What I'd Do Differently If I Started This Blog Today

An honest re-do of the 5th blog launch, with the 6 things I would change and the 4 things I would keep.

The exerciseThis post is an exercise I have been meaning to do for a while. The exercise is: pretend you are starting the blog today, with everything you now know, and write down what you would do differently. The exercise is useful because the post-launch hindsight is the part most "how to start a blog" posts skip. The post-launch hindsight is the part that tells you what actually mattered, and what was just the way I happened to do it.The exercise is also humbling. The list of things I would change is longer than the list of things I would keep. The longer list is the part that tells me the original plan was not as good as I thought, and the not-as-good is the part I want to capture for future-me and for anyone reading this post.The 6 things I would change1. I would start with the static site, not WordPressI started on WordPress because WordPress was what I had. The cost of WordPress was the 14 months of stalled writing, the permission issues, the managed-WP tier, the dependency surface, the deploy. The cost of static is the file copy and the small cost of learning how the static site works. I described the full reasoning in [the pivot post, but the short version is: static would have saved me 6 months of platform friction, and the 6 months would have gone to writing instead.2. I would write the first 10 posts before launchingI launched the blog with 4 posts. The 4 posts were not enough to feel like a blog. The 4 posts were enough to feel like a placeholder. The placeholder launch is the part that trains readers not to come back, because the placeholder launch is not a real blog yet. The 10-post launch would have been enough to feel like a real blog from day one. The 10-post launch is the part I would do today.3. I would pick the topic narrower, then expandThe site covers automation, GenAI, prompts, behind-the-build posts, releases, and process. The breadth is fine now. The breadth was too wide at launch, because the breadth made each post feel like a different blog. The narrower launch (e.g. just "automation how-tos" for the first 10 posts) would have made the 10 posts feel like a coherent blog, and the coherent blog is the part that makes readers subscribe. The expansion (to prompts, behind-the-build, releases) can come in posts 11-30, after the reader has a reason to come back.4. I would set up the post template on day oneThe first 5 posts had inconsistent layouts. The inconsistency was small (sometimes the lede was a , sometimes a ` `; sometimes the meta was inline, sometimes a separate). The inconsistency was the part that cost me time later, when I had to retroactively fix the inconsistencies. The template on day one would have been the part that prevented the retrofits. The retrofits are the part I have learned to not skip.5. I would buy the domain before writing the first postI wrote 4 posts on a WordPress install on a Hostinger subdomain, and then I bought the domain. The order is wrong. The domain should be first, because the domain is the part that gives the blog a permanent address. The WordPress install on a subdomain was a temporary home, and the temporary home is the part that signals (to me, and to any reader who looks at the URL) that the blog is not serious. The domain is the part that signals seriousness. The domain is the part I would do first today.6. I would commit to the cadence publicly on day oneI committed to the cadence in my head on day one, but I did not write it on the site. The "I will post 2-3 times a week" promise was a private commitment. The private commitment is the part that is easy to break, because no one is watching. The public commitment (e.g. "I post every Tuesday and Friday, and I tell you when I miss") is the part that holds me to it, because the public commitment is the part the reader can call me on. The public commitment is the part I would do today.The 4 things I would keep1. The voice profileThe voice is the part that makes the blog different from the other 200 AI/automation blogs. The voice is the part that turns a generic "I tried this AI tool" post into a "here is what I actually used, what worked, what did not, and what I would do differently" post. The voice is the part the [Blog Writing Factory prompt pack enforces, and the pack is the part that makes the voice consistent. I would not change the voice. I would not change the pack.2. The point of viewThe point of view is "I am a builder who ships small things, and I write about the small things I ship." The point of view is the part that filters out the posts that would dilute the blog (e.g. "AI is changing the world" think pieces, "here is a list of 50 tools" roundups). The filter is the part that keeps the blog coherent. The filter is the part I would not