Results and quality are often not entirely linearly correlated.
MrBeast said something along the lines of, spending 2x longer on a YouTube video might lead to 10x more views. The reason for that is, that a video that does decently well and a video that really goes viral are often quite similar in quality, it's just that last bit of effort to take everything from good to perfect that can 10x or even 100x your results.
In that way, it might be better to upload 1 really good video every month than 4 great videos or even 30 okay videos.
Results and quality are almost never linearly correlated. Often it swings one way or the other. 80/20 rule, or you spend time creating something very high quality and get outsized rewards for it.
It was something Marco Pierre White said on a broadcast once that has always stuck with me. He said "I've always said that perfection is lots of little things done well." Since I've started applying it in my own work, I have to say that I entirely agree with it.
This reflects my experience of late. It wasn't true for me in the past.
I think the issue is straightforward when you think about it, and the author basically spells it out, while missing one important fact.
When you're dealing with the long tail, it's better to have many small hits. Because long tail items are all pretty similar in their results.
But when you're dealing with the fat head, a slight improvement moves you dramatically up the graph.
Which means: When you're not very good, it's better to take many swings. Spending hours trying to be marginally less mediocre isn't going to get you much. Post often. Miss often. Score some hits.
But when you're good, it's more valuable to spend time becoming great.
On my newsletter[1], I like to think I'm in the middle. I'm not great yet, but I'm good enough that it pays to put in the effort. And I do see the results. The traffic and subscribers I get from a handful of high-effort posts (where I spent days on them) far outweigh the results of all my other posts combined.
Additionally, I think this will only become more true as time goes by, because AI is making is so easy to create low-effort content. That long tail is getting awfully long.
I already wrote and deleted 3 opening lines to this comment so I think I get where the conclusion is coming from. Maybe it has to do a bit with how much of the piece of writing is finalized or I guess put in order in your head before starting writing it.
I tend to have an abstract, unfinished thought on a subject and just start working it out while typing. Maybe I'm thinking linearly I dont know. But at some point I may hit a roadblock or change my mind and then go back and rewrite most of the previous material. It's why I rarely comment, it takes me a while.
So if this time is included in the writing, when it really is idea formulation and research then the time required to get to 300 words skyrockets.
In the context of a blog, especially if you aim for a certain style,flow etc I guess this is even more prominent. The more time you spent writing for the specific goal the better (an faster) you will get at it I suppose, but still I think some people just have a way of thinking that more robustly tranlates to text.
In the sense that you can link sentences like you link thoughts and get to a conclusion rather via a well articulated piece rather than what I (and possibly the op) get which is a mess of sentences referencing different points in the cognitive process that may just get to a dead end.
This has not been my experience at all. The more I publish, the more interaction I get. The more interaction, the more people keep reading (i.e., subscribe).
But this is clearly the author's experience, and if I were him I'd be deathly curious to determine the optimum number of blog posts per period for maximum interaction. One per year?
You definitely have the benefit of starting in 2007 and producing consistent content, which means you're pretty darn good at blogging at this point and can produce quality content faster, so I can see why posting more often benefits your situation.
I'd say it depends on who you're writing for, or what you're optimizing for. If you're somehow going for virality (viralness)?, then continued optimization in that regard makes sense. If you're writing about more niche topics or ideas, I don't know if the same applies. Not that refining your ideas is bad. Just that, as in the "common widsom" he gives at the beginning, a bigger surface area is likely to yield more stiff with a broad appeal. I've written a few things that for example got some upvotes and views on HN and elsewhere, and some that were complete flops. And I have no idea which ones are going to get noticed or be interesting - it certainly doesn't relate to what I find interesting. But I'd rather just write stuff I think is interesting, and hope occasionally someone else finds it useful, that try to optimize for others interests, which I think even if successful would probably be more boring overall.
Reminds me of an interesting comparison between the minds of the painters Raphael and Leonardo - the former was much more prolific (and successful at the time) because he had a template and would crank out art, whereas Leonardo would spend countless hours digging into the underlying context (literally studying muscles and tendons under the skin), which led him to make more impactful pieces by being able to do things like lighten the shadow on a chin due to the reflection of the light on the rest of a body, rather than a hard shadow.
I think this is a misunderstanding of the recommendation.
When you're just getting started on something new where the ROI is going to be minimal, it makes sense to not put a ton of effort on perfecting when you do not have a clear understanding what people want/need.
After you have a sense of what people are consuming, and therefore have some measure of an ROI, it may make sense to focus on repeating what worked.
I've read one of Judea pearl's latest books (called I think "the question of why") and while I found it extraordinarily convincing to make the case that the "data first" approach to inference is fundamentally flawed, I also couldn't extract anything quantitative from his graphs. Maybe my fault, it's on my to-reread list. Thanks for the link.
