Guide Writing and Cookie Cutter Builds

Yesterday there was an excellent post on Blizzard Watch written by Zoopercat, one of the creators of Ask Mr. Robot. In it, she shows through data (sims and surveys) that the community’s perception of a talent (or any other character combat customization) effects usage far more than the actual value of the talent would imply. The suggestions from cookie cutter builds, such as those found in guides, tend to present choices as black and white, and discourage customization, even when case-by-case choices are nearly equal or even more optimal. I highly recommend reading it.

Community Best

I’m not a theorycrafter; I don’t come up with mathematical proof that one choice is better than another. Rather, I prefer organizing and teaching subjective play. As a writer of guides, I’m often faced with community pressure to align my words with the accepted choices. And because of the type of non-confrontational person I am, I often cave to meet that pressure, against my greater goals of teaching understanding over insisting on a single correct choice.

For example, early in the expansion, I recommended haste in my brewmaster guide. Haste was a poor defensive stat for brewmasters, but it made play feel better and the defensive loss was actually pretty small in the scheme of things, especially at low gear levels. It was one of my many suggested builds, all offering solutions to different problems. But after many comments and complaints about how bad haste simmed, I finally removed that stat priority from my guide. Not because I believed it to be bad advice (I still give people that advice when they come to me with a resource management problem, in private), but because I got frustrated with constantly explaining my reasoning.

I have similar stories regarding Healing Elixirs, Chi Wave and Zen Sphere, the resource generation talents, and others. My ultimate goal with this guide is to encourage mastery and love of the class, but I fall short of that goal because I only have so much energy to argue in favor of customization in contrast to sims.

Basically, people love black and white answers. They like feeling right when they “choose” the correct answer, and they like feeling superior to people who choose the wrong answer. Subjective choices cause frustration. A lot of that comes from historical precedents. WoW started and spent many years in a state where “choices” were decided by objective algorithms and there really was a right and wrong answer. Some talents simply were better in every situation. But in the new talent system introduced in Mists, that old perception rarely applies, even if we try to force it. We as a community rarely distinguish between “best” and “best on average” or “best for very skilled people” or all the various reasons something might be the current favorite.

Two Audiences

Guides generally have two different audiences: “I want to jump in and play” and “I want to understand”. The vast majority of a guide’s readers will be in the former group, what I’ll call the quick’n’dirty players. They can range from new players who want to tank a 5 man with a freshly boosted monk, to experienced players who want to try a new class, to casual players who want to spend their precious few hours of free time playing a game instead of reading about it. They are interested in getting a introduction that is easy to digest and follow. Cookie cutter builds are targeted to this type of reader, and they often get frustrated when the answer to their customization questions is presented as subjective. They build the body of a readership, and are really important for establishing an audience.

The latter group, the “I want to understand” people, or what I’ll call the comprehensive players, are more interested in why these choices were suggested and how to master the spec they are already familiar with. These are a smaller group that usually starts out as a the quick’n’dirty type, but eventually want to know more. They may want help solving a particular problem, or they want to customize their character to their liking, or they simply want to be better at the game. They want explanations for everything, beyond the “do this”. Cookie cutter builds can actually harm this group because cookie cutter builds don’t offer nuance. A talent row that is actually competitive might come across as inflexible because the majority of cookie cutter builds suggest a single talent. Brewmaster healing talents and resource talents often hit this wall, because each row has a popular choice (Chi Wave and Ascension, currently), but there are good reasons to change that choice when need arises.

You might recognize the “quick’n’dirty” label as the same I use for my raid tanking guides. Those guides simplify everything down to an extreme level to especially cater to people who don’t want to read an essay for every boss (that’s me!). They’re my most popular article on the site. But they have an advantage in that no one goes into them expecting a complicated explanation for everything. However, that is the regular expectation for my class guide.

My class guide needs to satisfy the needs of both the quick’n’dirty players and the comprehensive players. I can’t do two guides, one in overview and one in detail, like I do with boss guides, because the difference is subtle enough that the average reader might be confused where to start. When looking for a boss guide, you know whether you’re attempting normal for the first time or where you’re pulling mythic, so a guide can label itself appropriately and you’ll read the right one. But a class guide can’t clearly differentiate between quick’n’dirty and comprehensive because you as a player might not know which one applies to you just by reading the title.


Many guides solve the problem of picking an audience by catering only to quick’n’dirty readers. That’s where you get the most bang for your buck. Like I said before, these readers vastly outnumber comprehensive readers, and you only have so much time to spend on a guide that you don’t make money off of. And if you do make money, getting the most eyeballs possible is even more important.

I have tried to cater to the comprehensive players in the past, but that was often confusing to quick’n’dirty players and led to a lot of frustration on both my part and the many people who contacted me asking for more specific guidelines. I tried having two guides, a beginner’s guide and a detailed guide, but a beginner is a little different than a quick’n’dirty reader and very few people classify themselves as beginners (it’s a loaded word).

Currently my guides start with a suggestion at the top and offer explanations below, which I think is the best compromise, but value is still lost in the compromise. People still might easily assume that suggestions are law, and the existence of cookie cutter builds still discourages experimentation, as explained in the article I linked at the beginning.

In light of these thoughts, I experimented with my talent page to perhaps encourage thinking along the lines of “best for this situation” instead of just “best”. For now, I still begin with the “Quick and Dirty Talent Build” for people who want just that.


And then I quickly go into what I hope is still an easy way to evaluate talents even for a quick’n’dirty reader.

Screen Shot 2015-12-10 at 1.22.11 PM

Note the tags, the “Easiest” and the “Most Useful on Average”, etc. My hope is that starting with the why, and wording that “why” in the tersest of language, people will identify their needs and pick talents to match them. I want to avoid arbitrarily defining what is “best” (usually “Most Useful on Average”).

I’ve been trying to think of other ways to display information normally catered to comprehensive readers in a way that’s easily digestible for quick’n’dirty players. In some cases, that can be accomplished by wording things succinctly, or emphasizing a succinct reason, like through boldface. As talents progress, especially in Legion where most talents are hoping to be interesting choices, I want to see if I can encourage thoughtful choices while still being easily accessible.