How Time Estimation Saves Time, Sanity, & Your Developers

The Problem Every (New) Studio Faces

"It is a truth universally acknowledged, that a [game studio] in possession of a [project], must be [overscoped].” - Jane Austen (probably)

While not in any way unique to graduate student projects, around mid-project we’d found ourselves looking at a backlog of too many tasks with not enough resources to complete them. As the producer for the Art team, I was starting to sweat. We’d eked out a passable Vertical Slice (VS) milestone, but now needed to fill out four levels, or 'tracks', with hero pieces, custom environmental buildings, environmental mod-kits, VFX, and more.

Being a brand-new producer, and not wanting to give a hard “no” to ideas, I’d been sitting idly by, hoping that the Artists would randomly “figure it out” and get things done by the next major check-in.

A Desperate Cry

As part of the sprint retrospective we noted that our Level Designers were struggling to be patient with our “ad-hoc” method of reporting asset statuses… if we reported them at all. At the same time, we were worried about what we could accomplish in the next Alpha milestone.

The Art team and I found ourselves crying out: “Please don’t kill the Artists!”

A Simple Approach

Being a simple guy, I went with the simplest, arguably most classic, solution: an Excel sheet. The Level Design Producer had made a backlog Excel document with the Level Designers' art requests already recorded, so I made some simple time estimations to see what was possible with the time and resources we had left:

A quick note if it wasn’t already apparent, these were back-of-the-envelope calculations. I didn’t know how long it would really take each artist to make a hero asset or create a bubble effect, but that was actually fine. Why? Because even with rudimentary math, it revealed how little we’d been thinking about the realities of time and scope. We could tweak the numbers however much we liked, but the big red numbers on the top of the sheet showed we were being way too aggressive. We were nearly 18 hours over budget—more than a new Artist for the team could do in the time left! These results were rather grim. My estimations showed there was not even enough time to tackle our current list of assets. However, the Art Team and I just needed a measurable, reliable calculation to show that our intuition was correct. We finally had the evidence we needed to start making the hard cuts and to stop piling on wishful art features.

An Amazing Conclusion

Dreading the lead group's reaction to the findings, I shared the Excel sheet at the next lead meeting. To my surprise, people understood we were well and truly over-scoped and that we needed to do something about it!

But I needed to follow a few important steps in order to help people accept this result:

Several decisions came out of that meeting. Firstly, we aimed to scale down the number of hero assets art requests. Secondly, the team eventually would cut of one of the four racetracks.

Important note: I never pushed for the removal of a level, it was a decision made by the game designer after he considered both the Excel sheet and other factors. While it’s fine to mention a potential way to reduce the workload for the team you represent, it’s important to let the lead in charge weigh the options and make the final decision.

Everyone Sucks at Process, That’s Why We Do It!

Even with the removal of a large amount of asset requests and trying to balance the Excel sheet hours, the Art team still didn't finish all of their tasks—but that’s ok! The real benefit that came out of it was practicing a process that helped us better understand what was possible. Time estimation is hard, and it’s almost impossible to get right. The same is true for many planning processes, but we get better over time.

For my team, even a rudimentary time estimation helped us get a reality check on what was possible for our game—and that’s all it had to do! Process doesn’t need to be beautiful, it just needs to be effective.