I’ve been fixated on removing my Google and Meta accounts, writing first about frustration with data collection, then an archive of past Instagram content.

Meta has a steep safe-guard (dark pattern) of preventing Instagram account deletion until 1 month after the request is made. I get the precaution: it would be awful to lose an account within 24 hours if there was a data breach. Apple has long had a 2 week policy (2018, 2021, 2025) for account recovery, which sounds breezy in comparison. Meta has plenty of other dark patterns, like the byzantine steps to opt-out of AI training, or exposing and selling phone numbers provided for 2-factor authentication (2FA).

Writing this post, I learned the term roach motel to describe services that are easy to join and hard to quit. It perfectly describes the feeling of entering a service only to feel trapped for fear of data loss or difficult removal processes, the two subjects of today’s post.

Data backups

I was pleasantly surprised by Google and Meta’s willingness to let me download my data, though the processes were still awkward.

Google will email saying they received your request:

But they do not send an email when your data is ready to download. I assumed it would take a day or so, then checked a few days later. I downloaded my data in time, because notably, it expires after 7 days.

Google data is downloaded in zipped folders. Thankfully Google Docs and Sheets are saved in Microsoft format (.docx , .xlsx ), not just links to a web browser, which is what happens if they’re synced to your desktop via Google Drive. It’s nice knowing the data exists on my hard drive, not a Google server forever. I’ve started using LibreOffice and .odt files, and Dropbox sync for document editing.

Meta provided downloads to my Instagram media within a day or so.

Viewing the media in an organized way requires an HTML file:

This HTML file refers to media in their custom folder structure, with every filename scrambled:

It’s anyone’s guess as to why the filenames are scrambled. Maybe they’re anonymized overall… or Meta just doesn’t want you playing around in here. Oddly enough, the scrambled names still sort by upload date which was enough for me to repost them in a useful order on my memories page (…after moving them out of their dated subfolders, one by one).

Overall these two experiences were not great, but workable. Much work remained.

Repetitive login strain

While it’s painless to set up a new email, it’s a commitment to change all your accounts to a new email address. First, I unpaired any accounts (around 75 !?) that used Google sign in, then I used my password manager to check which logins were still important vs. those that could wait.

In most cases I remembered my password, or relied on my password manager. In some cases, I hadn’t logged in for a while and needed to reset before changing the email. Even with a fresh password you need to provide 2FA with the old or new address. One website (craigslist, of all places) asks you to confirm on the old and new email before the switch is complete. It’s okay if you skimmed that paragraph.

This process brought up my conflict about passkeys. On one hand they solve the problem of password management, but on the other, I’m skeptical when a company uses biometrics for security. In some countries, law enforcement do not need your permission to unlock a phone through biometrics.¹ This is dystopian. Our devices hold personal information and law enforcement already have an agenda if they decide there’s something of interest on your device. Without your consent, they can use your body to access the device. Would a PIN passkey feel any more secure, allowing broad access to passkey-enabled sites simply because someone has my device and a PIN? Password managers have the same “master password” vulnerability, but are far more vulnerable without the device recognition technology of passkeys. I’m still thinking on this one.

To state the obvious: the longer you use an email address, the harder it is to change.

I understand why we can’t have an “email migration” or “delete all my accounts” button. As I said, I had around 75 accounts paired via Google login, and still others via email login – it’s become hours of work to switch to a new email, and I’m still checking on accounts I almost forgot. I’m lucky there wasn’t a catastrophic Google mishap that inspired my choice to leave, as the exposure across numerous sites and logins is hard to quantify. And I’m one person.

There’s the possibility of real people using my old email. Do I keep an auto-reply / auto-forward? For half a year? A year? I’m a gig worker, and sometimes I interview for work that doesn’t line up, but people reach out a year later. Do you head off these concerns with a big “bcc announcement”? Email consistently has the highest success rate for work in my industry, so I think I’m giving it appropriate weight.

Finally, there’s sites I’m not using. Do I log in and delete the account? Or do I let it decompose and the company worry about it? Call me sentimental, but there’s something sad about abandoned accounts. Once I delete the old email, some company server will get an auto-rejection, then delete the account or keep getting auto-rejections and waste bits of bandwidth over time. How much data is sent to dead email accounts every day? If there’s a chance a company will keep emailing a dead account, and every email has a little energy cost, is it more ethical to delete accounts for sites I don’t want to use anymore? The mind boggles.

On the subject of registries…

Does unsubscribe even work?

Maybe I’ll revisit this in another post, but sometimes it feels like changing addresses is the best approach to avoid marketing emails. I swear I’ve unsubscribed 4 or 5 times from New York Magazine and Rolling Stone, and they still message about “stories I’m missing”.

The ideal Unsubscribe: Scroll to the bottom of the email, click unsubscribe, a new tab opens and says “Your email ___ has been successfully unsubscribed”.

Harmless: Same as the above, with an additional link to the main website you unsubscribed from. Text saying it’s safe to close this window and no further action is required.

