Tech vs. media is a reductive false construct when the real villain is hoarded wealth and massive corporations who view people as metrics. Neither industry understands how the other works, and both (all) industries are filled with a bunch of shady players along with some very good people. Both tech and media industries urgently need to slow down and fix their massive systemic racism problems.
I wish tech folks gave far more credit to journalists for the extremely hard, public work they do. And I wish writers and editors understood algorithmic technology and data-driven content better, which is why I write this newsletter.
Like many of you I’m reimagining systems and trying to find my ideal version of the future, knowing that we can’t just completely demolish the systems of the present. I fully believe many of the tools we have access to right now are democratizing discourse and should be maintained. Yes, they need to be better. But this week I’m focused on naming what works.
No review this week, and honestly, I looked for software to review, but nothing was really sparking any joy. I was discouraged when I read the following description on Product Hunt: “Make ads like a machine.” No. I do not want that future.
So if you have software you’d like me to review in the future, please, send an email or leave a comment!
If you’re new here you can
The 1990s internet as a creative haven
Like any good writer, I’m a punkass anti-authoritian in search of an audience.
When I was a teenager I coded and published my own websites, making logos with neon grunge fonts on black, typing up my angst poems, watching my stat counter go up 20 visitors each day, sharing my site with friends I made on IRC or joining webrings.
I read their websites instead of teen magazines. I felt like I knew the girls who published them, even though they used aliases like Starshine Cutegirly or Fake Plastic Life. (If you’re curious, my web aliases were Mischief, Sweet Little Agony, Rave and Christian Suburbanite, from what I recall.) I became friends with some of the girls and exchanged emails with them.
I had real life friends, too, but it wasn’t until my second Livejournal that I even connected with them outside of AIM. I thought my writing and poems were too nerdy for sharing, and I didn’t know any other girls who coded (I also didn’t ask). Also, they were my secret diary websites, so: secret.
I developed a keen sense of internet literacy early, and I knew my parents were watching kindof, so I shied away from any website that wasn’t clearly produced by a teenager. I never much liked reading the flame wars in newsgroups, and I was only in search of more photos of Billy Corgan or facts about Velvet Goldmine.
For me, the internet was always a safe place where I could connect with likeminded makers who understood me in a way Seventeen could never approach. Magazines and radio and tv were already hot garbage in the late 90s, thanks to deregulation and two decades of unchecked white wealth and commodity fetishism.
Even Sassy had stopped printing by the time I was a teenager, and Bitch and Bust were hard to come by. You had to get to the Borders magazine rack at the exact right time of the month to snag a copy.
On the internet I discovered new bands and zine distros and suck.com and art experiments that were nothing but fun.
I recognize my privilege. I know that it’s not like that for everyone. I know how lucky I was to have a DSL-connected computer in my bedroom that my loving parents let me use more or less unchecked. But I see myself in a lot of the TikTok teens today, learning to use the tool and the algorithm because it’s fun and it’s new and it’s different.
In the Internet sea I willfully swam as far away from racism and sexism as I could because I saw plenty of that in the real world, on tv, in magazines. It wasn’t why I was there. Pure luck and a low profile kept me from online predators (also encountered plenty of those in the real world). Even though I dreamed of being internet famous, I never really broke through.
Web search before Google was terrible
When Google was released I thought it was so cool because I could find Tori Amos and Radiohead websites so much quicker. I never wanted to find Rolling Stone content about my favorite bands because I was already reading those magazines the second they arrived. I wanted depth, what we now call longreads. I wanted to read women who were writing about music, and they were mostly online. I wanted to stan. I wanted to share.
Before Google, web search was a minefield. Do you remember Ask Jeeves?* Hotwire? You never found what you were looking for, unless you were looking for porn or basement nerd types, verticals the internet has always favored. I remember wishing for a universal search engine before Google existed.
Google won the early search wars and became the monolith it is today because it always brought better results. From the beginning, it surfaced much better websites. It surfaced independent media. It surely didn’t democratize media, but along with Wikipedia and IMDB and all the information it was crawling, it made web useful long before mainstream publications did.
