Three weeks ago we published our 200th blog post. Not over three months, not with a team of ten people working full-time, but as a coordinated stockpile strategy that required careful planning, a drip publishing mechanism, and an honest decision to compete for topical authority in answer engine optimization.

This is what we learned.

Why We Did This

The short answer: we wanted signal in the AI search space. Fast.

By early 2026, it became clear that answer engines (Perplexity, OpenAI search, Google’s AI overviews) were pulling citations from published content. Google increased crawl frequency to content that ranked well for broad informational queries. The models themselves were being trained on fresher content more often. The playing field had shifted from traditional SEO (rank for keywords, capture traffic) to topical authority and publication velocity (own a subject area, get cited).

We run Instant Press Co. because we believe in the power of earned media for B2B brands. We help clients place stories in Forbes, VentureBeat, TechCrunch, and Entrepreneur. But we also write a lot. We have the infrastructure. We have writers. We have templates. So we asked: what if we took our own advice?

What if we built topical authority on PR, publication placement, and AEO itself by publishing 200 pillar posts in rapid succession? Not as a growth hack. As a test of our own playbook.

We decided to do it. We’d write the posts. We’d schedule them. We’d release them at a rate of 10 per day over three weeks. We’d measure what happened to our indexation, our crawl frequency, our visibility to AI models, and our ability to be cited.

How We Planned It

The planning phase took a week. We didn’t guess at topics. We started with Semrush.

We pulled keywords across four categories: PR and media relations, AEO fundamentals and strategy, publication placement and earned media, and practical guides on content, outreach, and pitching.

The filter was strict: Keyword Difficulty under 35 (no high-authority fights), search volume above 30 (not vanity traffic), and relevance to what we actually do. We excluded brand keywords and queries about our specific competitors. We wanted to own the water, not just a single pond.

Semrush returned about 300 viable keywords. We mapped the strongest 200 to frontmatter data: one keyword per post, organized by category and sub-topic.

Then we clustered. AEO posts linked to other AEO posts. PR fundamentals fed into advanced PR strategy. Guides connected to deeper content. We created a topic map in Markdown (a simple index) so writers could see which posts needed to link to which other posts. No broken clusters. No orphan content.

We assigned write dates based on category distribution. We wanted 60 PR posts, 50 AEO posts, 50 publication posts, and 40 guides. We spread them across writers (we have six on staff) and set deadlines. Each writer got 30 35 posts assigned, with a staggered schedule so we had batches of finished work coming in every few days.

Parallelization worked. We had the first batch of 60 posts written and reviewed within eight days.

The Drip Publishing Mechanism

We use Astro for the site. Static generation. Markdown files in /src/content/blog/. Each post has frontmatter with a publishDate field.

The key insight was simple: we could schedule the publish dates in the frontmatter, build the site multiple times per day, and only include posts with a publishDate less than or equal to the current date.

Here’s the logic in the Astro build:

  1. All 200 posts exist in the repo as markdown files.
  2. Each post has a publishDate (ISO 8601 format).
  3. The build filters out any post with a publishDate in the future.
  4. When a build runs, only published posts appear in the site.

We set up GitHub Actions to run four builds per day: 6 AM, 12 PM, 6 PM, and midnight UTC. This meant posts could be scheduled at any time, and they’d go live on their scheduled date without any manual intervention.

The publishDate values were staggered: October 1 at 6 AM for the first batch of 10 posts, October 1 at 12 PM for the next 10, and so on. We spread 200 posts across 20 days. Ten posts per day, every four hours, like clockwork.

Vercel picked up the builds and deployed them. No webhooks needed. Just GitHub Actions pushing to the main branch on schedule.

The whole system cost us nothing extra. It runs on infrastructure we already had.

What Worked

1. Indexation speed exploded.

Google’s crawler showed up on day one and never left. By day five, we had indexed 50 new pages. By day ten, 120. By the end of the three weeks, all 200 were in the index.

Compare that to normal publishing pace. We’d release one or two posts per week under normal circumstances. It takes 4 6 months to build 200 posts worth of indexation signals at that rate. We did it in 21 days.

Google’s crawl frequency to the site jumped from an average of 18 crawls per day to 80. The bot came back constantly, picking up new pages, following the links between posts.

