At the end of 2025, I deleted my LinkedIn to reassert control over my digital footprint. I couldn’t confidently mitigate the platform’s inherent privacy risks, and, at the same time, I noticed that I could get most of its purported benefits by other means. I wish I didn’t have to be so mindful of how I interact with digital products—that’s what I’m working on. If you’d like to follow what I’m up to, I can add you to the email list for my annual update. Send me an email at william@[this website] to be included, or to get in touch for any other reason.
Let me dismiss a few concerns that immediately accompany the phrase “I’m not on social media”. First, I’m not in imminent physical danger (as far as I know). Nor are there skeletons hiding in my résumé. I had plenty of connections; in fact, I was only a few dozen short of the vaunted “500+” badge. I have only a couple lukewarmly conspiratorial views about big tech companies, and those had nothing to do with this decision. I do take myself, my networking, and my career seriously, and in fact, that’s precisely why I’ve chosen to remove myself from LinkedIn.
I work as a lead engineer on a cyber intelligence team for a large American financial institution. As such, I am especially mindful of my digital footprint: that work exposes me to a small, but real possibility of interest from organized criminals and espionage groups. In 2025, as one example, the cybercriminal group Scattered Spider targeted two members of Google’s cyber intelligence team, attempting to publicly blackmail Google into firing them.Google called their bluff, and federal authorities soon arrested the teenagers—because who else would be behind such theatrics. I don’t evaluate my personal risk to be particularly high, but I do consider it good professional practice to stay mindful of what’s on the internet about me. After all, if I’m not willing to follow my conclusions about software and cybersecurity to their ends, why should anyone else follow them?
On the surface, LinkedIn has a fairly standard privacy policy. They sell all my data to third parties, albeit “anonymized”.Associating anonymous data with its source individual is such common practice that Wikipedia users have seen fit to dedicate a whole page to it. They also sell any data I set as publicly viewable on my profile directly, non-anonymized. All of that is typical practice for a social media company. With LinkedIn, the salient risk comes from how much information a user is encouraged to enter, and how much of it is public by default. A typical profile will contain a first and last name, a headshot, contact information, employment history, and education history. A more complete profile might include résumés, career milestones, photographs, or details about hobbies or interests. LinkedIn also displays my connections on my profile, and shows my degree of connection (and any mutual connections) to anyone searching for me. Nearly all of this information is, by default, publicly viewable and searchable. The platform offers me the ability to hide almost all of this information, but only by sifting through a dozen confusing privacy menus and hoping I’ve configured the right combination of controls to keep me appropriately protected, but also appropriately visible.
Besides any additional risk from my work, this level of information exposure can be dangerous. It is shockingly easy to use a LinkedIn profile to turn two or three pieces of information—say a first name, city, and job title—into someone’s full life story. A gamer or a polemic commenter might be doxxed. Scammers will use public information to draw victims further into their scams.AI-driven spearphishing scams using LinkedIn profile information exploded in popularity in the first half of 2026. Someone under physical threat—from an abuser or a stalker, for example—might expose themselves to harm. So far, I’ve been fortunate enough to avoid those situations, but as always, an ounce of prevention is worth a pound of cure.
Of course, although it has its own unique flavor, LinkedIn is hardly the only online platform with privacy risks. And if all I had online was a single LinkedIn profile with minimal information, it wouldn’t be all that useful to a malicious actor. And thanks to GDPR, I can even request that they remove all of the information that they have about me, should my risk level jump up. But like many, I had a messy trail of breadcrumbs—old forum accounts from middle school, public Strava activities, discarded dating app profiles—and I had a LinkedIn account for long enough that they had sold much of my data to data brokers, which one GDPR request wouldn’t do anything about. A common response, at that point, is to throw up one’s hands and say “all my data is already out there; what can I do?” Why go to the trouble of cleaning up a LinkedIn account—or anything else, for that matter? But we actually have more agency than the common cynicism suggests, for a simple reason: storing data costs money.
