Last week, we implemented a new feature in a database that is expected to save a metric boatload of time. It uses an API (a software intermediary that sends data between two systems) to verify that an organization is officially a nonprofit, according to the IRS database systems. Prior to this feature, staff manually pushed a button to retrieve the data. Now, the database uses scheduled code and “triggers” to run itself on a schedule or in particular situations, like when an application is submitted. That equals a lot less button pushing, which is a win! It took about 100 hours of work to build it and more than a year of passive effort (for some of that time, it was an ignored back-burner project, whoops).

How long will it take until we gain back those 100 hours? Who knows! Can we really say that we are “saving time?” For me, the jury is out.

But is it still worth it? We think so! We want tedious processes like this to be automated so that we can reduce “human error” and put our brains to other things. Or perhaps just work less? Nah, I take it back.

hot take alert !!!!!

In my career as a Problem Solver/Spreadsheet Whisperer, I try my hardest to mythbust the sneaky premise that having the “perfect” system will “save you time” so that you “no longer have to think/do” “any” administrative/operationsy work. I think this is a false promise and a dangerous one at that.

One way that I undermine this story is by writing the history of movement operations during the 1950s and 1960s American Civil Rights Movement. Turns out that Dr. King and his crew spent a heck of a lot of time on operations (at least according to FBI wiretaps), but those stories typically don’t make it to the social movement story mill.

Sophisticated database systems don’t save time. There, I said it! Instead, we replace that time – or more time – to do sophisticated processes which are necessary for our missions.

Why? I’ve said it before and I’ll say it again – administrative work is part of our mission, not a distraction from it. Donor acknowledgements, event registrations, appeals, impact measurements, newsletter segments, nametags (with correct name and pronoun, of course), logistics, contracts, accounting… these are ESSENTIAL functions of any organization, and when they are well resourced and running smoothly, they dovetail and enhance program offerings, whether that is adopting puppies or winning political campaigns.

jevons’ paradox

In the mid 19th century, British economist William Stanley Jevons wrote a book about burning coal. He basically argued that as the availability of coal increased and the price decreased, people’s energy use would increase. For many people (then and now) this was counterintuitive. As the cost goes down, why don’t people use the same amount of coal and pocket the difference? Or a bit more coal up to their prior fuel budget? In fact, many studies since then show that people not only “replace” their saved energy cost with extra energy but they actually EXCEED their prior energy cost and super-exceed their prior energy use. In one example, studying the usage of lighting as lightbulb technology has improved:

Our efficiency gains haven’t reduced the energy we expend on illumination or shrunk our energy consumption over all. On the contrary, we now generate light so extravagantly that darkness itself is spoken of as an endangered natural resource.

David Owen, The Efficiency Dilemma, The New Yorker – Dec 12, 2010

This presents a serious dilemma in the field of energy efficiency. For example, how do we square the reality that as cars burn fuel more efficiently, most people just drive more? What if making energy efficient cars costs more energy, too? This line of thinking quickly leads me to nihilism… good thing there are some smart economists and activists who are working on these questions and reject this notion as a false binary. (Thank Gd for queer ancestors who teach me that all binaries are false anyway!)

I’m no economist, not even of the armchair variety. And I haven’t read any books by/about Jevons. I’m worried about the implications of this dynamic under capitalism and what to do about it. I wish I had something smarter to say, but whether energy prices should go up or down is outside of my area of expertise. Truth be told, I’m not even sure if this theory can be applied outside of macro consumer behavior, but I’m going to try it out. Because it’s my blog dammit! And I’ve been holding onto these questions for years. I didn’t know there was a name for it until yesterday!

(That’s my giant disclaimer for this blogpost… now that that’s out of the way, let’s keep moving.)

what does this mean for databases?

With the “fourth industrial revolution” (is this term used outside of the Salesforce-sphere?) I notice an imperative to automate-all-of-the-things, and yet people are working longer than ever. It seems like the time we “save” never gets used for leisure, but rather for more production. I have some big questions/critique of this, but that probably belongs in another blog post. (If you have them too, leave me a comment!)

Because the reality is that I think many impact-first organizations have an administrative debt that that is overdue.

We gotta fix that imbalance before we can pocket the saved time! And that might be an “asymptote” anyway! We might never reach the point of being “done” or having time left over. I have to be ok with that, because change is constant and I feel that we have an ethical responsibility to use appropriate technology in mission-driven work.

