Our monthly office hours for paid members is today! Join us at 1p Central Time to chat about all things value-based design. Happy to answer any questions or pressing issues you may currently have. Find out what 1p Central Time is in your time zone here. If you’re a paid member, our meeting link is at the end of this email. If you sign up between now and 1p, we’ll send the meeting link your way.
In last week’s letter, I wrote about the world we exist in as designers after LLM’s broad-scale introduction. Now I’d like to talk a little bit about the labor aspects of LLM, which are, like it or not, heavily influencing those who buy design as of press time.
There’s a lot of noise out there. Almost all writing on this topic is unreadable. I sort of get why people are getting loud, though, considering western civilization is slowly collapsing and nothing feels remotely safe or permanent anymore. Kinda makes sense why people would think LLM is a cool safety blanket of future-forward manifestation, eh?
Fortunately, I think I’m one of the only people trying to write about any of this who isn’t resorting to hyperbole. That is because I am deeply suspicious of hyperbole, and I’m even more suspicious of hyperbole when it’s applied to something that is both structurally unproven and broadly disliked by most people, while also possessing elements of addictiveness.
So let’s step back and reiterate some first principles:
- People buy design. Design is a B2B industry. We sell to businesses. Sometimes our “customers” are big businesses that hire us for full-time jobs. Others hire us as clients to do one-off projects.
- Design is a form of leveraged power. As a designer, you are responsible for load-bearing, consequential changes to a product or service. You are, in some form, being handed the keys to the proverbial business sedan.
- Executives make economic decisions. That means their first duty is to protecting their profit imperative, not to buying design for its own sake.
- Design must be proven. Yes, design is a low-risk, high-ROI activity, but unless that is proven to those in power, they’re not going to buy it.
- Everybody designs, and everything is designed. It can either be designed by you, informed by your own expertise. Or it can be designed by someone else, probably badly.
If you agree with these, then we’re in for a smooth ride. Importantly, none of these principles have anything to do with LLM. They are evergreen and largely unchanged over the past 40 years. Any new technology is introduced in the context of these ideas.
Now, let’s look at that list and think about what LLM does for those who buy design:
- LLM creates the perception of reduced labor cost. Because LLM is already palpably reducing labor costs for developers, and because LLM can output images, it’s reasonable to believe that LLM will, at some point, allow designers to do more with less. It kind of sucks right now, but maybe someday? Someday soon?
- Executives are already making more & more design decisions, so there’s increasing pressure on strategic work to deliver economic value. Strategic work is consolidating at the top of the pyramid, which is also shrinking. As a result, the bottom is dropping out for production work.
- In short, LLM makes executives think that they can design. And they can! Because everybody can design. Knowing that design is a research-focused process, though, the question becomes: should they design?
- Finally, executives do what’s trendy. At one point, “design thinking” was trendy. Now it’s LLM. The hive mind is real.
None of this is great! It means that designers are losing power, and some of them are going to lose their jobs. It also means that executives might prematurely embrace LLM as a design tool, forcing it on departments that are unprepared for it.
Smart executives will do two things: pay attention, and move slowly. Wrecking your team to embrace a brand-new, unproven technology isn’t a good look. Doing nothing is also not a good look. And due to the aforementioned hyperbole, there isn’t a legible middle path.
That means you’ll probably start to find executives polarizing into a few camps:
- Those who take a wrecking ball to their teams and try to LLM-ify the whole organization yesterday.
- Those who throw a button at the top of their app with a sparkle emoji and claim that they’re embracing the future. Nobody clicks the button. In a few years, the button goes away.
- Those who do nothing, or appear to do nothing.
The second is the most common that I’ve personally seen, but I’ve seen a bunch in all three.
Knowing that labor is about to shift in many of these organizations, here’s what LLM could do for designers, in case high-quality tools ever happen to be built for them:
- Production work gets faster, eventually. I haven’t yet seen a LLM tool that quietly reduces friction and makes production easier, but in enough time it might happen. Lord knows I’d like smarter guides & layout, more robust design systems, and not much else. Is any of that ready for us yet?
- Because of the drop in time it’ll take to perform production work, that means the work of selling design will become far more important than it already is. I often joke that my job is 5% designing and 95% shipping. That’s about to get closer to 100%.
- Many others have mentioned this point, but we’ll probably see a big shift towards strategic thinking, which will make it even more imperative that design reclaim its power-wielding authority in an organization. It will also make research more obvious as an essential component of design, which we sadly lost sight of over the years.
- Smart organizations will use the reduced overhead in designing to create small, actionable moonshots that they can use as competitive leverage. Every LLM-using developer I’ve seen now talks favorably about how it has eliminated their “hack project list.” Think about what that might do for developing small projects within an organization.
In short, I just spent two years talking about leveraged power dynamics and the utter necessity for design to reclaim its authority by focusing on the generation of value. Enter this thing, which I don’t like but is here and making buyers weird. Fortunately, smart people still rely on other people when it comes to leveraging expertise. That’s where you come in. That’s where focusing on value comes in.
Remember always that LLM is not a substitute for your taste. At the end of the day, you’re responsible for what ships. Your taste is the final arbiter of what happens. And no LLM can outdo your taste.
In short:
- Executives are embracing LLM because other executives are telling them to or doing so, not because LLM is in & of itself an appropriate technology for the job.
- Executives will become more suspicious of buying design because they incorrectly think they’ll be able to do it themselves. There will be a messy find-out phase.
- It’s incumbent on us to remember & convey what design is: a process that generates customer- and business-focused insight. Then we have to convince others that such activities are both expert (someone well-trained in the practice should do them) and profitable (we should be hired to do them).
- We’ll spend less time pushing pixels, and more time using our skills to exercise our taste and work within a company’s power dynamics. This is probably a good thing? But it’s not here yet and nobody is really building for it.
And so I don’t know. I don’t feel super great about all of this, but it’s also better to know what motivates buyers of design right now – because that’s how we learn how to move.
A brief administrative note
Over the next few weeks, I’m going to jump around a few other sub-topics here, because a lot of you have said you liked reading this. If you don’t like it, though, and you want me to just get back to the design stuff, please hit reply and let me know. I might make a little sub-list for some people to read? LLM is probably insufferable to many of you? You didn’t expect any of this, right?
Oh, and one of you asked what LLM I use in my own job. I use Whisper locally to transcribe usability test & customer interview recordings, and nothing else. Most of my job is Slack posts, Zoom calls, and writing long things about LLM that I’ve been thinking about privately for two years. If you think that disqualifies me from writing about a thing, fine, your loss.
The rest of this post is available to members. Upgrade now to get:
- Every lesson on how to level up your career with value-based design
- Full access to our 10+ year archive
- Monthly office hours with an engaged community
- Early, discounted access to all new releases