Companies collect data either to sell it or to support their customer experience. With that lens, anything but the most basic information loses value quickly: it matters to advertisers if someone searched for “cute puppies for sale” three days ago, but it doesn’t if it’s been three years. So, most companies have data retention rules—expiration dates on your data—that will automatically delete most things saved about you after a certain period of time.Retention periods typically range from six months to ten years, depending on business, and technical, and regulatory constraints, and aren’t usually public information. So, by adopting privacy-oriented technology practices and cleaning up what I’ve put online,In most cases, to protect user experience, companies won’t delete data that you have published yourself: things like old Facebook posts or YouTube videos. Those, you have to go remove manually. Data brokers will also retain data for longer time periods, but generally respect takedown requests. I have been able to do something about my online privacy. It may take a few years to come to fruition, but without much additional effort.
Before discarding LinkedIn entirely, its benefits deserve exploration. Maybe the tradeoff is worthwhile—after all, many people have decided to make it. First of all, some people see a LinkedIn profile as a mandatory shibboleth for the modern worker, akin to monotone circumlocution or a disingenuous affinity for golf. Statements to this effect often arise from dated or overly simplistic beliefs about an esoteric but absolutely rigid socioeconomic hierarchy. I don’t consider this a substantial point. And I expect that all this explanation will reassure any business leaders who may worry I’m some sort of malcontent. But second, many claim that LinkedIn is an essential way to keep up with what’s going on in your field and in your network, or just that it is the easiest way. I have found that for myself, for both software engineering and cybersecurity, it is not. But there’s a lot more meat on this bone.
Software engineering does not change as much as it gives itself credit for. The most popular languages and frameworks have been on top for decades.Seven of the top ten most popular programming languages from the 2025 StackOverflow Developer Survey were created in the last millennium. For my craft as a software engineer, there is no reason to check the news daily. I prefer to make a dedicated effort to learn a new tool a handful of times per year during long, focused sessions rather than scroll through LinkedIn. The latter approach of undirected microlearning feels productive when unexamined, but in reality is riddled with serious problems. Although LinkedIn presents itself as a useful and effective source of business news—industry trends, layoffs, new attacks—the issue is twofold. First, the information on LinkedIn is not crafted to inform, and second, the platform presents it in a way that creates unnecessary anxiety and encourages superficial cognition, all in pursuit of the platform’s user engagement metrics.
Consider the provenance of the information on a LinkedIn feed. A disturbing proportion of it comes from illegitimate accounts run by state-sponsored agents trying to push a narrative beneficial to their country or politics.Precise numbers are difficult to estimate. In 2022, NPR and The Washington Post both reported on disinformation about COVID-19 and the 2020 US presidential election spread on LinkedIn. But suppose I can filter that out—a tentative supposition, as disinformation actors are good at their jobs—and I only encounter posts from legitimate accounts: businesses and individuals. A problem remains: even these accounts are not trying to inform me. In fact, they’re often actively trying to mislead me.
Businesses and individuals post on LinkedIn for a few reasons, but very rarely to educate. Almost always, it is to persuade or sell their audience on one thing or another. Salespeople will cherry-pick engineering problems or cyber events to persuade potential customers that their product is worth buying. Executives and employees will share inflated or dramatized stories about their accomplishments to attempt to establish their brand. Businesses will post jobs to boost recruiting and satisfy legal conformities, not because those jobs actually exist. Combine all those perverse incentives with the common idol of many social networks—engagement—and you end up with insufferably quixotic “LinkedIn influencers” working in concert with overeager social media marketers to reduce my feed to an inscrutable morass of buzzwordy corporate drivel.Or recently, an AI-generated simulacrum of such.
Though annoying, there is nothing profoundly unethical about most of these activities, to be clear. But their goal is not to provide me with useful information. If any of these actors might have done so, it is by coincidence or, more dangerously, done strategically to lead me to a conclusion they want me to have about them. Maybe it’s clearest in the contrapositive: if you want to publish verified, important, new information about your industry, is LinkedIn going to be your choice of platform? At best, I expect you would use it to announce that information’s release elsewhere: on your own website, in a press release, in a white paper.
But suppose, against the odds and the incentives of The Algorithm, I could curate a high-quality feed filled with people or organizations I trust to inform me: people who might use LinkedIn to point me towards useful conference talks, blog posts, or white papers. Even in that supposed utopia, I still run retina-first into the litany of issues inherent to algorithmically-generated infinite-scroll feeds.
First, and at this point in the public awareness, frequent, unstructured “doomscrolling” through social media creates chronic stress. The self-reinforcing unease of either missing out on something (or worse, finding something) at the bottom of an infinite feed perpetuates a cycle of anxiety punctuated by mind-numbing microdoses of unhealthy dopamine release. From a psychological standpoint, the phenomenon of burnout is best understood as the accumulation of stress over time. From a practical standpoint, the phenomenon of burnout is clearly a natural result of spending one’s downtime scrolling through infinite feeds.
