in those scenarios while the whole team has gotten bigger rewards and great careers.  The dedication to team result in delivering values. This holds true all the times.
  • Invest in resources – Hire best talent. Further send them for trainings, Encourage cross KT culture, plan their travel if their teams are in different country city, constantly keep in touch with them and make them feel that the org does care for them. Many a times in bigger organization individuals are lost while it is primary responsibility of a project manager to ensure they are being taken care of. One of the tool I used heavily is to ask individuals their aspirations and constantly seek opportunities to fulfill them. You may not be able to do it all but creating a path is kind of easy. In fact, many a times I see individual aspirations are so easy that it can be fulfilled in no time and little effort.
  • Process improvements – Let scrum master handles processes while constantly review to see what improvements can be made further. Improving on test coverage, adhering to standard practices, encouraging people to take risks etc. For instance you see all retrospective items are great while after a meeting we don’t do anything about it (Unfortunately most agile team has this challenge). Well, here you can have action items and ask team to assign that. The next retrospective should start with older action items. You can always work with scrum master to get processes better.
  • 1×1 Meetings – This is very powerful tool. As a manager, I would strongly recommend you to keep talking to team members. Appreciate them for their great work. Constantly provide feedback on areas they should get better. Understand their perspective and identify action items for you. Make yourself available for team all the times. I am being often told by managers coming from traditional model that don’t be easily available for your team as you may end up doing more and people won’t value you. I am completely against of that idea. I think you should meet up your team as much as you can.
  • Empowerment – Find right people and empower them to do certain things. Build your next set of leaders. Trust people. If you manage multiple agile teams then build leader within each team and empower them. Practically you shouldn’t be very busy all the times. You should  be available for your team as soon as possible when they need you.
  • Improve system – You should be constantly looking at everything in the system and find opportunities to improve them. Improving productivity, quality, trust, competence, delivery, processes and many more.
  • Seek challenges – The seeking challenges is always better option over embracing challenges. You should be constantly seeking challenges and building the same culture around that.
  • Fun place to work– Create an environment where everyone feels like coming to office. There should be lots of fun. Believe in starting your day with fun and it should be throughout. Don’t have a constraint that I would allow you to have fun when we achieve something. Although you do celebrate every victory but fun is something should be there all the times.

Always remember that things which you can do directly, that’s great while things or tasks which agile doesn’t allow you to do it, get yourself involved indirectly and get them accomplished. At the end of day, your biggest priority is make your team successful. If you can achieve that, rest all fall in place.

Story Points Vs Time
 

The story point is high level estimation based on complexity before the work begins on story while the hour based estimation is just more concrete estimation where effort is represented in hours. The amount of work a team can accomplish during a single sprint is called Velocity. It is calculated at the end of Sprint by adding all the story points which are done-done. 

Velocity is a measure of the amount of work a Team can tackle during a single Sprint and is the key metric in Scrum. Velocity is calculated at the end of theSprint by totaling the Points for all fully completed User Stories

Now the question is how does story points relate to time and this is certainly one of the debated topic among sprint teams and I see confusion all around. I heard from individual team members many a times that story point can never be related to time in any way. One cannot say 1 story point equivalent to 2 hours or 5 hours or any other number. The second statement is right while this creates a notion that story point has no relation with time. When you have 3 story points, it could take 3 days or up to 50 days. This doesn’t seem right and this is actually not RIGHT. “Whatever time it takes is fine after-all one is putting on effort” effect is what visible .

The reality is, the story point states EFFORT. The estimation for team member X can be 10 hours for 3 story points while other it could be 20 hours. The moment you say effort, it is size. But the size is relative. Hence it proves the point that it is not exact hours while it’s relative hours. The estimate is for team and they should be able to say what is the biggest size in terms of story points can be picked which would be done in sprint. Moreover if we say, team velocity is 40 story points that means we can complete 40 story points for this team in N week sprint.
If one doesn’t want to relate this to size at all, then I suggest them to stop doing estimation as in that it case it would be just waste of time. Typically a story point suggest min and max time which might be taken to accomplish that story. For instance 3 story point for team P means effort between 3 and 5 days. It could be plus or minus but the point I have is to have some indication with respect to size.