Thin ice: Confirming which type of emails you’re unsubscribing – should have an option to “unsubscribe from all”.

Terrible: “Sorry to see you go!” or anything that makes a person feel badly for unsubscribing. Forcing the user to log in again to confirm the unsubscription. Worse, forcing them to log in but the website magically forgets the unsubscribe request and instead goes to the home page, forcing the user to go back to the email and click “unsubscribe” again.

Hell: If you’re using a VPN and there’s friction around “you seem to be logging in from a different location, would you like to change domains?” Then you change domains, lose your login, and start over. Usually happens with e-commerce sites. You may successfully submit your request, and there may be text on screen saying “It will take __ number of business days for us to process your request, in the meantime, you may still receive email communications from us”.

Unfortunately the FTC’s Click to Cancel appeal did not hold in a conservative federal court, so we’ll be dealing with unsubscription sludge for the forseeable future.

And those are “legitimate” marketing emails! Companies I’ve willingly given my address because I was interested in their product or service. I don’t even touch the content in spam and phishing attempts because those things look radioactive.

It’s easy to ignore spam email, but I find spam calls and texts extremely annoying. I used Incogni for a month using my old email address and information, though I still get the occasional spam call. I’ve been on the Canadian Do Not Call registry which maybe helped? I shouldn’t have to do forensics to prove this. Multiple spam calls a week is still too many. Weekly voicemail notifications for an eerie silence that I delete, except for that rare time someone needs a second to think before speaking. At least once, someone started speaking right as I reflexively pressed delete. I then cursed loudly, no clue as to who left me a message. I’ve heard it’s effective to request data removals yourself, so I’ll see if I can bring the calls down further.

Is there still FOMO?

The mourning process for Instagram has been complicated. I sort of miss scrolling to see what my friends are up to. But as I keep saying: social media is not a social life. It’s important to remind myself those images weren’t just for me. I may be aware of my acquaintances’ lives, but did that meaningfully affect my day-to-day?

At first I was sad losing contact with some friends-with-benefits. And yet, some people I wanted to give my number to actually didn’t respond to DMs, reminding me of the low-effort nature of many acquaintances.

I’ve noticed some listlessness at home, probably caused by a few things: waiting to afford new furniture for my apartment, going through a personal loss, hot weather, my dog developing separation anxiety, getting used to my new kitchen and local food options… All of these things are manageable, but it’s now easier to feel alone without a social feed as a reminder of the people I know. I wasn’t expecting to miss this about social media, and it may not even be real! I may be mislabeling a social feed as a kind of “ambient support”, even when that moment-to-moment feeling may have instead been closer to distraction, comparison, and melancholy. I specifically named Actually Missing Out as a feeling that came up when I saw friends doing things in Vancouver that I couldn’t join when I was away from the city.

There’s a weird catch-22: looking at a social feed fuels FOMO because you’re looking at all the events you weren’t invited to (or possible events you can go to soon), you have a slightly larger contact list than you do phone numbers, so there’s an illusion of possibility. Conversely, avoiding a social feed triggers very different FOMO, because you just don’t know what’s happening on a platform. You may have a deep conviction it’s not healthy for you, but there is a real loss of connection worth acknowledging.

Sometimes it’s helpful to distract from things that can’t be solved today (e.g. I won’t buy furniture for at least a month), but there’s a lesson here: where are instances I’ve happily chosen isolation with a social substitute? Times I laid down and double-screened the night away, when I could have used help, or even just a fifteen minute check in?

Conclusion

It took cumulative days of research to take my data seriously. The tech giants are so embedded into our lives that leaving platforms isn’t just mentally taxing, it feels like a genuine loss and period of uncertainty. Going through old logins was like ripping up a carpet; some parts were fine while other parts were tattered, stained, and fully due for removal. Thankfully there was never severe damage underneath (to my actual life or finances), but part of me wondered why I didn’t make this effort sooner. And there’s the key word: effort.

All of this is in pursuit of a word that feels like it’s lost meaning: privacy. When Google and Meta say privacy, do we believe them? Have you read their privacy policies lately?

I don’t want my email provider spying on emails or browser behavior to build their AI model. I don’t want my social life to be mediated by ads and algorithms. I especially don’t want to support platforms that neglect queer and progressive issues. I want to feel well-rounded, pursuing authentic interests, with the space and energy to do so. Google and Meta products feel like they want the exact opposite: to put me in a niche and tell me what to like, with ever-sharpening attacks on my time and attention.

Notes

1. In Canada, police need a warrant or court order to access your phone. Among the most vulnerable countries are China, Singapore, the UAE, Iraq, Egypt, Bangladesh, Iran, Russia, Germany, Thailand, and Turkey. The US, UK, and Australia barely missed the top of this list. Of personal relevance, “[US] customs police have broad powers that allow them access without judicial oversight, and they may insist people hand over their passcodes, too. In many states, mobile phone searches can be warrantless. Recent data also suggests that state and local police forces are using surveillance tech from the FBI, while protecting the secrets of the technology.”