*What kind of elitist nonsense was Ask Jeeves? Why did you have to make your internet butler answer your foolish questions?
Search engine optimization (SEO) as a means of connection
When I moved into digital publishing I learned to optimize for Google because I wanted to learn how to make the content I was creating find its audience. I’ve always approached it from that lens:
SEO helps your content find the people who want to see it. Good SEO has nothing to do with advertising and has everything to do with quality, relevant content.
All that said, oh yeah, Google the company is fucking evil. The utopian promise of digital search and discovery can never be achieved by a private company run by elitist men who don’t understand that they have to massively flush the original porn-and-basement-creep dataset. In a recent organic search product release, Google described the offering as “a big step forward in democratizing access to digital commerce,” which, ha, Google e-commerce search optimization in 2020 has nothing to do with democracy or people’s access and everything to do with being able to afford an additional FTE who can read and understand digital optimization.
But the good parts of SEO, how the algorithms that power search help people find the content they want — those are still brilliant things! They are still working! The current wave of anti-racist social change is powered by social networks and search engines that help people share and find information they would have no access to otherwise! And even if Page, Brin & co. are douchenozzles, non-ad-powered organic search is a great concept.
Were we to completely rethink how tech and media operate in regards to organic content discovery and data collection/surveillance, here’s what I’d save about organic search:
1. Availability of publicly available anonymized, mass audience search data.
Yes, it’s technically high-level surveillance, although typing a phrase into a search bar and getting a result is no different (to me) than going to a free concert and getting counted at the door.
Search query aka keyword data is amazing! If you are the kind of writer or editor who cares about building an audience, search queries can map out exactly what topics your audience cares for and the language they use to talk about those topics.
When SEO folks talk about search queries and keywords, we’re looking at the volume of people who type a certain phrase into Google’s search bar. The general ranges of query volumes can be accessed by anyone with a Google Ads account, but most SEO folks use premium tools to get more accurate estimates.
Search query or keyword research will tell you that 9,900 people every month search “Instagram algorithm” versus only 3,600 who search for “Facebook algorithm.” On the surface that means that more people are interested in learning how the Instagram algorithm works. Looking at these two terms with no further context indicates that the larger number of Facebook users are less interested in how the service works in 2020, or that maybe they don’t know there’s an algorithm running in the background.
If I were looking to write an article about either of those algorithms, then, I would look more deeply into other keyword research that questions that people were searching around those algorithms to ascertain where the gaps in knowledge lie, and what their intent really is.
Search data identifies what the general public knows or where they are a confused about a certain topic. Coupled with critical thinking and deep reading, query data can be immensely valuable in understanding how the general public understands a topic. Not everyone goes to a library, but everybody uses search.
SEO keyword research is often presented in a list of numbers and terms, organized in terms of “opportunity” without understanding intent. The opportunistic or growth hack point of view can lead to a misuse of the data, comparing two terms based solely on bigger-versus-smaller numbers.
In contrast, high quality keyword research is organized by topic and entity, builds connections between the content creator/planners and the audience.
Good search professionals spend time understanding what the audience wanted to find when searching those queries — what we call intent. But it’s great data. (Noting here that although it is publicly available, the data is not publicly owned, nor is it easy to find or access without training. That’s problematic! But the data itself is good.)
Writers, editors, content creators, publishers and search professionals should collaborate on determining who is searching these terms and what they indicate about the audiences they want to reach. The search query data is the foundation of a plan, a content strategy or an information architecture designed to serve the audience’s interest.
It’s not data that should be used only by one team or one agency at one juncture, nor should the data be retroactively applied to already written content. (I.e., you shouldn’t have someone “SEO-ify” a document if no search data has been used in the document’s creation).
Search data provides a wealth of knowledge and background for a writer on literally any topic. Google provides keyword data because they are selling ads, but when you use that data for content creation, you’re just weaving in the language of your audience. You’re saying to your audience, I hear you, and here are your words reflected and the answer you want to find.
2. Using search data for accessibility, transparency and clarity
We all know that in marketing and business people use silly words that mean nothing. Every business wants to “leverage solutions for better outcomes” and so on and so forth, you know what I’m talking about.