I have seen individuals just change the story point for a given story if they can’t complete the same in stipulated time. I understand the estimation was not done right or probably we might not have done estimation but adding/changing story point or hours post the work is being done is pathetic. It completely wastes the time spent on estimation and further we lose all the value gain with estimation. This practice is primarily meant for inflating numbers to have better reports. This MUST be stopped.

I have been often asked that the team is not able to predict well because they are new or may not be competent enough to find the right story points. How should we handle that as it is important to be predictable as you can’t run the business in vacuum? When your boss or business ask how long it takes to complete the effort, you still have to answer probably by looking at release planning and estimation. This shows another point why we have to relate the effort to hours (Not the exact) and here min and max does the job.

As we are talking about story points, there are many teams who would just add story points to story while for individual tasks they don’t assign hours. The burndown chart in that case is driven by task (Task based or hour based) which is in a way weird because you suddenly see ups and downs. The whole value of one of the most critical artifact in agile is not being materialized. The planned and remaining hours is important in individual tasks.

What happens when you are in the sprint? How to make it more meaningful?
 

There is enough motivation I have to write this article as the whole intention is to answer the questions which I have been asked multiple times pertaining to sprints I dealt with.

Some of which are listed below but not limited to

– How can we ensure sprint success?
– Should we do more or less?
– Why continuous improvement?
– How can I be successful? How can I grow faster?
– How to get better at estimation?
– Am I following the right set of processes?
– Is pairing good and how should we pair?
– Am I being heard? Should I challenge people? What if the team is very senior compare to me?
– The standup happens in the evening because of the offshore and onshore model? How to make it effective?

And the list is endless.

Let’s assume you are already in sprint cycle. One of the important questions is how do you provide an estimate for items which are already prioritized for sprint?  The majority of newbies teams with limited exposure to agile struggle due to unknowns during mid-sprint which was not accounted for. The estimation might have given with many assumptions. The predictability is the biggest concern in most agile teams. One of the easiest ways is to pre-groom stories well in advance.  Typically for two weeks sprint, I encourage the team to spend 10-15% of time daily during the second week of a sprint to get as many details as possible for next sprint stories, create tasks, identify dependencies, quick spikes to find out the feasibility.

The typical myth most people have that we should be adding only tasks having business values while it might be a look good factor for product owner but from a developer standpoint, this is something we should avoid. In addition to functional and non-functional requirements, a team should add testing, documentation, and everything that they would work on as a task. This given better clarity and sharing of work.

When you constantly spend time in grooming next week sprint story, you are very confident how much worth of work you can do it in next sprint when you get on to sprint planning as your estimates are good enough with confidence that what you are going to build. As this practice continues, there is a point which comes approx. in 3 to 4 sprints where the team doesn’t even pick a work which is not clear.

In summary, when a team starts with sprint, during the planning meeting, everyone talks about stories and tasks (Tasking is done). The goal is to add/remove items might not have been considered and provide estimates. The commitment is made with lots of confidence. You can never avoid unknowns while the chances of getting it reduced significantly increases.

Just to prove above point, once I have enforced SRS (Software Requirement Specification) in agile where it must have to be created and approved before planning a day in order for that item to be considered for planning. The team productivity dropped for 30-40% in first two weeks to get settled with new process inception and spending time in creating a document while as the time passed, the team who was not confident in doing 35 story points, hitting close to 60+ story points. All that happens in less than 4 sprints. Worth it!! Isn’t it?

This approach is phenomenal when the team is co-located while teams which are not, it can be slightly modified. For instance, the team at location x can pre-groom few stories and location y does the same with another set of stories.