Second, the emotional potential of a snippet of information—its latent ability to cause happiness or hurt—is inversely correlated with its temporal and spatial distance from its recipient.Psychology describes this phenomenon with the construal level theory. In other words, I don’t care that much, on an emotional level, about a devastating typhoon that made landfall in Guangzhou thirty years ago, but I care a lot about a hurricane that’s headed for my hometown on the East Coast next week. For me, the decision between having a strong emotional response to industry news or processing it at a higher level of abstraction is easy. I work in a complex and technical field. I can’t have my day-to-day decisions guided by fleeting emotional responses to irrelevant events. Imagine if I had my team recode all of our software from Python to Haskell because I read a LinkedIn post decrying that all non-Haskell teams would be cast into the permanent underclass when the AI singularity came. If that reality truly comes to pass: first, God help us all, and second, we still have—at the very least—a few days to respond. Structuring the information in a way that encourages sober reflection can only help us. Real-time feeds—and also things like “monitoring the situation” dashboards and even news push notifications—rev emotions from zero to redline dozens of times per day. That alone is enough to induce some level of anxiety. When a post about layoffs amplifies that now-hair-trigger nervous system, it sets you off for the whole week.
Third, passive information consumption is just not an effective way to learn. Research shows that activities like rereading highlighted notes or even listening to a talented teacher don’t, on their own, cause learning. Passive studying creates a metacognitive illusion that one has learned a subject, while in reality, all it provides is familiarity with a subject.Psychology Today’s article “The Feeling of Learning Can Be a Psychological Illusion” and its citations are a great read. Have you ever followed a lecture completely, and then a few hours later looked at your homework, only to realize that you have no idea how to do it? That’s the difference between familiarity and understanding. The brain believes it has learned, but it has not. Scrolling through a LinkedIn feed works the same way—the brain doesn’t get much out of it, even though it can feel like it does. A little while ago, I came across a tool for self-reflection. The results were enlightening. It’s a simple question I ask myself: what did I read two posts above what I’m reading now? If I can’t answer, I’m not learning. I could almost never answer.
An infinite-scroll feed tricks the brain in all three respects. But the obvious objection to abstinence is pragmatic: won’t I miss something important? It’s a fair concern, but in practice, it’s unlikely.
I generate nearly all of the value in my work during long, focused blocks of time on projects planned weeks or months in advance. In this capacity, I find reading the news once per day unnecessary. I certainly need to understand the way technology is going, but once a week is more than frequent enough for me to keep tabs on that. Of course, others in different roles, like intelligence analysts briefing a security operations center, may require a different cadence. But my job is not to react to the present, it’s to plan for the future. That requires a steady mind. I believe this may be an inevitable conclusion of the digital age: “keeping up with the news” can no longer be considered an axiomatic virtue. The degree of “keeping up” must be intentional, or we will go insane. I consume technology news more deliberately than LinkedIn encourages, and mostly from periodical newsletters and print publications.To its credit, LinkedIn offers a weekly summary email of posts from your feed. Were I to attempt to use LinkedIn as a source of information, this would be the only way I interacted with the platform.
While I have a number of problems with the quality of information I get from LinkedIn, I do agree with the general consensus that LinkedIn is useful for networking. Few decisions are entirely without consequence; this is the primary tradeoff I’ve had to make in staying off the platform. However, given my career mission, I’ve found satisfying workarounds to the thorniest issues stemming from my absence. On the whole, networking is a two-armed beast: first, I’d like to follow what others are up to, and second, I’d like to broadcast to others what I’m up to. Broken down, both can be managed without a social media platform.
The first arm, following others, has two segments: first, passive, serendipitous discovery of news and opportunity, and second, active search for information.
