personal site as compounding asset — the 18-year retrospective

I started this blog in December 2007 from the college computer center in Madurai. The first post was titled Let us walk on 2008. The blog’s name was I like the way I am. The domain was on Blogger before it moved to a self-hosted WordPress, then through several hosting providers, and finally to Astro on Vercel in May 2026.

Eighteen years and 96 posts later, the compounding is real and is concentrated in places I did not expect. Here is what compounded, what did not, and what I would do differently if I were starting over.

what compounded

Three things, in order of how much they mattered.

Credentials and credibility. The single largest compounding effect of running a personal site for 18 years is that the site itself becomes a credential. When someone Googles my name, the first three results are: the blog, my LinkedIn, my X profile. The blog is the longest-running of the three and the only one I fully control. Anyone evaluating me as a hire, a collaborator, an advisor, or an investor lands on the blog and forms an opinion within five minutes of reading.

That opinion has been overwhelmingly net-positive. The hires, advisory invitations, conference invitations, podcast appearances, and book deals over 18 years that came directly from “I read your blog” — that list is in the hundreds. Each individual instance was small. The cumulative effect was enormous.

Relationships with strangers. The reach of a personal blog is not measured in monthly visits. It is measured in the people who reached out after reading something specific I wrote. The 2018 ML book post led to three job offers and four book deal inquiries. The 2020 culture-building post led to two startup advisor seats. The 2024 brand loyalty post led to a year-long consulting engagement.

The conversion from “stranger reads my post” to “stranger reaches out” is roughly 0.01-0.05% of readers. The conversion from “stranger reaches out” to “meaningful collaboration” is much higher — maybe 10-20%. The cumulative number of strangers turned into collaborators across 18 years is enough to materially shape the trajectory of my career.

Writing skill and clarity of thinking. I am a better writer now than I was in 2007. I am also a better thinker. The two are causally linked. Every blog post I have written forced me to take a vague thought and make it specific enough to defend in public. That discipline accumulates. The me of 2026 can write a 3,000-word essay in two hours that would have taken the me of 2007 two weeks.

The thinking part is the underrated benefit. Writing a blog is a thinking exercise disguised as a publishing exercise. The output (the post) is the byproduct. The internal change (clearer thinking) is the real return.

what did not compound

Two things, despite my hoping they would.

Direct ad or affiliate revenue. I have made approximately ₹0 in direct revenue from the blog over 18 years. I ran ads briefly in 2009-2010 and the revenue was below the threshold of being worth managing. I never built a paid newsletter, a paid course, or a paid product directly tied to the blog. The blog has been net-negative in direct cash flow (hosting costs minus ad revenue) for almost all of its existence.

If the goal had been to make money directly from the blog, the strategy would have been wrong. The blog has paid for itself dozens of times over through indirect effects, but never directly. The direct-monetisation path was always a mirage.

Search traffic at scale. The blog has never had more than ~5,000 monthly organic visitors at peak. For an 18-year-old site, that is unimpressive. The reason is that I never optimised for search. I wrote what I wanted to write, in whatever frequency I wanted. The SEO playbook (topical authority, internal linking, keyword targeting) was something I did not apply systematically until 2025.

If you measured the blog by search traffic, it would look like a failure. The actual value came from a different metric — quality of inbound connections, not quantity of inbound visitors. The two metrics are weakly correlated. Optimising for the quantity metric would have produced a different blog with worse outcomes on the quality metric.

the 2022 compromise as stress test

The September 2022 webshell compromise on my old WordPress install was the most disruptive event in the blog’s history. Seven empty PHP files dropped at the server root, the site briefly serving suspicious content before I caught it, an entire weekend cleaning up.

The site survived. The traffic recovered within a month. The reader relationships were unaffected. The compromise itself turned out to be a non-event in the long-run impact on the blog’s compounding.

What it did do was force me to confront the architecture. WordPress was the wrong tool. The compromise made that undeniable. Four years later I migrated to Astro. The blog continued to compound through the migration with no visible discontinuity to readers.

The lesson: the substrate of the blog matters less than the writing itself. Move hosts, change platforms, switch from WordPress to Astro — none of these affect what the blog actually is. The compounding survives platform changes if the content survives them.

the 2026 migration as renewal

The May 2026 migration to Astro was the largest infrastructure change in the blog’s life. Markdown files in git, version-controlled, deployed to Vercel’s CDN, zero ongoing maintenance.

The migration produced one immediate benefit: I started writing more frequently. The friction of WordPress (logging in, dealing with the editor, worrying about plugin updates) had been suppressing my publishing rate for years. Without measuring it, I had assumed I just had less time. After the migration, I shipped more in the following four weeks than I had in the previous four months.

The substrate matters in this way. The right substrate makes the desired behaviour easy. The wrong substrate makes it hard enough that you do less of it.

what I would do differently

If I were starting today, with the knowledge of 18 years of running this blog:

Start on a static site generator. Skip the WordPress phase entirely. The maintenance cost of WordPress was the single largest hidden tax on the blog’s first 18 years. Astro, Hugo, or Eleventy would have produced the same content with one-tenth the operational overhead.

Optimise for the rare reader, not the average. The valuable readers are the rare ones who reach out, hire me, become collaborators. The average reader is a transient visitor who bounces. Most blog advice is calibrated for the average reader (more posts, shorter posts, SEO-optimised titles). Calibrating for the rare reader means deeper, more honest, less optimised writing.

Stay narrow. I wrote about everything for 18 years. The compounding would have been larger on a narrower focus. The blogs that produce 10x the credential effect at the same word count are the ones that wrote 100 posts about a single topic, not 100 posts about 100 topics.

Trust the compounding even when it looks slow. Most years on this blog looked unsuccessful by any short-term metric. The compounding only becomes visible at year 5, year 10, year 15. Most blogs do not survive long enough to see it.

the meta-lesson

A personal site is a compounding asset on a 10-20 year horizon. The compounding is invisible in the first 3 years, modest at year 5, real at year 10, transformative at year 15.

Most people quit before year 5. The ones who do not are the ones for whom the blog becomes a structural feature of their careers, their relationships, their thinking.

If you have a blog, keep writing. The first 5 years are not the return. The first 5 years are the investment. The return starts to show up in year 6 and accelerates from there.

If you do not have a blog, start one. The decade-from-now version of you will be glad. The decade-from-now version of you will also be the only one who can verify whether this was true.

I am writing this from year 18. I can verify that it is.