There is another interesting thing which I have seen in offshore and onshore are sharing the same stories. The challenge is what I work (assuming I am at offshore), I need to hand off to onshore and vice versa. For 6 day worth of story, I and my counterpart need to understand what other person has done. The wastage of time in handoff and further understanding work done by counterpart every day by both ends adds extra hours.

I have seen instances when a story which can be done in 6 days took more than 6 days because of the addition of another person. Sounds funny? Yes, it is. Even if in extremely positive scenarios, it would add 50% extra time. A lot of people think we can avoid dependency and everyone should know everything, while it is funnier that the horizon of learning is limited for an individual if you look at long term. I am not saying that we shouldn’t do that but we need to be wary of taking a right decision. For instance, if there is a production issue or a complicated story which we want more than one person to learn/work or a story must have to be completed in little lesser time. With these kinds of practices, trade-off is understood. Compromising productivity to achieve temporary priority is fine in this case.

When the team is aggressive, they pick more points and to make up that quality is being compromised. Or the team may not be able to do that, which is even worse as it demotivates the team. I always tell my all teams to start with less (less than they think they can really do it). Further, as we progress, keep on taking more and more. The constant improvement is really important.

As discussed earlier, the pre-grooming or creating SRS very much fix this issue.

In order to have your sprint successful below is what I think are very critical

– Trust each other in team
– Give focused hours. Core working hours are really important
– Celebrate every victory
– Appreciate your team members
– Provide feedback openly
– Fail faster

The “I” factor usually becomes a challenge in many sprint teams. I will pick the certain specific piece and create dependency or I am more competent, and I would do my own over working with a team. The moment you see people like this in your team, get that fixed or move those out of the team. They just create a nuisance and spoil team culture. This is just a myth for an individual as it’s very easy to replace anybody in agile. The only worry, I see here is spoiling team culture.

As an individual team member of team, the success of him/her is

Constantly learning, learning faster
Team Success
Fun working

I have seen less competent team while better at supporting and trusting each other ended up doing much better in terms of quality, quantity and achieving business goals compared to high experienced teams with team culture lacking. Agile is all about self-organizing teams which move up in ladder by helping each other to grow. It is so beautiful that everything is visible and individuals with “I” and “We” just do not grow. If you are not able to openly give feedback to your team member because you are scared, you would end up being just mediocre.  I do less because sprint success is important and does it matter if I work just 2 hours a day because I am completed all the work I planned. Sounds familiar?

Another interesting observation what I have from some of the developers is if I don’t contribute it is anyway not visible because I can show I am pairing or stuck on complicated issues or whatever. It’s is a just wonderful myth. The agile gives so much of clarity that everything is visible.

When it comes to releasing software, I would say release as many times as you can. This would force you to think a need of having lots of tests. Again one of the common challenges is, most teams focus well on the unit test while they hardly create integration test. That is bigger killer in long run. A lot of great teams are given 20% to 30% time in a sprint just to refactor code, working on extra tests which are not there or working on nonfunctional requirements. The challenge is if it’s not planned, it doesn’t happen. The human psychology must have to be addressed. If I am given a work on Monday and I am expected to do this by Friday, I would do that on Friday even if I could have done that in 2 days.  In addition to that when you have a task it has to be done.

How great self-organizing team you have, it is always good to add work in sprint board.

The fear is another cause of doing less or not doing RIGHT. Many times teams are afraid that they may not be able to complete the work hence they pick less and even though work gets completed in advance, they keep picking less. The extra time remaining is good for learning new stuff and working on tech debts. 1/2 day or one-day learning is decent for two weeks of sprint while for tech debt and other stuff, planning results better outcome.  When the team trusts each other, they as the team always take credit for success and accountable for failures. It sounds better. Indeed it is. Agile encourage you to fail while with every failure you should learn and avoid repeating that mistake. When you are working on existing complicated product, and while working on the story you see there is an opportunity to fix existing bug or refactor some code, I would say you should do it. Or plan to do it ASAP.

