One month of daily posts: what shipped, what worked, what is next

Thirty posts in thirty days. The last one is the one you are reading. Time to look at what happened.

The premise was that the existing weekly content calendar was leaving real gaps: persona breadth, long-tail search intent, plugin-feature surfaces we had not explained yet, and a small founder/retro thread. The 30-day daily plan was the experiment. One post a day, every day, May 24 through June 22.

Here is the honest retro.

The numbers

I am going to keep these directional rather than precise, because I want to write about what to do next more than what we hit on the dashboard. Traffic is up. Newsletter signups are up. Three of the posts (the ones I will name below) drove most of both.

The shape of the numbers across the month: a slow build for the first week, an inflection point around post 10 or 11 (when the link-graph started compounding because every new post linked to several earlier ones), and a plateau in the last week where the day-over-day numbers stopped surprising me. The compounding effect is the part I will keep believing in for the next 30 days.

Plugin installs are also up, but the attribution is hard. The blog is one of several surfaces driving installs; isolating its contribution would take attribution tooling I have not set up. The directional sense: the blog is doing real work in the funnel; the magnitude is in the range I would have hoped for, not better or worse.

Three that landed

The plugin pile-on (June 7). The Sunday-read slot, framed as “five plugins one DesignSetGo app can replace.” This one got the most LinkedIn engagement from non-developers in the audience: agencies and consultants who saw their own plugin lists in the description. The lesson: persona P3 reads strategy posts and shares them, more than I expected. I had been thinking of strategy posts as P2-facing.

If it breaks, your site is fine: the safety pillar with receipts (June 14). This one got picked up on Hacker News (modest, not viral). The three concrete scenarios with mechanisms is the shape that travels. The “marketing claim with receipts” framing seems to be a format that the HN/skeptical-developer audience rewards.

The four boring parts of WordPress your vibe-coded app secretly needs (May 25). A Monday morning concept post that pivoted on a single framing (auth, content modeling, user accounts, admin UI). It drove the most direct “I get it now” replies. Several of the replies turned into newsletter signups. I think this post will end up in the long tail as one of the highest-referrer posts, because the framing is reusable for other channels (social cuts, podcast intros, intro decks).

Three that did not

DesignSetGo and WPML: how multilingual sites read through the bridge (June 15). A useful post for the specific audience that needs it. Almost no one is in that audience. I knew this going in; it was the right post to write for the support burden it removes, not for the traffic it drives. In retrospect, this should have been documentation, not a blog post.

The CSP allowlist, explained (June 8). Same shape as WPML: useful, technical, narrow audience. The right people read it. Not many people read it. Same retrospective: this is documentation that ended up on the blog because the writing voice was easier to draft than the docs voice.

Build a DesignSetGo app from inside Claude.ai: the Connector walkthrough (June 16). I expected this one to land harder. It did not. My read: the audience that already uses Connectors in Claude.ai is small, and the audience that does not is not motivated by the “you can do it without leaving the chat” framing. The tutorial is correct and the path works; the framing did not find the audience. Worth trying a different angle next month.

Three surprises

Comparison posts overperformed. The Thursday slot (CFF, Gravity, Bubble) was where I had the lowest expectations and the highest engagement. People search comparisons. The comparison posts have a much longer tail than the tutorials. I will weight Thursdays more aggressively next month.

The agency-focused posts underperformed. The freelancer-pricing checklist, the consultant’s three apps, the in-admin dashboard tutorial. I expected these to compound through agency networks. They did not, much. My hypothesis: the agency audience does not consume blog posts the way I do. They consume Slack messages and Twitter threads. The blog posts are right; the distribution is wrong. Next month I will treat agency-focused posts as social-first, with the blog post as the durable backup, rather than the other way around.

The vertical posts underperformed in raw traffic but overperformed in conversion. The mortgage calculator post and the TDEE post brought less traffic than the strategy posts, but the visitors who arrived from them were dramatically more likely to install the plugin. The long-tail search intent (“wordpress mortgage calculator,” “wordpress tdee calculator”) is converting at much higher rates than the broader strategic queries. The next 30 days should have more of these, not fewer.

Three things I would do differently

Cut the WPML post. Or, more honestly, make it documentation. Not every “we should explain this” need is a blog post. The blog is a channel; the docs are a reference. I conflated them for that post.

Pair tutorials with screen recordings. Several of the tutorials would have been twice as effective with a 30-second screen recording of the result. I shipped them without because the recording is its own production cost. Next month I will record three or four short clips upfront and use them across the posts that benefit.

Plan the founder posts a week ahead. The Friday and Sunday founder posts are higher-quality when I have had a few days to think about them. The ones I drafted the morning of were noticeably weaker. Easy fix; I will keep a Friday-and-Sunday backlog two weeks deep.

What I am not doing differently

A few things I considered changing and decided to keep:

Daily cadence. It works. The compounding from the cross-link graph is real. The audience expectation, after a month, is “there will be a post tomorrow.” Switching back to weekly would lose that.

Long-form. No 300-word posts. The pressure to write shorter, more frequent posts is real and several smart people have made the case for it. My read: the long-form posts compound. The short-form posts get a spike and disappear. I will keep the 800-2000 word targets.

Founder voice on Fridays and Sundays. This is the part of the schedule that keeps the blog from sounding like a docs site. I will keep the rotation. The agency-pricing-checklist post (June 19) is a good example of why: the same content in third-person voice would not have hit as hard.

The next 30 days in two paragraphs

I am going to keep the daily cadence and the weekly rotation pattern (Monday concept → Tuesday P2 tutorial → Wednesday P1 tutorial → Thursday comparison → Friday founder → Saturday vertical → Sunday long read), with adjustments based on the retro: more comparison posts (Thursdays compounding), more vertical posts (highest conversion), fewer agency-focused posts on the blog (move them to social-first), and a documentation pass on the technical-narrow posts so they live in docs/ instead of blog/ going forward.

Topical focus for the next 30: the WooCommerce vertical (we have 12+ example bundles in examples/ that have not been written about); a second pass at the search-intent slots (this month’s mortgage and TDEE worked; I want a pricing-page-for-X, a quiz-for-X, and a booking-widget-for-X each in the next month); and the founder thread continuing to explore the agency pricing model since the June 19 post raised more questions than it answered. The new 30-day calendar will be in the repo within a week.

The shorter version

Daily cadence works. The cross-link graph compounds. Comparison posts and vertical posts are the highest-leverage formats. Agency-focused posts belong on social, not on the blog. Pair tutorials with screen recordings. Next 30 days, with adjustments.

Thanks for reading any of these. If you have something you want me to write about in the next month, send it.

Further reading

• The 30-day daily content calendar: the plan this month executed against.
• SOCIAL-STRATEGY.md: the social channel that pairs with the blog.
• CONTENT-CALENDAR.md: the weekly strategic calendar that runs alongside the daily one.