change.3. The Gumroad landing pageThe [shop page exists from day one, and the shop page is the part that gives the blog a real reason to exist beyond "build an audience." The shop page is also the part that gives the CTA in every post a real destination, instead of a "subscribe to my newsletter" CTA that has no value on day one. The shop page is the part I would not change.4. The 4-folder file structureThe folder structure (index.html, posts/, style.css, feed.xml) is the part that scales. The folder structure is the part that lets me add a post in 5 minutes (copy an existing post, edit, save, drop in posts/). The folder structure is the part that does not require a CMS, does not require a build step, does not require a server. The folder is the site. The site is the folder. I would not change this.The composite: the better launchIf I were doing this today, the launch would look like this:Buy the domain (Day 1)Set up the static site with the file structure, the style.css, the homepage, the about page, the start-here page, the shop page, the contact page, the feed.xml (Day 1)Write 10 narrow posts (automation how-tos, all in the same voice) before announcing the site (Days 2-30)Announce the site publicly with the cadence commitment (Day 31)Post 2-3 times a week, every week, with the same template (Day 31 onwards)Expand to prompts, behind-the-build, and releases after the first 30 posts (Day 90)Audit the cadence and the topic mix at the 6-month mark (Day 180)That is the better launch. The better launch is 1 month of pre-work, 6 months of steady posting, and a topic expansion at month 3. The better launch is the part I did not do. The better launch is the part I would do if I were starting over today.What this post is forThis post is for anyone who is considering starting a blog. The post is the part that gives you the hindsight before the launch, which is the part I did not have. The post is the part that says "here is what to skip, here is what to keep, here is the order I would do it in." The post is the part that takes 6 minutes to read and saves 6 months of trial and error.The post is also for future-me. Future-me is going to start the 6th blog, or the 7th, or the next project. Future-me is going to want to know what the 5th one got right and wrong. The post is the part that documents the 5th one, and the documentation is the part that makes the 6th one better.What I would tell past-meIf I could go back and tell past-me one thing, it would be: the 4 posts and the WordPress install were not a launch. The 4 posts and the WordPress install were a draft. The draft is the part I was treating as a launch, because the draft was what I had. The draft is not a launch. A launch is a domain, a static site, 10 posts, a shop page, a feed, and a public cadence commitment. The launch is the part I did not have for 14 months, and the 14 months were the cost of not having it.I would also tell past-me: the voice is the part that matters most. The voice is the part that makes the blog different from every other blog on the topic. The voice is the part that turns a generic post into a post that sounds like a real person wrote it. The voice is the part that the prompt pack enforces, and the prompt pack is the part I would not skip, even on day one. The voice is the part that compounds. The voice is the part I would protect.Who this is forAnyone who has a blog and is curious about what they would do differently. Anyone who is starting a blog and wants the hindsight before the launch. Anyone who is in the middle of a blog and is wondering if the choices they made were the right ones. The answer is: probably some yes, some no, and the post is the part that tells you which is which.The honest partThe 6 things I would change are real. The 4 things I would keep are real. The composite launch is the launch I would do today. The composite launch is also the launch I did not do, and the not-doing is the part I am not proud of. The not-proud is the part I want to capture, because the not-proud is the part I would tell past-me, and the telling is the part that makes the next launch better.The 5th blog is not perfect. The 5th blog is a real blog. The real blog is the part that has 26 posts, a small audience, a working shop page, and a steady cadence. The real blog is the part I would not trade for the perfect blog I did not start. The real blog is the part I am proud of. The real blog is the part I want to keep shipping.Want the system I used to ship the 5th blog?The Blog Writing Factory is the system I used to write the 26 posts on this site. The system is the part that removes the 30 minutes of decision-making that killed the first 4 blogs. The system is the part I would use on the 6th blog, the 7th, and the next project. The system is the part that makes the writing happen.Buy on Gumroad. $14

Send me the rough edges if you try it. I read every message.