If your team or PO is discouraging you from doing that, talk as a team. Your goal is not just to fix issue what customer reported while you should be constantly striving to make your product better. The priority can be discussed and trade off can be identified while If you don’t do it, it would never happen.

Let’s talk about scope creep. The scope creep is nothing but change of scope or change in requirements

If it’s in middle of sprint, there are only two ways to manage it

– Push it to next sprint
– Remove the same amount of work to pick new work (Do it only when it’s VERY important). Agile completely discourages it but when there is a prod issue or something similar to that sort pops in, you might have to take wise decision)

If it’s outside of sprint

Agile is excellent, it depends on the SOW you have signed. Primarily, you can take something out or you can add another sprint to your total number of sprint or whatever you have agreed on.

I personally don’t allow a team to pick work which comes during the middle of the sprint. It creates lots of confusion. Until and unless it is extremely important, you should not be picking it up. The extreme urgency like prod issue comes from customer etc. Many of the things can be planned by keeping some 10 to 20% of bandwidth to manage these support issues. If there is a huge work which would spoil the sprint, the best thing to do is to start a new sprint.

Another practice which I feel goes pretty well when you have offshore and onshore model, is to have quick sync meeting in morning. Typically standup happens in the evening which is nothing but post-mortem or status meeting who is attending that in evening. It doesn’t have the benefits of stand-up at all hence meeting in morning for such team adds lots of value.

The demo and showcase are very important, and I always prefer a practice of showing it to team and PO as and when it’s available. How great the requirement elicitation is, there are always things which get missed. I have seen teams doing Agile with waterfall model and I call that as “waterfall agile”. Most of the team members work on development for first 70% of days and in the end, the focus is on QA which is ridiculous. Ideally, you should work on development, testing, and showcase. If possible push that code to prod as soon as you can and don’t wait for the whole sprint to get completed. There is an exception to this is when you have bigger products and many customers are using it. This kind of scenarios you should be looking at a number of tests you have and what kind of dependencies to be addressed.

Agile is really very simple hence it doesn’t allow you to do too much tailoring, unlike waterfall method. Every process is simple but it has to be done diligently. You don’t work as individual while you work as a team.

PMI-Agile Certified Practitioner
 

Project Management Institute – Agile Certified Practitioner

The Agile Certified Practitioner formally recognizes your knowledge of agile principles and your skill with agile techniques. This is one of the most valued certification when it comes to Agile.

You can get all the information around this certification at

http://www.pmi.org/certifications/types/agile-acp

You must earn 21 PDUs (Professional Development Units) to be able to appear for this exam. You are expected to earn at least 30 PDU every three years in order to main the status of this certification.

PDU – One PDU can be earned with one hour of activity (training). As per the PMI “The professional development units(PDUs) are the measuring unit used to quantify approved learning and professional service activities.”

 

 

When to use Agile?
 

Agile – The agile means able to move quickly and rapidly.  Precisely the ability to both create and respond to change in order to fulfill the fast paced dynamic business environment refers to agility.

Agile software development refers to a set of protocols for building software under which requirements and solutions evolve through the collaborative effort of self-organizing cross-functional teams. It advocates adaptive planning, early delivery, evolutionary development and continuous improvement, and it also encourages rapid and flexible response to change.

The big question is when to use Agile. The world is moving towards Agile and one of the study shows that majority of companies would be very much Agile by 2020 and if not, it would be very difficult for them to survive. Although Agile adds value to process stream in most case while there are certain situations where it may not be apt.

