SEO is no longer influential over the kinds of clients you want.
SEO is no longer influential over the kinds of clients you want.
So I'm tossing around an idea in my head, of implementing a search engine that computes a 'redundancy penalty' on every web page it indexed. The measure of redundancy would need to be multi-scale and consider repeated phrases, points made, and paragraphs as well, so not necessarily a simple problem.
If you think this is something worth pursuing just let me know by commenting below.
First, effective writing will almost always restate the main idea somehow; you'll want your algorithm to be more sophisticated than "demote content that seems to repeat itself."
Second, your novel algorithm will be expensive to compute, so you'll need some other signals to filter out very low value pages before you pay to analyze others.
Finally, your technique will be valuable as a ranking signal until you (or someone else) begins to use your algorithm to help steer non trivial purchasing decisions. At that point, someone will dress up their low-value (or negative value) content to appear "high value" to your algorithm and your output will be filled with spam. You can delay this effect by keeping your user-base small, so consider selling private access to institutions who want a unique view of the internet for research or intelligence purposes.
Good luck!
You are falling into the trap of thinking that Google and its competitors are victims of this. They are not.
They are knowingly letting themselves get "exploited" because it turns out that those spam sites contain ads (that may be Google's) or analytics (that may be Google's), or make the search result page ads (from Google) look good in comparison and more likely to be clicked.
The heuristic to detect and block spam in general is very simple. Spam is there to get you to either buy something (the signal is the presence of a credit card form, "buy" call to action, or links that ultimately lead to the former) or view ads (signal there is presence of ads).
This would be very hard to game because those signals only work when they are visible (and obnoxious), which also makes them trivially detectable purely at the visual level by a crude classifier (trained on website screenshots with ads vs the same page with an ad blocker).
The resulting penalty score can be used as a ranking signal, so that all other signals being equal, a result without ads would rank higher than the one with ads for a given query.
This is not rocket science. The problem is that the mainstream search engines have the same business model as the spammers (and profit off each other), so there is no reason to suddenly slaughter the cash cow. The myth that search engines are "victims" to spammers/SEO provides very convenient plausible deniability so they have no reason to disprove that either.
A plausible and sustainable funding model is a good sign for sites.
This isn't a complete blanket ban on ads or the commercial web. If there isn't a profitable way to run a website that provides the content matching the search query, the ad-infested website will still come up first.
But now, a hobbyist, non-profit or even commercial enterprise that doesn't directly sell to you has a chance to outrank the ad-infested garbage.
That's where we are right now. I'm not sure that this is a solvable problem.
Effective corporate writing.
This goes back to the era 50 years ago when public speakers were taught "Tell me what you're going to tell me, then tell me what you're telling me, then tell me what you just told me."
It works fine for certain long-form speeches, but as the length of the document/talk decreases, so does the need for repetition.
For the length of 90% of what's on the internet, repetition becomes redundancy and nuisance filler.
Sentence 1: hints that the quote is wrong, because only corporate writing is repetitive.
Sentence 2: expands on this: corporate speech has been repetitive on purpose since the 70s.
Sentence 3: contrasts corporate writing, which is repetitive, with shorter content, which does not need to be repetitive.
Sentence 4: concludes that the quote above is wrong, because most Internet content (which is not corporate writing) does not need to be repetitive.
Sentence 1: specified that quote is about one specific kind of speech.
Sentence 2: explains origins of the quote - history. It is the history part that makes this sentence valuable and adds something to the reader (me).
Sentence 3: states that this is not effective to the kind of content we talk about - internet articles.
Sentence 4: concludes that repetition is irrelevant and wrong.
-----------------------
The way you tl;dr it, the poster would need to completely change the topic to "not repeat themselves". What they did was saying something new with each sentence, just keeping the "repetition" topic on.
Thanks
You very much not want it repeated the way it repeats now. Effective writing does not involve writing three similar paragraphs, it makes readers skip those paragraphs.
The internet articles we complain about are in the latter category.
Even videos like YouTube shorts, which are meant to be short, have this problem. They are seriously padded with repetition. Here's thing, have you seen this before? Few people have seen this. Do you know how it works? I'm going to show you. People are surprised when they find out how it works. But I explain it.
There must not be stats on choosing "never suggest this channel again", I select that pretty consistently when I'm looking for short form content and these videos waste all that time.
And of course LLM can detect duplicated content - even if it is rewritten, it should generate a similar meaning vector.
Spaced repetition, apparently it 'aids retention'. It's the same reason TV ads now run with a second shorter ad for the same thing later in the ad break.
I suspect that not coincidentally, blogosphere as used to be got much much smaller. Writing is not rewarding when no one finds it, so people just dont do it.
I think that AI will make it even worst. When everyone fires ChatGPT to get compilation of stuff, there is even less reason to write. So you dont.
Just make the site useful, informative and well presented.
To find out how users interact with it: Ask them, and/or do tests. Use the actual relationship with users to improve the site.
Even ignoring privacy concerns, this mostly avoids bad UX patterns like "cookie banners", makes the sites _significantly_ faster and lighter. You end up with something that is higher quality.
Exceptions to the rule exist, but that is always my baseline recommendation.