On "Quality"

Just a reminder to everyone making suggestions:

The definition of “quality” is conformance to [user] requirements.

–P.B. Crosby

The metric by which to measure the value of a change, a feature, a rule, or an option is simple: Is this something of value to the users?

If it is, users will come to you. If it’s not, you’re wasting your time and hurting the project.

If you think your idea is so amazingly great, and it absolutely must be implemented, and all the complaints are wrong, because once they use it they’ll understand how great it is… I have one word for you:



Yes, very good point.

1 Like

Clearly you’ve never owned a pet rock before.


“Quality” in software starts at the level of individual code changes. In this context it means “does it work” and “does it break anything”. I’ll be writing more about this in an upcoming blog post.

That’s just one perspective though. More directly related to your point:

The metric by which to measure the value of a change, a feature, a rule, or an option is simple: Is this something of value to the users?

This is part of the reason for our petitions process: it allows user requests to determine the broad direction of ClassicPress rather directly. Looking at the top few petitions, we see a clear theme that users want more performance and less bloat:

It’s not a coincidence that this is going to be our focus for v2.


Clearly you don’t understand the premise.

Pet rocks gave the customers what they wanted.

The 70’s was an interesting time which gave rise to odd products like the pet rock and the chia pet. But… that’s because the market wanted these products.

They were “quality” products because they met the requirements of the customers. Those requirements were mostly based on novelty, frivolity, and fun. People bought pet rocks because they were stupid and useless. We’d just come out of (our loss in) the Vietnam War, and we needed a distraction. Pet rocks were so amazingly stupid that they gained attention–and market share–because they distracted us from reality.

They provided a quality product for the intended market. Everyone starting in marketing dreams of creating the next “pet rock”.

TMI. Just a joke.




Quality for me is not giving customer what he wants. So I can say that since people “want” Tamagotchi, Tamagotchi Is a quality product. WTF?!?!?
Quality is about finding the sweet spot by balancing what customer needs with what he wants, and offering the best service possible with the means, skills and resources I have available.
Just my two cents.
Oh by the way, Tamagotchi Is a Digital Toy involving a person to take care of a virtual pet. Pet is going to diw if unattended… (Don’t know if all you folks know them, here in Italy that really stupid thing was a hit among youngsters and oldsters, everyone wanted to have one for the sake of demonstrating how responsible they were keeping it happily alive).


Good points all around.

I’d like to add that quality is much more than meeting requirements and providing value – consistency and durability surely can’t be overlooked. There are other facets, too. Consider many of the plugins found in the WordPress space – they meet user requirements and provide value, but, once you start digging around under the hood, you realize that the code quality is often limited by a number of factors.

My point is: the term quality is actually referring to level of quality, and the level of quality will be a direct result of how the software comes together on a wide variety of fronts that most end users never notice. The following page offers a palatable summary and describes the standard I’m working toward in Code Potent’s offerings.



Define “best possible service”–that’s just another word for “quality”.

If the customer wants you to give them a product and never be seen again, then anything additional you do is “bad service” (it’s annoying).

And deciding what the customer needs is a minefield (Remember: WordPress decided that users need Gutenberg) which often leads to bloat. The US automotive industry is a perfect example of this.


What I mean is if the customer asks me to set up a newsletter, I should do the job as best as I can with my skill set and resources, I must present him with at least three solutions to choose from with detailed explanation of what they do in order for him to choose.
Sometimes customers arrive with a “blurry” request, for example, they want to improve their site frontend but aren’t detailing what needs to be changed. That is where I can ask them what they need the site to do and balance it with what they want - they may want a stellar branding as they have seen on another site, but this branding could be the wrong move for them, it’s up to me integrate this want of theirs with the need for the site to obtain certain results. And again is up to me to perform as better as possible. this along with what @Code_Potent has outlined in his last post is quality.
Just giving customers what they ask isn’t quality at all. Again, if you as a customer want something useless and stupid like a Tamagotchi, I can’t call the Tamagotchi “quality”. Tamagotchi is performance, that is responding plainly to a request without giving any value.


You’re making assumptions. I’ve worked with a lot of companies doing newsletters, websites, etc. In many cases your method would be wasted effort and frustrating to the client. They want someone to “just do it”. They don’t want choices, they don’t want detailed explanations of each choice. They want an “expert” to make the decisions for them and know that they’re the right ones. A “quality” marketer–for them–is one who steps up, makes the choices, and lets them work on their own business. They would see your way as “poor quality service”.

You’re falling into the trap of thinking that your way is the right way for everyone. You’re confusing “quality” with “best it can be”.

Let me give a real-world example. Razors.

For personal use, I “require” a razor that has a floating cartridge, an open back, a textured grip, a nice heft, and a reasonably-long life. That’s quality for me.

Tattoo shops and surgeries “require” a cheap, single-use razor that gets rid of most of the hair, and has a failure rate of 10% or less. That’s “quality” for them.

It’s giving plenty of value. It’s giving entertainment at a very low price. That’s valuable to lots of people (as shown by the fact that they bought the toys).

NB: “What the customer asks for” is not the same as “what the customer wants or needs”. Don’t confuse the two.