While moving towards agile, one should evaluate following answers to below questions

  1. Requirement Definition – Changing or Constant
  2. Experience and Skills of a Team – Highly skilled or Newbies
  3. Change – Frequent changes or Less
  4. Resources – Dedicated or floating
  5. Timelines – Fixed or Flexible
  6. Documentation – Less or more
  7. Customer Involvement – Continuous or Intermittent
  8. Physical Location of resources – Co-located or Distributed

The mission critical applications or products probably still better be on waterfall or traditional model. Probably the 8 points mentioned above should be given a thought before taking a decision of choosing the right set of processes

Agile Manifesto and Core Principles
 

The agile manifesto says

Individuals and interactions over processes and tools

When I first time read this statement, I was little puzzled in terms of identifying the focus or intention behind this starting point. I had following question in my mind

  • Should we give little/no preference to tools or processes. Or if we do, to what extent.
  • Why is communication is so important? Why can’t we get it done utilizing tools so called ‘Agile Tools’?
  • What does it mean by individuals?

The whole world make use of processes and tools. The tools like JIRA has gotten lots of popularity.  At a time you feel that things can just go right even though there are distributed teams or people don’t interact much. The real truth is agile focus on people, communication and internal interaction.  In order to understand the real intention just replace the word “over” with “before” and it would make better sense. One cannot replace processes and tools completely while the communication gets more priority.  The collaboration, interaction and internal trust makes self organizing teams. The lack of self organizing teams result in ‘Handicapped Agile’.

Working software over comprehensive documentation

The comparison here is with waterfall or traditional development methodology where the customer requirements are documented start to end with huge sets of artifacts. I have personally seen team spending more time in documentation over building software. It is virtually impossible to document every piece of requirement in the beginning for most projects. Moreover the sign off from customer is just a formality as customer is dreaming of working piece over what is there in requirements. The elaboration of requirements itself is complicated and further there are so many unknowns.  The Agile believes in incremental development  and continuous delivery of working software to ensure the work is being done with respect to customer expectations.

It is good idea to have limited documentations while this statement every agile developer has taken it so seriously and developed a myth there should not be any documentation to be created at all.   Having product document, user manual or specifying common issues etc. would help project to run smoother.
Customer collaboration over contract negotiation

There is a common myth what I have seen in most of the projects I have gotten into that is “We don’t need a contract with customer and we would do it as we move forward with development”. This is one of the most misunderstood statement among developers and folks getting on to Agile. We obviously need contract and while the customer involvement gets more priority. Let me give you an example to understand this better.

Assume that you want to build a connector to connect to your IAM product to customer application to perform provisioning. The product roadmap and release planning is over post feasibility. As per release planning we have divided the work into number of sprints say 2 weeks 24 sprints. At this point as we constantly showing the working product to customer, the changes are welcome. The scope creep can be managed in this case is either by removing something from sprint of similar size which is not being worked up or adding that item in future sprints (not the work in progress). This might result in adding one or two sprints to overall 24 week sprint (Possibly we can reduce that while this kind of chances are fat)

Adding one or two sprints or probably more won’t disturb customer as at the end he would have a solid product and moreover the changes are proposed by customer which could be because market dynamics have changes, customer has missed the requirements or could be anything else.
Responding to change over following a plan

The successful product matters over complicated plans. Agile is extremely adaptable and typically go with release schedules while each release comprises of one or more sprints. The water fall model simply follow the plan and any change in requirement is very tough to include and further adds lots of cost to it. The flexibility of managing scope creep makes Agile superior. The focus is always respond to change. I have personally seen 6 weeks sprints have changed to 20+ weeks while customer was really happy as he has gotten what he wanted.

Here you have principles behind Agile Manifesto

Our highest priority is to satisfy the customer through early and continuous delivery of valuable software.

There are number of reasons

  • Customer gets confidence by constantly getting the working software.
  • The early time to market adds lots of value.
  • The market dynamics helps to make changes to products.
  • More than 60%+ software projects are scrapped before they are complete. This percentage can be reduced and further if you product is not good enough you got to know well in advance and you can take a call whether you want to continue or stop it.