Every industry has its own jargon, but people typically don’t search with meaningless jargon. They search with meaningful words! SEO investment leads businesses to words that clearly describe their products, versus the jargon version of nothing.
Writers in both editorial and advertising have always had a habit of expanding their vocabularies, throwing in $5 word when a .25 will do. Both novice and experienced salespeople are pros at creating meaningless sentences filled with buzzwords. Good SEO eliminates that vague or highfalutin’ or badly written language in favor of specifics.
For example, if you were a person searching for web comics, you would type in “comics” and not the highly theoretical and infrequently used “sequential art.” Good SEO logic wants you to use the word “comics” in your headlines, information architecture and a few times throughout your content. Because most people say “comics” or “cartoons” before they say “sequential art.”
Lowest common denominator search optimization is no good, of course. Good SEO practitioners don’t identify target language simply because it’s highly searched. They identify what the client or content creator has not already thought about, bringing new dimensions and ideas into the work.
Good search data makes content more accessible to a wider audience while generally improving the quality of content.
3. Using search data to dismantle existing ideas of what audiences want to read
Professional environments, especially those who hire employees with the exact same economic and social backgrounds of existing leadership, produce and regurgitate assumptions about an audience or customer base, even when those assumptions have changed or were never correct to begin with.
The core function of SEO is to align an organization’s content with what the audience actually wants to find, not what leadership’s “gut feeling” thinks they want to find. “Gut feeling” about an audience is more likely to replicate what leadership feels, and SEO data keeps that in check.
Whether it’s a manufacturing company with a content marketing program or a national news website, assumptions about what the audience knows or understands should be considered a primary function. And anonymized, unguided mass search data is every bit as valid as a poll or survey.
Every project I’ve ever completed in my nearly eight years working in search illuminates some new facet of an audience for my clients. And I’m in favor of any practice that leads to a broader understanding of a more diverse group of people.
4. Using search data to support a positive user experience
Search optimization — and the practice of digital optimization in general — should support better websites, higher quality advertising, fewer shitty pop-ups and ad units. Organic optimization means that you’re not trying to hack your way to success with dark patterns or weird psychological tricks.
Organic optimization means you’re relying on your content to find its best audience in the most accessible way possible. That’s it. It’s about opening content to everybody, where that content should be publicly available.
Search optimization argues in favor, always, of a positive user experience. Any future iteration of search needs to keep the user’s access to information in mind — even if that access is eventually membership-only. (Currently I am a fan of the 3-5 articles per month paywall for accessibility.)
I could go on for ages about what I’d change about Google and search as it is, and I don’t think Google should continue dominating search without additional transparency. Mass advertising-supported content from private media companies also needs to be rethought. The CPM model needs to vanish, finally, and independent digital media models need to be supported and explored.
But using search data as a gateway to understanding and creating content for an audience? That should never go away.
Because when you’re in search of an audience and want to create service content, news content, informational content, human-oriented content, sparkly poems, zine directories, anti-racist resources and artistic experimental websites, search data and the SEO professionals who understand it can be a means to a deep connection.
Are you picking up what I’m putting down? You can always
Content tech links of the week
I’m very excited to dig into the style guide of the Trans Journalists Association. (Also, I’ve noticed so many progressive websites with a version of that neon bluish purple recently that I’ve started calling it Woke Purple.)
Graphic Content: How Visualizing Data Is a Life-or-Death Matter on the new-to-me Frieze.com
Here is my favorite take on the pushback from writers in power who are upset that progressives are questioning them on Twitter, by Osita Nwanevu in the New Republic, the website section also in a shade of Woke Purple.
I wrote a few weeks ago about loving Spotify even with its algorithmic flaws, but I wish it paid artists better. (I buy records of artists I like but Spotify needs to pay more.) Here are a couple of explorations of how Spotify currently pays artists, one from Pitchfork and another from independent artist Steve Benjamins who breaks down the data behind the streams.
And to prep for next week, here’s Digiday on Substack entrepreneurs.