My experience shows me that as the hard stuff gets easier (and our capacity increases), new hard stuff emerges! And that’s not just novelty stuff, it’s real, mission-aligned, useful/critical things.

Here’s a real life example: Remember all those one-off surveys we used to do – they sit in separate Google Sheets in a drive that no one looks at. Let’s hook them up with the database so that we can measure change over time and remind people who haven’t filled it out yet. Sure, that’s going to be a lot of work, but it’s also going to be useful! We can do that now, because we have a reliable list of contacts in the system. More work, more impact, more potential! No time saved.

And another example: Now that we sorted out our donation classification system, it’s time to deal with the “junk drawer” of mailing lists in the email-blaster-platform (think, platforms like Constant Contact, Mailchimp, etc). So many organizations that I support upload new lists every time, leading to a lot of clutter (and sometimes extra charges) not to mention wasted time and inability (in some cases) to respect contact mailing preferences. Yikes! Once again, more work, more impact, more potential! No time saved…

what does this mean for nonprofits?

I’m slowly coming to the conclusion that anyone peddling a database or system that exclusively saves you time time is offering nothing but snake oil.

I think we should be asking ourselves, is saving time on administration the ultimate goal? My personal response is a clear NO WAY! My “north star” is solid processes + spending time on enhancements that support our mission + continuous listening/improving our tech. That might take more time, or less time, but either way, it’s the right thing to do.

I think we need to be honest that having a useful database system (and surrounding culture) is going to be a resource intensive process. It’s going to use time and attention, rather than save it. I think that’s a good thing! Yes, we have limited capacity, and yes, this area merits a portion of that energy.

In many cases, implementing a shiny, new platform will not only cost upfront $$ (implementation, data moving, training, change management), but also on-going costs (release management, automation, third party applications, moving more processes into a centralized system, professional development etc). Many, many orgs end up creating an entirely new position for their staff System Administrator. While I think this is a positive development, I don’t see how, on one hand, we hawk time saving, and on the other hand, we need to hire a brand, new, expensive position. Perhaps we are shifting the work from a decentralized to a centralized philosophy. I think there are many merits to doing it this way, but we should be transparent about WHAT THAT MEANS!

Especially (and I’ve seen this happen time after time) – when we start being able to track more sophisticated data about our programs/donors/what-have-you, we end up with MORE TIME spent on data collection (not less!) but also more/better data collected at the end of the day!

So. If that’s a win or not likely depends on your personal opinion and the relative import of whatever data process occurred. But by no means would any reasonable person describe this as a time saver!

what does this mean for nonprofit database practitioners?

Friends, we are facing an uphill battle but I think we are up to the challenge. In my assessment, we have technology debt, limited resources, and technophobia/impostor syndrome to overcome. Luckily, we ALSO have powerful tools, powerful community, and a powerful story to tell about what becomes possible when we commit to enable our systems (and our orgs!) to reach their full potential.

With these three steps, I believe we can garner the benefits of Jevons’ paradox (save time and put that “extra time” to good work!) while reducing the harms (runaway, unquestioned use of limited resources).


    We need to have truthful conversations about the complexity of various database or database-adjacent projects! Instead of promising to “save time,” let’s resolve to tell the truth about who’s time is saved versus who’s time is utilized.

    When managing a complex database system, I think there are lots of temptations to put time into projects that are perhaps better left un-done. It’s important to ask how the given project aligns with the mission/needs of the organization or if we’re doing it because (1) one person prefers it that way, or (2) it is cool/shiny.

    I’m generally a good sleeper, but anxiety about the nonprofit industrial complex can genuinely keep me up at night. If we let database-tinkering stay as a siloed, “overhead” activity rather than a mission-essential, all-hands effort, then we risk relegating database management to the heavily mythologized “nonprofit admin bloat” phenomenon. We also end up with worse systems because they are farther away from the staff, donors, volunteers, program participants, clients, and more who make the database come to life.

what do you think?

Are your systems saving you time? If yes, I’d love to hear about it and would be keen to publish a rebuttal to this post 🙂

Is there another thinker who has a better explanation than Jevons? I admit my ignorance. This is your open invitation help me develop these ideas!

Are you worried about nonprofit overhead? Do you have ideas about how to right-size and justify the cost of good systems for resource-constrained nonprofits? I want to hear your ideas! (FWIW, I think this is an overblown critique when many/most nonprofits are running on a shoestring).

One thought on “on jevons’ paradox

Leave a Reply