Welcome changing requirements, even late in development. Agile processes harness change for the customer’s competitive advantage.

The changes are always welcome. The customer gets what he wants and it would not hurt developer or company building the software.

Deliver working software frequently, from a couple of weeks to a couple of months, with a preference to the shorter timescale.

You know where you stand and the same is with respect to customer. This is probably the biggest benefit of agile.

Business people and developers must work together daily throughout the project.

The biggest reason for most of the projects failure is the team working in isolation with no clue whether they are on track. Working closely with customer constantly not only gives confidence to customer while at the same time benefits to team to be aligned well with customer with respect to their requirements.

Build projects around motivated individuals. Give them the environment and support they need, and trust them to get the job done.

The most efficient and effective method of conveying information to and within a development team is face-to-face conversation.

Working software is the primary measure of progress.

Agile processes promote sustainable development. The sponsors, developers, and users should be able to maintain a constant pace indefinitely.

Continuous attention to technical excellence and good design enhances agility.

Simplicity–the art of maximizing the amount of work not done–is essential.

The best architectures, requirements, and designs emerge from self-organizing teams.

At regular intervals, the team reflects on how to become more effective, then tunes and adjusts its behavior accordingly.

After years in Agility
 

It’s been few years since I started following agile methodology. When I look back it’s interesting and happy journey. Was thinking about the book (User stories applied) which introduced me to Agile world. Back then, I read the book like a story book. Now I re-read the book after two years after practicing Agile. It was a fantastic read! I thought of sharing few fundamental things for being in an agile team.

  1. Team > Individual : When you are part of an Agile team, the focus shifts from individuals to the team . This will be a big change for those who come from waterfall model. Although during the standup meetings people talk about their achievements for the day and the next actions, it’s not the individuals work alone that make sprints successful. Things like, taking up pending tasks, helping team members complete tasks, pair-programming, looking for the overall success of the business etc. are thing which make significant impact than an individual’s ability to write optimal code/test.
  2. Where is the Design : When I came from the waterfall model, I was surprised that there is no definite design phase. But soon I realize that design is happening in iteration over a period of time. While being in a sprint, Design happens when we do sprint planning and tasking, design happens when we do the pair programming, it happens when we do TDD. This way, there is more scope for the developers to think about code and design, than a single monolithic time slot to think through design.
  3. Agility is different from being in an agile team : Having an agile mindset is completely different from being part of an Agile team. Being part of Agile team gives us leverage to do iterative development, do the ceremonies of agile process etc. But when we have not imbibed the principles and the manifesto behind Agile, it’s not Agility.
  4. Communication : Focus and Communication are fundamental to any successful sprint team. When there is no direct communication, thoughts & opinions are not converted to actions. Be it daily standup meetings or technical discussions/pair-programming/or during sprint planning, connected and collaborative people can only create beautiful software.
  5. Practice, practice, practice : There is no testing phase, there is no business verification phase, nor there is any formal review process. How can we sustain quality ? how can we insure reliability ? The answer to these questions come from the software development practices like TDD, BDD, emergent design, CI etc. Agile provides a way to use these practices. Apart from the acceptance criteria for a story, Definition of Done(DOD) governs the acceptance criteria for all stories. When we have DOD covering these practices, it becomes a standard for ensuring quality. There techniques and practices in combination with Agile process can help to deliver continuous business value while ensuring good quality.

There are an indefinite list of advantages for doing iterative development over waterfall and vice-versa. Although Agile, helps to deliver software faster to clients, there can be projects where regular waterfall model might help. We need to be choosy in selecting our process for each and every project that we work on. Provide your inputs and thoughts on this topic in comments.

What is “Pair Programming”? Is it worth doing it?
 

Pair programming is an agile software development technique in which two programmers work together at one workstation and share the same keyboard. One person (Driver) writes the code while the other person (Navigator) reviews it and at the same time thinks about the big picture.