Actually, I believe the claim in the article is precisely that spending more time leads to better posts and that this leads to more subscribers. Basically, the author feels that he achieves more (in readership growth, but also on other dimensions he cares about) by taking the time to write one really good post instead of using that same time to write two mediocre posts; he contrasts this with the standard advice that publishing frequently is critical to building an audience.
Whether this is true today or not, wait a year or two until we're saturated with a million AI spam articles from an explosion of ai generated websites that all cross-reference each other for SEO. If you don't figure out some way to stand out, you're going to be in even more trouble than everyone else
It's been what ? 8 months since chat gpt is out there ?
Not much changed, if it was just a matter of shitting out SEO riddled articles to become interesting we'd know it by now. SEO spam only gets you so far, you still need relevant content.
In 8 months? 8 months? Go to udemy and look up “writing” and look at the student counts for the pages of course results. People need time to get their balls rolling. Everyone is still wondering whether to try their schemes, the schemes are only ideas today. Give it a couple more years for everyone to start executing them. It only takes one person and a modest investment to write a million words with ai and spread it across 100 new websites. What about when there’s a million people doing that, trying to compete?
You grow as a writer with practice, but you grow readership with quality. The incentive between quantity and quality can be at odds for a personal blog. I think it comes down to what your goals are for your writing. Personal growth or professional recognition.
I wonder if this has something to do with ChatGPT. Maybe Google and other services are downranking now if a site churns out frequent, lower-quality articles because they assume it's AI-generated.
MrBeast said something along the lines of, spending 2x longer on a YouTube video might lead to 10x more views. The reason for that is, that a video that does decently well and a video that really goes viral are often quite similar in quality, it's just that last bit of effort to take everything from good to perfect that can 10x or even 100x your results.
In that way, it might be better to upload 1 really good video every month than 4 great videos or even 30 okay videos.
I'd imagine the same goes for blogging.
I was very confused here until I realized you rank the intensity of 'really good' and 'great' in a different order than I do.
Great > really good
https://www.visualcapitalist.com/word-sentiment-scale/
I much prefer the quality over quantity model.
The most important thing we've learnt is to appeal to the algorithm of the platform you're on, not the humans who consume it in the end.
https://austinkleon.com/2020/12/10/quantity-leads-to-quality...
I think the issue is straightforward when you think about it, and the author basically spells it out, while missing one important fact.
When you're dealing with the long tail, it's better to have many small hits. Because long tail items are all pretty similar in their results.
But when you're dealing with the fat head, a slight improvement moves you dramatically up the graph.
Which means: When you're not very good, it's better to take many swings. Spending hours trying to be marginally less mediocre isn't going to get you much. Post often. Miss often. Score some hits.
But when you're good, it's more valuable to spend time becoming great.
On my newsletter[1], I like to think I'm in the middle. I'm not great yet, but I'm good enough that it pays to put in the effort. And I do see the results. The traffic and subscribers I get from a handful of high-effort posts (where I spent days on them) far outweigh the results of all my other posts combined.
Additionally, I think this will only become more true as time goes by, because AI is making is so easy to create low-effort content. That long tail is getting awfully long.
[1] For the curious: insurgent.ca
I tend to have an abstract, unfinished thought on a subject and just start working it out while typing. Maybe I'm thinking linearly I dont know. But at some point I may hit a roadblock or change my mind and then go back and rewrite most of the previous material. It's why I rarely comment, it takes me a while. So if this time is included in the writing, when it really is idea formulation and research then the time required to get to 300 words skyrockets.
In the context of a blog, especially if you aim for a certain style,flow etc I guess this is even more prominent. The more time you spent writing for the specific goal the better (an faster) you will get at it I suppose, but still I think some people just have a way of thinking that more robustly tranlates to text. In the sense that you can link sentences like you link thoughts and get to a conclusion rather via a well articulated piece rather than what I (and possibly the op) get which is a mess of sentences referencing different points in the cognitive process that may just get to a dead end.
But this is clearly the author's experience, and if I were him I'd be deathly curious to determine the optimum number of blog posts per period for maximum interaction. One per year?
When you're just getting started on something new where the ROI is going to be minimal, it makes sense to not put a ton of effort on perfecting when you do not have a clear understanding what people want/need.
After you have a sense of what people are consuming, and therefore have some measure of an ROI, it may make sense to focus on repeating what worked.
I've read one of Judea pearl's latest books (called I think "the question of why") and while I found it extraordinarily convincing to make the case that the "data first" approach to inference is fundamentally flawed, I also couldn't extract anything quantitative from his graphs. Maybe my fault, it's on my to-reread list. Thanks for the link.
Not much changed, if it was just a matter of shitting out SEO riddled articles to become interesting we'd know it by now. SEO spam only gets you so far, you still need relevant content.