Breaking Down Offline Editorial: Gantt Chart View (Part 2)

In part 1 of this series I describe the process of using a Gantt chart as project management tool for unscripted television post production. In this post I will zoom in on one specific part of the editorial process: offline editing. And meditate deeply on how we think about editing as an activity in the project management scene of the word.

In project management terminology:

An activity is a component of work performed during the course of a project. Activities take time and consume resources; you describe them by using action verbs. Examples of activities are design report and conduct survey.

This is tricky, because your activity is someone else’s project. For example: Change Car’s Oil might be an activity to you; drop car off at the mechanic and pick it up 30 minutes later. But from your mechanic’s point-of-view Change Car’s Oil is a series of small activities that require unique resources and multiple stopping points.

There are some flaws with Car Oil example, but you can also think about something like renovating a house. As a homeowner you might see macro activities like: renovate bathroom, rennovate kitchen. But your General Contractor sees hundreds of smaller activities that need to be planned in coordination with each other. As a homeowner you hire a GC so you don’t have to worry about “the details,” but knowing the details will help you understand the cause of delays when they eventually crop up.

The Post Supervisor faces a similar dilemma

One way of looking at the post schedule is macro: “The team has six weeks from start to rough cut.” Therefore, you’d see this component of offline editorial as one distinct activity. If you are a Post Supervisor you probably see your schedule this way.

An editor sees something entirely different. They see hundreds of smaller activities. Even something like “cut act one” is broken down into much smaller activities: cut scene 1, cut scene 2, cut scene 3, cut transitional segment between scene 1 & 2, cut transitional segment between scenes 2 & 3, cut bump out, etc etc etc…

This difference of perspective isn’t a problem, until you have to understand and explain why a cut is late.

If you only see the macro picture, you can know that a cut is late. But you won’t have the information you need to troubleshoot the cause. Even if you breakdown your activities into something more ganular, like acts, you still won’t have the information you need for a diagnosis.

How Changing a Car’s Oil is Different from Editing

At this point it might be logical to think that the answer is breaking down offline editorial into more distinct activities; the “cut scene 3, cut transitional segment between scene 1 & 2” from above. This is how a general contractor and mechanic does it. But editing is different for two reasons.

Firstly, the order of activities is usually not important. An Editor can work on a scene in Act 3, then score Act 2, and finish the day assembling a transitional montage. Without critical dependencies, a Post Supervisor can go mad trying to track all of the editor’s activities each day.

Second, more importantly, the creative process is messy by its nature. Perhaps you’ve seen this inspirational meme around:

Creativity doesn’t happen in a strict linear fashion. It requires exploration and false starts. The truth is that every creative endeavor, even network television, will never be finished, only abandoned. More often than not, post production is an exercise of accomplishing as much as possible in the allotted time.

The point of applying project management techniques to television production is to maximize the amount of time your team is doing the fun creative work instead of waiting for resources to become available. It’s not to limit creativity, but to unleash it!

Posting an Unscripted Show: Gantt Chart View (Part 1)

The Gantt chart is a tool used by project managers to illustrate a project’s schedule. It features a list of activities vertically along the lefthand-side. The horizontal axis is time: bars represent the amount of time to complete an activity, and the total timeline is the amount of time required to complete the project. Most importantly, a Gantt chart shows dependencies, which are critical to understanding the order of operations for a project’s activities to reach completion. I used the excellent tool OmniPlan and built a Gantt chart of one episode of an unscripted show’s post production workflow. (You can see the chart in all its 5798 x 1018 glory here.)

We’re not building a 747, but an unscripted television show is still a complex project with a disconcerting number of procedural requirements (i.e. Network Notes) that can delay a project indefinitely. And multiple hard logic processes that require the work of a skilled technician. Let’s look at each of them.