2. Internal link structure meant something.

Because we clustered topics and linked between related posts, the crawl pattern wasn’t random. Google followed internal links aggressively. Posts about “AEO and topical authority” linked to “how to structure AEO content” linked to “topical clusters for AEO.” The bot followed these paths and indexed all of them within 48 hours.

We didn’t use anchor text keyword stuffing. We wrote real links that made sense in context. But the crawlers picked up the semantic relationships anyway.

3. Entity consolidation happened fast.

We noticed something subtle in the Search Console data. Posts about “answer engine optimization” and posts about “AEO” started clustering together in Google’s systems. The entity for Instant Press Co. strengthened. Posts were no longer isolated; they formed a constellation around our brand and expertise.

This is where topical clusters for AEO becomes more than a theory. Google can see when you own a topic cluster, and it rewards the signal.

4. AI models noticed quickly.

Within two weeks, we started seeing citations in Perplexity responses. We’d ask about “answer engine optimization strategies” and Instant Press Co. URLs appeared in the citations. This happened faster than we expected. We attribute it to the velocity signal and the topical authority consolidation.

We don’t have access to the training data for other models, but based on the Perplexity citations, they’re picking up our content.

What Surprised Us

1. Duplicate content concerns didn’t materialize.

We were worried about thin content or duplicate value across 200 posts. The KD<35 filter meant we were writing about simpler, more specific queries. Many of the posts are genuinely useful on their own, not just variations on a theme.

We did find a handful of topics with real overlap (two posts both touched on “media relations strategy”). We merged those before publishing. But we didn’t face the wall of duplication we anticipated.

2. Conversion actually improved.

We ran 200 pieces of content. 200 chances for someone to find Instant Press Co., learn what we do, and ask about services.

Our contact form submissions doubled during the three-week publishing window. We’re not linking to paid offers in every post, but the sheer volume of entry points into the site increased qualified traffic.

This isn’t novel. More content = more leads. But the speed of the signal was sharp. Within the first week of publishing, we saw the lift.

3. Backlinks came naturally.

We didn’t do outreach for these posts. We didn’t ask anyone to link to them. But linking happened. PR professionals found posts about media relations. Content strategists linked to AEO content structure. A few posts got picked up by newsletters.

The topical authority position plus the publication velocity made the content linkable. Quality content published fast attracts attention faster than quality content published slow.

What Surprised Us in a Bad Way

1. Maintaining quality at scale was hard.

We’re proud of the 200 posts. But we were also on a schedule. A few posts could have used another round of editing. A couple have typos we caught after publishing (fixed quickly, but still).

Quality doesn’t break at scale, but the margin narrows. If we do this again, we’d add a QA pass for every 50 posts and adjust timelines accordingly.

2. Internal linking data entry was tedious.

We created a map of which posts should link to which other posts. Manual data entry of 400+ link relationships across 200 posts is error-prone. A few posts didn’t link where they should have because we missed them in the spreadsheet.

Next time, we’d build a linking map tool or use a wiki syntax to define relationships upfront and auto-generate the links during the build.

3. The publish mechanism created a timing problem.

We scheduled 10 posts every day at regular intervals (6 AM, 12 PM, 6 PM, midnight UTC). This is good for consistency and visibility to crawlers. But it’s bad for time zones. Our audience isn’t evenly distributed across 24 hours. We should have clustered publishes around peak traffic times instead.

Results and Next Steps

It’s been three weeks since we finished the series. Here’s where we stand:

We’re continuing to publish new content, but at a normal pace. One or two posts per week. The 200-post stockpile is the foundational layer. New posts build on top of it.

We’re tracking what happens to these 200 posts over the next six months. Will they continue to rank? Will citations increase? Will the topical authority signal continue to strengthen? We’ll know more by August.

But here’s what we know now: publishing 200 posts in three weeks, backed by a drip mechanism and a topical cluster structure, moves the needle fast. It’s not a vanity play. It’s a strategy that works.

If you’re building topical authority for answer engines, velocity matters. Clustering matters. Internal links matter. Freshness matters.

This is how we proved it.

Have questions about the strategy, the tools, or the results? Hit reply. We read every message.