How to do it right

  • Understand the requirement well before you start. Spend few minutes and discuss with each other.
  • Agree on one small goal at a time.
  • Support each other
    • If you are a driver, focus on small tasks and quickly complete it avoid bigger issues.  Trust navigator to your safety gate.
    • If you are a navigator, constantly review the code and think of a big picture. You don’t need to dictate the code.
  • Celebrate victory when the task is completed or you resolve a problem.
  • Changing roles few times in a day helps (Driver to Navigator and vice versa)

Do’s

  • Encourage pairing. Do not worry about slight productivity loss in beginning. This is something I have seen many teams when new teams are formed as they don’t have experience in the pairing.
  • Start with a trial. Do not force individuals to pair. Let them decide who wants to pair with whom.  Changing pairs constantly helps.
  • Pairing for more than six hours a day is not advisable.
  • Individuals should switch driver and navigator role when they get bored.
  • The Large monitors and good leg room are essential along with co-location which is absolutely mandatory.
  • Trust and support each other.  The team culture plays a critical role.

Benefits

  • Improves software quality without impacting time to deliver.
  • The focus and energy involved are much higher hence chances of making mistakes reduces significantly.
  • The better articulation of complexities and hidden details of coding tasks reduces the human errors.
  • The requirement elaboration and definition of done usually be better understood.
  • Build trust and help individuals to be better skilled.
  • The partners can switch when frustrated or stuck. The work doesn’t get compromised. The change in responsibilities once again adds energy to work.
  • New recruits come up to speed more rapidly in a pairing environment.
  • From a developer standpoint, pairing is enjoyable and valuable activity. I have seen developers who resisted pairing the first time, eventually loved it and found it to be much really useful in terms of learning, more engagement, better quality and successful careers.

Challenges

  • It’s social skill and it takes a time to do it. The best pair programmers exactly know when to say let’s try your idea first. Do not expect the outcome of pair programming from day one. It takes some time to do it in right way.
  • No benefits are expected if both the programmers are not actively engaging themselves. It’s “programming it loud’ methodology” hence it is essential that the driver and navigator constantly communicating. The silence kills the benefits of pairing.
  • If the two people have personal challenges, the pairing cannot be forced upon. The trust and mutual understanding between the two people is absolutely necessary
  • Experience mismatch is another bigger challenge. The senior programmers often want to have more control and give a little room to junior programmers.
  • View pairing as one person watching, the other person doing the actual work. That becomes boring and disengages the person watching, eliminating any real benefit from this practice.
  • Pairing should be avoided for very simple tasks or tasks which are very clear and can be done in little time.
  • Pairing needs to be done by two. The moment the third person added to it,  it cannot be called pairing anymore.
  • The co-location is absolutely mandatory. The absence of co-location where the work is shared between two programmers is called sharing over pairing.
  • Force the pairs or identify them ahead of time which may not be right in many scenarios. The best approach is to let the pairs form and swap by their own.

Myths

  • The pair programming is mentoring while it should never be. Programmers often pair with somebody with the goal to learn technology and domain. That is called either as mentoring or knowledge transfer. This should be treated as knowledge transition and expecting less than one person productivity is a fair expectation.
  • Two people working on the same story but different tasks individually is considered pairing while IT IS NOT! It is sharing the work but not pairing.
  • I have often seen individuals complete half the task and hand that over to another team member in evening sitting in a different country and assume that it’s pairing. This is again sharing of work and comes with extra cost due to unknowns, handoff and understanding each other’s work.
  • Doing agile requires pair programming. The reality is Agile manifesto never talks about pair programming.
  • The initial resistance of programmers that pairing is not a right thing to do. Most programmers like it when they try it.  Others don’t do it right and start believing it’s a waste.
  • The pairing would reduce the productivity to half. This is one of the most debated topics.  I have personally experienced that when it is done in right way, it improves the overall productivity of more than two people working individually.