Quantity over Quality

Reading Time: 10 minutes

In this series, I’m exploring the influence of white supremacy culture on tech (and broadly, professional) culture, following Kenneth Jones and Tema Okun’s list of characteristics as a guide. Here’s the introduction and the full series so far.

Let’s talk about the fourth element of white supremacy culture: quantity over quality.

Here’s the description of Quantity Over Quality, verbatim:

Quantity over Quality

  • all resources of organization are directed toward producing measurable goals
  • things that can be measured are more highly valued than things that cannot, for example numbers of people attending a meeting, newsletter circulation, money spent are valued more than quality of relationships, democratic decision-making, ability to constructively deal with conflict
  • little or no value attached to process; if it can’t be measured, it has no value
  • discomfort with emotion and feelings
  • no understanding that when there is a conflict between content (the agenda of the meeting) and process (people’s need to be heard or engaged), process will prevail (for example, you may get through the agenda, but if you haven’t paid attention to people’s need to be heard, the decisions made at the meeting are undermined and/or disregarded)

This one, I confess, I had concerns about when I read the name. Here’s why: we have already discussed perfectionism, the first element of white supremacy culture. To me, “quantity over quality” sounds like the opposite of that.

This suggests that white supremacy culture is simultaneously too much and too little of something. This can be true, but it’s not actionable without additional details about the contexts that separate when it is too much and when it is too little. Without that, the action we have to take is to guess, on no information, which extreme we’re at and which direction to go in any given situation. Which, I’ll be frank, is not particularly helpful.

However, upon reading the description, it seems like the thing we’re talking about here isn’t “amount of something over its quality” but rather “quantitative metrics over qualitative metrics.” This is not mutually exclusive with perfectionism.

quantitative qualitative

In fact, the last point from the “Quantity over Quality” description exemplifies a team perfecting a quantitative metric of a meeting’s “success” (number of items covered out of number of items on the agenda) over addressing deep-seated issues in the team’s approach:

  • no understanding that when there is a conflict between content (the agenda of the meeting) and process (people’s need to be heard or engaged), process will prevail (for example, you may get through the agenda, but if you haven’t paid attention to people’s need to be heard, the decisions made at the meeting are undermined and/or disregarded)

I even see some defensiveness in this one: the use of the agenda to redirect the conversation away from the possibility of criticism and toward “safer”, predetermined topics. I suspect that the further we get into the elements of white supremacy culture, the more we will see them layer on top of one another.

Where Quantitative Metrics Help Us

First of all, quantitative metrics do help us. They give us one means of observing change and a shorthand for communicating about these changes. We have statistical methods for understanding what they do or don’t mean (although these are often omitted or misused, from the corporate boardroom to medical labs).

As relevant as how we measure things, I think, is what we measure. We’ve run across examples of this before. Here’s one: implicit bias trainings don’t effect change on the company culture because employee progress on inclusive behaviors isn’t measured. I made a rubric for that, and that rubric has made its way into about a dozen tech companies’ hiring and promotion practices (that I’m aware of). This is an example of a situation where cultural stagnation arose, not from the prioritization of a quantitative metric, but from the absence of a quantitative metric.

I don’t think Jones and Okun are suggesting that we shouldn’t use quantitative metrics. I do think it would be possible for white people to interpret it that way, so I felt the need to erect a guardrail here: quantitative metrics are useful, when used correctly (i.e. with statistical rigor), and when applied proportionally to our priorities, such that the absence of accountability does not artificially lower the priority of something that matters to us.

We see this reflected in the antidotes section for this element.


  • include process or quality goals in your planning;
  • make sure your organization has a values statement which expresses the ways in which you want to do your work;
  • make sure this is a living document and that people are using it in their day to day work;
  • look for ways to measure process goals (for example if you have a goal of inclusivity, think about ways you can measure whether or not you have achieved that goal);
  • learn to recognize those times when you need to get off the agenda in order to address people’s underlying concerns

In Addition to Quantitative Metrics

As much opportunity as we have to improve in our use of  quantitative metrics, we have even more opportunity to improve in our use of qualitative metrics because we rarely use them We pay less attention to them than we do to numbers. We lack statistical methods for deriving meaning from them (though I’d argue that numbers are just as susceptible to self-serving interpretations as stories or conversations are).