As discussed, infinite feeds are not an ideal platform for passive information discovery. The signal-to-noise ratio is extraordinarily low. However, LinkedIn is how many of my colleagues, both current and former, share information about what’s going on in their careers. To close this gap, I shoulder the burden of keeping up with them in other—preferably in-person—ways. This doesn’t match the frequency with which I’d be able to keep up with them on LinkedIn, but the signal-to-noise ratio is much better. As a disclaimer, I’m still actively refining my process for doing this and I can’t say much in the way of its efficacy. I can say, however, that a 30-minute chat satisfies the ethical fulfillment of keeping up with connections far better than liking a LinkedIn post does, and that both the quality of information I receive and my absorption of it are dramatically better. As a supporting anecdote, I got my start in cyber intelligence by invitation from a former coworker who I had previously hosted as a friend for a New Year’s Eve party. Trust comes from real relationships. LinkedIn can be a decorative sprinkle on top, putting you at the front of someone’s mind for a few moments, but it’s a poor tool for building real connection.
To the second branch of following others, the active search for information, I must admit that LinkedIn is a valuable tool and the title of this essay is a partial lie. I do have a LinkedIn account. It is devoid of personal information and has zero connections. I use it—quite effectively—for research into companies. There is a downside here, which is that I don’t have access to second-degree connections. However, in my experience (and having had many connections myself), these are really only useful for the extraordinarily narrow case of understanding work environments at large companies. At a small company, you’re unlikely to have second-degree connections at all. At a large company, they won’t be able to aid a highly bureaucratic hiring process much. I mostly use my research LinkedIn to find groups of companies, names of notable contacts, and posts by those companies or people. Morally, doing this makes my decision to abstain from LinkedIn a bit selfish: if everyone were to follow suit, it would ruin this utility. On the other hand, if everyone were to begin following suit for the same reasons, the market should quickly adjust to satisfy my desires (and everyone else’s, in this Kantian scenario). On top of that, the best successes I’ve had in active searching were not on LinkedIn. For example, in college, I cold-emailed the innovative iOS weather app Dark Sky for an internship with a heartfelt endorsement of their approach and a practical suggestion on what I would add. The CEO spent half an hour on the phone talking with me about it.Unfortunately, the whole company was 12 people at the time, and they were onboarding two new engineers. So, they didn’t have room for an intern. He told me to send over my résumé after graduation, but Apple bought them before then. In hindsight, maybe I should have followed up anyway—but then, my security sensibilities wouldn’t be as refined as they are today, and you might not have ever had the distinct pleasure of reading this. Good research requires depth, which LinkedIn doesn’t facilitate or even encourage.
The other arm of networking is the problem of broadcasting what I am doing to others; in other words, making myself known. I’ll extend an apology that I have made myself harder to find for well-intentioned recruiters. However, that apology does not extend to any recruiter who ever messaged me on LinkedIn. The number of times I was offered a salary less than half mine for a dead-end position, offered a job outside my field, or even greeted with the wrong name was exorbitant. I found no value, over the course of years, in a variety of market conditions, in having my information exposed to the general public for the purposes of mass-market hiring.
That being said, I do want my connections to be able to keep up with what I’m doing, and without much effort on their parts. Of course, in-person conversations are reciprocal, but I can do more for others. So, at least once per year (possibly more often for significant events), I send an email newsletter. I don’t expect others to entirely disrupt their usual patterns to accommodate my preferences, and a concise newsletter fits nicely into a normal digital life. If you prefer paper, share your address with me, and I will happily send you one in that format, too.
To that point: the most alarming possibility to me in deleting my LinkedIn was that I would imply to my connections that I didn’t value them, that I didn’t want to know what they were up to, or that I wanted the world to leave me alone. Quite the contrary—I’m a social person, and I wish we had better-designed social media platforms so that I could engage with friends in ways that would deepen and meaningfully maintain human connections. Unfortunately, that isn’t the world we live in.
So for now, if you’d like to get in touch with me (or if you’d like for me to keep up better with you), I invite you to send me an email at william@[this website]. I check that inbox a few times per week. If I’ve given you my number, send me a text, WhatsApp, or Signal message anytime. Since I don’t have a profile for you to glance at, here are a handful of ideas for things I do to keep up with people instead. Especially if we’re friends or colleagues, I invite you to initiate any of these with me as well:
- Get coffee, tea, or lunch,
- Send a short letter, postcard, or Christmas card, or
- Schedule a recurring Zoom or FaceTime call.
Even if you do none of these things: if you ever discover something I might do to help you, I am still here. I (genuinely) cherish unexpected texts from friends or colleagues I haven’t spoken to in years. You won’t, however, find me on LinkedIn, trying to maximize my exposure to people and information at the expense of my privacy and my sanity.