Two hours a week. That is the entire content budget a solo founder needs to build an SEO engine that compounds: one hour to write, thirty minutes of keyword research, thirty minutes to publish and review.
The reason that works, and the reason most founders never see it work, is that SEO does not reward the person who publishes the most. It rewards the person who publishes consistently, in one place, about one problem set, for long enough that Google stops treating the site as an unknown.
Here is the loop, the checklist, and the timeline you should actually expect.
Consistency beats volume, and it is not close
Two sites publish 24 posts. Site A ships all 24 in a six-week sprint, then goes quiet for five months. Site B ships one a week for 24 weeks. Same word count, same effort, wildly different outcomes.
Site B wins for three concrete reasons:
- Crawl frequency follows publishing frequency. Predictable fresh URLs train Googlebot to come back on a shorter interval. Faster discovery means faster indexing, which means the feedback loop that tells you what is working runs weekly instead of quarterly.
- Topical authority accumulates, it does not spike. Twenty-four posts around one problem set, published steadily and linked to each other, read as a site that owns a subject. Twenty-four posts dumped in six weeks read as a content dump.
- You get 24 chances to learn instead of one. Every post you publish changes what you target next, but only if you are still publishing when the data arrives. Site A's entire sprint was planned before a single result came in.
The sprint-then-silence pattern is the most common failure mode in startup SEO, and it is usually a symptom of treating content as a project instead of a process.
The two-hour weekly loop
Block the same two hours on the same day every week. Protect them the way you would protect a customer call. The loop has three parts.
30 minutes: pick the target. Not brainstorming. Picking, from a list you already built. Open your keyword map, take the top unwritten row, and confirm it still looks winnable by checking who currently ranks for it. If you do not have a map yet, build one before you write anything: query, page, intent, CTA, twenty to thirty rows deep. A full quarter of topics, decided once, means you never spend a Tuesday morning wondering what to write.
60 minutes: write one draft, start to finish. One sitting, no research tangents, no editing while you write. You already did the research in the first block. The structure that keeps this to an hour:
- Open with the concrete number or result. First two sentences, no throat clearing.
- Three to six H2 sections, each answering one question the searcher has.
- One example that is real, from your own work or a documented case.
- A closing CTA that matches the intent of the query.
If a post takes three hours, the target was too broad. Narrow it and the draft gets faster.
30 minutes: publish and review. Title tag and meta description written deliberately, not auto-generated. Two internal links out to older posts, one internal link in from a post that already gets traffic. Then inspect the URL in Search Console and confirm it is discovered. Last five minutes: check last week's post and note whether it is indexed yet.
The keyword research block and the writing block should never happen in the same hour. Deciding what to write and writing it use different brains, and merging them is the single biggest reason a "one post a week" plan quietly becomes one post a month.
Why weekly specifically, and not daily or monthly
Daily publishing is a trap for a solo founder. Quality falls off, the pages get thin, and thin pages actively hurt you: they dilute topical focus and, if you are running programmatic pages alongside them, they make the indexation problem worse rather than better.
Monthly is too slow to build a habit or a crawl pattern. Twelve posts a year is not a content engine, it is an occasional newsletter, and the gap between publishing and feedback is so long that you never learn anything from it.
Weekly is the interval where all three constraints line up: it is sustainable next to a full-time build, it is frequent enough that crawlers re-learn your schedule, and it produces roughly 50 assets a year. Fifty pages, each targeting one intent-qualified query, is a real surface area.
The month-four trap
Here is the part that decides whether any of this works: almost nobody quits at month one. They quit at month four.
The timeline for a new domain is brutally consistent. Months one to three are the invisible phase: pages get indexed, almost nothing ranks, traffic is effectively zero. Months four to six is where the first real signal shows up, usually a handful of long-tail terms landing on page two, then page one. The compounding, where older posts keep climbing while new ones stack on top, generally starts around month six.
Month four is the exact moment where you have done the most work and seen the least return. The graph is flat, you have shipped 16 posts, and quitting feels rational. It is not. It is quitting one month before the data would have told you whether the strategy works.
Two things make month four survivable:
- Track leading indicators, not traffic. Impressions, indexed page count, and average position all move well before clicks do. If impressions are climbing and position is improving, the system is working even when the traffic number says zero.
- Pair content with something that moves faster. Content compounds slowly because authority builds slowly. Directory submissions lift Domain Rating on a 3 to 6 month clock, running in parallel, which means the posts you are shipping in month two land on a stronger domain by month five.
When we grew promptbuilder.cc from a fresh domain to 50.1K monthly clicks, the mechanism was not one viral page. It was pages stacking on a domain whose authority was rising at the same time, week after week.
The weekly checklist, copy this
Same day, same two hours, every week:
- Open the keyword map, take the top unwritten row
- Check the live SERP for that query, confirm it is still winnable
- Draft in one sitting: number up top, 3 to 6 H2s, one real example
- Write the title tag and meta description by hand
- Add two internal links out, one internal link in
- Publish, then inspect the URL in Search Console
- Check last week's post: indexed yet?
- Once a month, instead of a new post, refresh the best page sitting on position 6 to 20
That last line matters more than it looks. Somewhere around post 12, refreshing an existing page that is already close to page one will beat writing a thirteenth new one. The cadence stays weekly, the output type changes.
Plan the quarter before you write the first post
The reason a weekly cadence survives contact with a real founder schedule is that the hard decisions are made in advance, in bulk, once a quarter.
Sit down for two hours, once every 13 weeks, and produce a table with one row per week: date, working title, slug, primary keyword, a three-line outline, and which CTA it points to. That is it. When Tuesday comes, there is no decision left to make, only a draft to write.
This blog runs on exactly that document. The 13-week plan behind these posts lives in the repo as a plain text file with one brief per week, and every post you are reading was written from a row in it. It is not a fancy system. It is a list that removes the weekly decision, which is the only thing standing between one post a week and no posts at all.
Consistency is not a personality trait. It is a calendar you filled in ahead of time.
Want the full step-by-step version of this playbook, plus a new breakdown every week? Grab the free guide, and follow along on X.