Where quantitative metrics can show us what has changed (or not changed), qualitative metrics help us understand why. I teach a class for computer science master’s students, and in almost every class they answer at least two surveys (I have a whole post in the works about surveys, which I’ll link here when it’s ready, but in the meantime, here’s a taste). In one of them, I ask both quantitative and qualitative questions. The results include nice pie charts demonstrating what percentage of the class would have liked more of X, less of X, or the same amount of X as we did in class. This helps me suss out what to change.

To figure out why it needs to be changed, I turn to the responses for qualitative questions like “elaborate on your experience with X” or “compare your experience with X to Y.” The answers here help me understand how to change things for the future. But it’s not the same as glancing at a number or a pie chart: I have to sit down with a cup of tea and read what my students had to say, and cross-reference that with who they are and how they learn (which I get from a survey that I send them at the beginning of the quarter) to come up with creative solutions that will work for them.

This gives us a massive clue as to why white supremacy culture favors quantitative metrics.

The fundamental distinction between quantitative and qualitative metrics is the time that we spend interpreting them.

BIG-ASS NOTA BENE: What I said was “the time that we spend interpreting them.” I did not say “the time that we should spend interpreting them.” As I mentioned before, it is frustratingly and terrifyingly common for organizations of all types to look at numbers and interpret them without applying the requisite statistical rigor. We often use numbers the way we use code review: it could be excellent if we gave it the time and effort it deserves, but since we don’t do that, it’s often pretty shite.

Unlike quantitative metrics, though, qualitative metrics don’t give us a shorthand with which to convince ourselves that we instantaneously know what they mean. Instead, we have to spend time. And white supremacy culture never has time. We’ve seen that while studying sense of urgency as a key element of white supremacy culture.

You know what else white supremacy culture doesn’t have? Attention. White people working in a white supremacist corporate environment don’t have to develop empathy stamina because the default experience is their experience. We have to actively train ourselves to listen to, and care about, other people’s experiences. You know this. We did a whole series on this.

So there are three skills we need to build to use qualitative metrics effectively:

  1. We have to practice making time. This is, in part, a function of our time management, and in part a function of our view of time—what deserves our time? What has to be done right now? What has to wait because right now we need to make time to listen, or read, or think?
  2. We have to practice putting things in context. Who is providing the qualitative data, and what informs their perspective? What unique experiences do they have, that we cannot access?
  3. We have to practice empathy stamina. Can we scrape up the humility to rely on other people’s assessments of phenomena that we don’t notice because we cannot access their experiences? The key to finding new, innovative solutions is in listening to and believing people who didn’t make the decisions that got us where we are. We can try, for sure, to imagine ourselves in their shoes. But even when we can’t successfully do that–even when our experiences are too different—we can engineer solutions from the ideas that they suggest.


We talked about how white supremacy culture favors the use of quantitative metrics over qualitative ones. Also: the same forces that drive white supremacy culture to favor quantitative metrics also drive it to use quantitative metrics incorrectly.

Quantitative metrics have value (especially when used correctly), and in some cases the problem isn’t that we measure things wrong—it’s that we don’t measure them at all, and in so doing, de-prioritize them by failing to build in any accountability to them.

That said, there is a critical role for qualitative metrics in understanding and acting on quantitative metrics. While quantitative metrics can give us the what, we turn to qualitative metrics for the why, which informs how to effect the changes that we want. This process, though, requires us to make time, and slow down, and listen, and understand experiences that are different from ours, and believe them even when we cannot identify with them. No matter how racist we think we aren’t, or don’t want to be, we need to build those skills to enact anti-racist data-driven decision making.

If you liked this piece, you might also like:

The rest of the posts in the (brand new!) inclusion category

The series about reducing job interview anxiety (especially for folks with a little experience)

The cost-effectiveness of pair programming (for folks who feel strongly drawn to working with others!)

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.