In the first two installments of best practices for postmortems, I touched on setting up a postmortem, Part I, and what goes into a postmortem, Part II. Today, I'll wrap up the series with closing thoughts on things to avoid and valuable next steps.  

Items to avoid during a postmortem

There are some actions that do not help us improve on our processes and should be avoided in a postmortem:

1. Do not call out individuals. The key is to identify what went wrong, not who caused the issue. Try and be as solution-oriented as possible.  
2. Do not go over the budgets in front of the whole team. If you feel it's necessary, cover schedules and talk about estimates, but we don't need to talk numbers with the entire team. Usually when we go over budget, it was from a larger planning issue that involved management.

What comes out of postmortem?

You must have a list of core action items after every postmortem. Otherwise the practice wasn't useful. Sometimes as an account service or project manager employee, you may not have a good idea for a solution to a technical problem. For these items, send out tasks to team leads in the other departments to help you formulate the list of action items. For each item, if possible to address immediately, have an owner and a timeline for completion. Other action items not immediately addressed will be reviewed by the management team and assigned appropriately. 

In the previous article on postmortems, we covered the basics of a postmortem: when it occurs, why it should be done, and who should attend. In this article I will dive more deeply into what exactly goes into a postmortem.

What goes into a postmortem?
Whoever is scheduling the postmortem needs to come prepared with an agenda. This should obviously be happening for all meetings, but it's especially important for a postmortem. Many times a postmortem will go off on tangents, so as the organizer of the meeting, you need to make sure you cover all the items within the time allowed, and this will help you keep people the meeting on track. Before the meeting, send out questions to be delivered back to you, then use the those questions to customize the agenda and send it it out a couple of days before the meeting.

Start by looking at the planning phase of the project. Items like:

• Initial conception (maybe even compare to final deliverable)
• Risks that were identified (or not) up front
• Adherence to the schedule
• Accuracy of estimates

Address the process throughout the project. Cover items such as:

• Were tasks coordinated and sequenced correctly?
• Did we miss any steps in the process?
• Were tasks with a higher priority identified?
• Did we find that certain steps in the process took longer than originally thought? Did this cause the project to slip?
• Were JIRAs assigned for tasks on the project? Was the information in them detailed and accurate? (Due dates, detailed instructions, commenting, etc)
• Was the project schedule kept up to date throughout the project?
• Did we address out-of-scope items appropriately?

Cover communication within as well as outside the internal team.

• Did the group hold weekly status meetings?
• Did the different groups respect each other’s expertise?
• Were there difficulties coming to a consensus on key decisions?
• Were there any issues communicating with third party groups or the client?
• Were stakeholders kept in the loop regarding changes to features?
• Did the group receive regular minutes after meetings?
• Were the goals of the project communicated clearly to the team?
• Were team members accessible to everyone to help answer questions?

If necessary, talk about team members and their roles.

• Was your role clearly defined on the project?
• Was it clear who had decision-making power at every level?
• Did the PM have someone to perform all the functions necessary for a successful project?

In part three of Successful Postmortems I will go through the items you should avoid in a postmortem and more importantly, the action items that you need to define, assign, and execute on.

Here are my visual notes of Nadya Direkova's talk, "Game On: Design Patterns For User Engagement."  

See these notes bigger →

Springbox had the privilege to work with the Visual Arts Center (VAC) at the University of Texas at Austin, developing a new brand identity for the institution. I was on the Springbox team that created the brand identity and recently attended the VAC's grand opening on Friday, October 24th.

The new VAC gallery is located on the southeast corner of the Arts Department. Approaching the gallery, a series of steps takes you up to the entrance. This was the first chance to catch a striking glimpse of the work we created. The front side of every step is painted such that while you look at it from the base, it creates the VAC logo, which we designed.

As you walk in, the gallery opens into a large area where the artist in residence displays their work. The current piece titled ZZZ’s, by the Los Angeles-based artist Ry Rocklen, is a collection of found objects placed around a sequence of tiled carpets, which expands up toward the ceiling. It’s the kind of piece you can immerse yourself in and get a feel from all angles. Ry goes into more detail with VAC Director Jade Walker in explaining the concepts behind the sculpture.

Walking further back on the first floor is an exhibition titled Unveiled, which includes pieces by University of Texas students and alumni. This exhibition's theme centers around transparency, clearly noted by Self-Portrait, a column of neatly folded and stacked t-shirts belonging to the artist Arturo Agüero, reaching around 20 feet high and ordered in a sequence that flows through an entire color spectrum. Upstairs another group exhibition titled Deconstruct and Reconstruction included several visiting artists’ pieces, including sculpture, video installations and paintings.

The gallery as a whole is a large space with many large areas designed to give the pieces the breadth and attention they deserve, allow visitors to immerse themselves in the work, and, most exciting for me, our branding work is presented throughout the space on all types of media. I won’t lie when I say it gives me a warm fuzzy feeling inside to know that we were a part of this new and exciting space for intersecting ideas in art.

Prologue: Tom attended the Flash in The Can 2010 event in San Francisco and has documented his experience. The opinions and/or views expressed in this series of posts are his own and do not necessarily reflect the views of Springbox. 

Last day of the conference. The first speaker for the day is Robert Reinhardt, who will be talking about Flash video and where it all fits in with HTML5 and Silverlight. Turns out, Matt and I have talked to this guy a few times. He’s written many books about Flash, including the Flash Bible, and is super-friendly and approachable. His talk was refreshing because it proves how far behind other technologies are when it comes to video. He explained how we need to help our clients understand who their audience is and what they want. I couldn’t agree more. 

We need to educate our clients to not think so much about where they will deploy their application and how it will be created. Instead, they should think about the problems they want to solve and who their audience is.  For example, Reinhardt had a client who demanded that an application work on the iPhone. Why? Because she had an iPhone. After analyzing site analytics and looking at a breakdown of visitors, it was revealed that less than five percent of Reinhardt’s client’s audience used iPhones. In the end, it wasn’t cost effective to build the same experience to work on the iPhone. However, what can be cost-effective is to tailor your mobile experience to fit the needs of specific users. For instance, the aforementioned client had varying hours of operation and a location that was difficult to find — two key items that likely concerned her mobile audience. In the end, the client’s mobile execution highlighted the company’s hours, provided its location and included a few specifics about products. Reinhardt’s client was able to address the needs of the mobile customer while deploying a version that satisfied the needs of desktop users.  

It was time to wrap things up for FiTC, at least for us. We needed to catch an afternoon flight, so we only had time for one more presentation. Grant Skinner, a well-known leader in creating all kinds of cool flash tools, games and applications gave a talk about how he formulates ideas, the direction of his current work, and where he’s focusing his future efforts. What I gained most from his insight is doing the most with the free time you have. We have so many distractions throughout the day, with current projects, social events, Facebook, Twitter, email, websites, videos and on and on. Most great ideas come out of projects that you’re passionate about. Make free time every day — even if it’s for only 20 minutes a day — to work on what you’re passionate about. There’s a good chance that some goofing around in your production tool could potentially lead to some really exciting client work. At Springbox, it’s easy to get lost in the day-to-day production for our clients, but if we can find that time to focus on new and innovative ideas, it may eventually lead to more quality tools for our clients. 

So what did I learn at this conference? Here are the key points: 

1. Flash is not dead. It’s thriving.
2. Mobile is King right now. Tomorrow, who knows?
3. Android supports Flash. Android is big and getting bigger by the day.
4. San Francisco is cold in the summer.
5. Passionate work will (almost) always be rewarded.
6. Adobe cares about its customers.
7. Flash nerds like to party.