Procedural Requirements

Procedural Requirements

These green dots indicate a milestone, in our case,  when a ‘cut’ is handed off to the network. At these points the work to be completed transitions from the production company to the network. Procedural requirements are activities that are required to happen in a certain order and can’t be skipped; for example you can’t work towards a fine cut until you’ve received rough cut notes from the network.

While waiting for network notes the production company has two choices; reallocate their resources to other work (example: have your editor work on another episode) or, put their resources on hold until the network sends their notes. Neither of these options is ideal.

Having an editor jump onto an unfamiliar episode is a violation of Brook’s Law and may actually cause more problems such as stylistic inconsistencies. Some production companies take this one step further and have their editors work on an entirely different show while waiting for the network to turnaround notes. Editors typically dislike this practice. And not every production company has enough work to move editors between projects.

On the other hand, putting your resources on hold may not be possible either. Many post houses won’t discount an edit suite rental if it goes dark for 2 or 3 days. And most editors aren’t ok with having 3 random days off without pay. So often the production company is left with no choice but to pay their editors for not working.

Finally, while the production company waits for network notes, they have no power to do anything besides wait. Therefore each of those looong orange lines represents an opportunity to cause significant overages. Having a plan in place beforehand can save money and personnel problems when dealing with network notes turnaround.

Hard Logic Processes

Before a cut can be sent to the network, certain work needs to be performed. For this show an Assistant Editor needs to replace Visual Effect and Graphic elements (such as interview back plates) and insert scratch Voice Over. Afterwards the work needs to be compiled into a sequence according to network specifications and exported to file for Network review.

Hard Logic Processes

In this instance I was asked to plan for one A.E. to replace Vfx while another A.E. replaced the voice over. This is a luxury most production companies don’t have. From a project management point of view the trouble is that if one of these activities requires more than the budgeted 3 hours, the entire process is delayed. Exports often have problems and require re-exporting. These are all opportunities to further delay the project.

Keep in mind that all of this occurs immediately before the more troubling network notes procedural requirements. The high cost of a bad output is the primary reason I created the Show Exporting checklist.

Monte Carlo Simulations

One of OmniPlan’s most intriguing features is its ability run simulations which estimate the likelihood that certain milestones will be completed by deadline. In order to utilize this feature, you have to estimate the amount of effort required for each activity. This is just a fancy way of saying that you’ll need to estimate how quickly an activity will happen in the best case, worst case, and most likely scenarios. For example: inserting VO usually takes 3 hours. If there is very little VO to replace and the process happens smoothly, perhaps it could take 1 hour. However, if the VO needs to be replaced for an entire episode it might take 5 hours. These numbers 3 hours, 1 hour, and 5 hours would be the Expected, Minimum, and Maximum effort.

Using data I’ve accumulated over the years. I estimated activity effort and run my first simulation. The results weren’t promising, but not surprising:

Monte Carlo Simulation

OmniPlan has 0% confidence that the show would complete on time, but expected us to be done within 8 days. When I saw this result I went through all of my effort estimates and reduced my worst case scenario numbers. The result was the same. Overall editing an episode has too many points of delay that even when many things happen on time, there are too many opportunities to lose time. In addition, editing is one of those things that always goes up to deadline, and often over, but never under. Rhetorically: how often has a producer been given five days to address notes and they come back and to say they only need four?

Coda

Using Omni Plan to map out an unscripted show’s post production workflow illuminates three valuable lessons.

First, waiting for Network Notes is probably the activity with the most waste generating potential. Production Companies would do well to spend an appropriate amount of time planning how they are going to manage their resources (equipment and editors) while they wait for network notes.

Second, Assistant Editors are responsible for activities that can cause considerable delays (nevertheless embarrassment) if problems come up. Therefore taking the time to find and hire knowledgeable A.E.’s should be the goal of every production company.

Finally, an unscripted show’s post production workflow has complexities that haven’t generated the thought worthy of it. In my next post I will dig into how I generated the activity lists and effort estimations.