Well, there are customers who want me to step up. But the majority I have do ask me options. Usually I give options, and point out my suggestion.
What I mean is “quality” is how “perfectly” you do the job. On the newsletter example quality can range from me installing a very poor solution to finding the right one for them. Instead Tamagotchi Is performance of an order. It has no level. There aren’t sophisticated ones and cheap, poor ones. There is just a single thing. So I call it performance of a task. Quality happens when someone decides to build a better Tamagotchi, maybe a metal one, unbreakable. Or one that is water-proof. In an attempt to perform better. But since it’s a cheap performance, there is no need to better perform. So “quality” value is less important than “price”.
What customer asks, wants and needs are three different things. But in a blog example… I always supply a basic skeleton (home, about, services, newsletter/mailing list). Not all customers “ask” for mailing list feature because “they want a way to have customers contact details, but they have no clue on how to do this” and I supply it because I know they may “need” it. That for me is finding a sweet spot among ask/want/need.
In some cases I choose for them, if they are the kind of customer you describe. In other cases I discuss the solution with them. (Majority of my customers are totally able to build the site themselves, but for whatever reason they prefer hiring someone to do the job. They want me to build something I can manage/maintain but also they want to retain control by having the final word on everything. They want options to evaluate, and they want to decide. Sometimes they listen to me, sometime not).

Anyone trying to define Quality needs to read Zen and the Art of Motorcycle Maintenance - an entire autobiographical novel dedicated to defining Quality.



@anon95694377 thanks for the suggestion :smile:.
I think there are so many nuances to this word that sometimes it’s not even possible to translate it properly.


The translation is in the OP: Quality is conformance to [customer] requirements.

I’ve been teaching this lesson for going on two decades. “Quality” is a meaningless word. It is 100% subjective. It’s even a common colloquialism: “One man’s trash is another man’s treasure”.

Any discussion of “quality” should be promptly tossed in the trash. When those creating/offering/selling the product focus on “quality” they invariably focus on “more” and “more complex”.

We need to focus on what the clients (users) want and need. Not what we think is best for them, but what they actually want and need.

This comes from asking questions, listening to feedback, looking at data, and accepting what all of that says.

What we think is “quality” (I’ve seen more than one person here insist that “quality must include X”) is irrelevant. The only definition of “quality” that matters is that of the users–because if we don’t give them what they want and need, we won’t have them. Make zero assumptions about what users want. Ask them. Then listen. Then give them what they want and need.

ClassicPress was founded because WordPress stopped listening to the users and decided that their definition of quality was the right one.

Let’s not repeat that mistake.


This is where you run into trouble, because each user’s wants and needs is different, and your choice is to either provide a generic solution that meets the majority or something quite complicated that can match each user (but at the cost of the user having to understand it all to be able to select it). This is where the WP philosophy of “Decisions, not options.” comes in. The software makes the decisions that the user doesn’t want to have to be educated about in order to make. People often mistake this for “don’t give the user options”, but it’s really about making sure that the options that are given are clear, and the average user has enough information to make the choice easily.

There are many things that a user doesn’t know enough about to be able to answer “What do you want and need?” because of the ramifications behind it. Example is newsletters: they may want to send emails to people, but don’t realize that a web server is a poor choice for that. Or they want a certain permalink structure, but don’t realize what that does to their SEO. Or they want images side by side on their page, and don’t realize it will look very different on a phone.
So listening to what they want is great, but it doesn’t affect quality.
The quality of a web site is subjective, but most would agree that it is poor quality when the content is not organized, the color contrast of the text is low, the page takes a long time to load, the links don’t work, there are overlapping elements, etc. These could all be due to the theme (I was reviewing one like this just 2 weeks ago). The quality of the sftware that produced the web site is also subjective (and made of many parts). It’s not so much about what you see (that’s the theme), but about the ease of use, security, speed for the users, but also the documentation, ease of extension, organization, consistency for the developers.


That philosophy was never implemented. The philosophy was always “growth at any cost” which leads to emphasis on backward compatibility which leads to nothing ever truly gets deprecated or just changed in a major way in order to not upset people too much.

In a project that truly believes in “decisions nor options”, the “classic” editor would have been deprecated with the introduction of gutenberg. Instead a lip service was paid to the mantra by forcing people to install a plugin instead of just having an additional setting somewhere. The cost? 5M+ annoyed users.

The true mantra WordPress actually operates by is probably more like “Bad decisions, not options”


Well, I don’t follow your logic there, and it’s not about quality.
But the “Decisions, not options.” philosophy has been cited to me several times in the past 2 years as a justification for a solution. It was from the mistaken notion that the user should not have options, though, instead of from the view that the user does not have enough knowledge or information to make the choice.
And you could say that that philosophy was actually behind the editor decision: WP devs decided so that users didn’t have to, or that users didn’t know they “needed” this new editor. (A lot of people love it, not sure why.)

1 Like

“Decisions not options” is just not about quality at all, it is about being opinionated. You can find positive sides in being opinionated and not, but quality is not an obvious ramification of following one of those schools of thought.

WordPress out of the box is very far from being a quality product. This is shown by the fact that no site has less then 10 plugins, with 5 of them used to disable core functionality. If we take WC out of the discussion as it is a product by itself, every plugin which has more than 1M active installs indicates that core is missing the relevant functionality that users need.

Of course, the first step to reach quality is to define what does it mean. For WordPress core it means number of installs, for site owners, it means ROI, for developers, it means ease of development, for people giving paid support it means tools and means to give it.