ViralHunt

I went all in on programmatic SEO and I want to share what happened, because the results were not where I expected them.

First the part I got right. We track viral posts across TikTok, Instagram, X, Facebook, Pinterest, Bluesky, Mastodon, Douyin, Tumblr, Reddit and Hacker News, and every one of those posts has a timestamp, a community, a hashtag set and sometimes a song attached to it. I was sitting on all that and only using it inside the dashboard. So I started turning it into public pages.

Ten pages that answer "when should I post on X platform", built from when the posts that actually went viral were published, refreshed daily, in four languages. Pages for which subreddit performs best for a topic. Same for Bluesky communities. Top TikTok profiles, and here is the part Im proud of, not just the profile list but the hours those accounts publish at, because we can see the pattern across thousands of their posts. Trending hashtags so you know which ones to use this week and not last month. And the songs that are pushing reels and TikToks right now.

Im gonna say something that sounds arrogant and I dont mean it that way. For this specific category, I think our data is better than BuzzSumo. BuzzSumo tells you what got shared. SocialBlade tells you who the big accounts are. Neither one tells you what time those accounts posted, or which community it landed in, and that is the part you can actually act on tomorrow morning. Ive paid for BuzzSumo, Hootsuite, Buffer and SocialBee for years and none of them ever gave me that.

So the pages are working. Traffic is up 1k and the discovery pages, the data ones, are the ones bringing it.

Now the part that is not working. The buyer intent pages, the "best tool for X" type, are barely moving. And those are the ones I built expecting them to convert. Turns out the pages that bring people in are the ones where Im just giving data away, and the ones where Im asking for the sale are dead.

So Im changing the bet. Ive noticed the keywords that convert best for us are comparison ones, stuff like "buzzsumo vs ahrefs", so Im going hard on alternatives and comparison pages instead of buyer intent. If people are already shopping, I want to be in that conversation with real numbers instead of trying to create the intent myself.

Two things Im considering next and Id like input from people who have done it.

Paid ads. Ive never run PPC in a category where Hootsuite and Buffer are bidding. Is it suicide for a bootstrapper at my size, or is there a corner of it that works if you go narrow enough on long tail?

Directories. Im sure I have a good product and almost nobody knows it exists. Which directories are actually worth the effort in 2026 and which ones are just a backlink you pay for? Im looking for the ones that send real signups, not the ones that send a spike and nothing after.

Currently at $84 MRR. Still a long way from the 5K that lets me stop freelancing, but the traffic side finally has a direction.

I started building ViralHunt ( https://viralhunt.io ) in public a few weeks ago and honestly the tool is helping me more than anyone right now. Trying to bootstrap this out and I will share you my experiences.

I designed this for my specific uses to manage multiple social networks, this is something that I wished it existed since years ago. I am a DevOps and always wanted to code something like this, but it would have taken me months and I needed to pay the bills, the thing now is that AI is making this easier for me to code, manage, deploy, etc.

I use my own special software every morning with my team. We track special topics with the radar, we decide everything on cards, what becomes an article, what is just an image post, what deserves a video script. My 3 employees work the whole flow in one place...

Before this we had everything scattered in Trello, Google Docs, Whatsapp chats, Buffer, and experimenting others like Notion and so. Does this resonates you?

Another thing that Ive been implementing and works great is the API and MCP to connect my Agent like OpenClaw, which tracks topics for me and suggests content with HTML/CSS templates ready, I just accept and do modifications. Just adapted Douyin and Mastodon too, translated all the Chinese content and turns out their trending page is full of educational formats that dont exist in the west yet.

Who I think this really helps:

1.- Entrepreneurs that need inbound content for their business, the radar shows what is already gaining traction in your niche so you dont create blind.

2.- Social media agencies, multi brand, multi network, one dashboard. If you manage clients you know this pain.

3.- And specially content creators like me that want to raise the monetization on their networks. Catching a trend 2 days early is literally money.

Im at $87 MRR, building this in public, all feedback is welcome. Link in the first comment. Bootstrapping without needing of investors and so, right now the traffic is going well with programmatic SEO and I will keep sharing my journey, my realistic goal for December is $1k MRR, first stop on the way to the $5k that lets me go full time on my own products. Hopefully it increases.

Im trying to do it without investing on ads, just building in public.

What do you use today to decide what content to create? Real question, Im collecting workflows.

https://www.youtube.com/watch?v=aIGZXiCaztQ

Douyin is TikTok's Chinese sibling. Same parent company, walled off from the west. Most tools skip it because the API situation is painful and everything is in Chinese.

We pull the trending videos and auto-translate titles and hashtags. 108,000 tracked in the last 3 days.

Since I create cultural, science and tehcnology content, Douyin is the proper network for this, misteriously it has more educational videos than Tiktok that is mostly used in the western. Why that could be?

On Douyin, roughly 1 in 4 trending videos was educational or cultural.

A forensic scientist reviewing movie autopsies.

Traditional glassmaking.

An AI game theory explainer.

Ran the same pass on TikTok's trending sample: low single digits. And half of those were false positives (a Messi goal is not education, no matter what my classifier thinks). People doing dances, people farming aura, lol.

Same company. Same format. Completely different diet.

I won't pretend I know why. Regulation, culture, algorithm choices, probably all three mixed.

But here's why I care: I run a science media brand, 8M followers. Educational content IS my niche. TikTok's trending page was always useless to me, pure entertainment top to bottom.

Douyin's trending page turned out to be a catalog of proven educational formats that haven't crossed over yet. The median trending video there has 1.1 million likes. That's the bar.

So the feature I added "for completeness" became the one I use most myself.

The roadmap keeps getting corrected by usage. Funny how that works.

Anyone else mining the non-English internet for formats or product signals? Curating this kind fo content from Asia is kind of a gold mine. Doing your own version and so. Did you know Douyin existed?

Bluesky is now a full network, not just another checkbox. You can schedule, publish and even post full threads to Bluesky from the same composer you use for everything else, and on top of that we track it like the big networks: feed catalogue, hashtags, categories, a trending radar and growth curves from day one. Fun fact: Bluesky is the only network we can sweep entirely from the server. No browser extension, no fighting anti-bot walls. Their open API feels like the internet 10 years ago, in a good way.

Programmatic SEO turned out to be our best marketing channel. Instead of fighting for 5 competitive keywords, we generate pages from data we already collect anyway: a Reddit community directory ranked by growth upside instead of raw size, a Bluesky feed directory, creator pages, hashtag pages, even trending sounds from Instagram Reels. Everything refreshes on cron. It's early but traffic is already picking up and the nice part is it compounds while I sleep.

Other stuff that shipped, rapid fire:

- ViralHunt is now an MCP server. Your AI agent (Claude or whatever you run) can query trends and schedule posts on its own. This is the one I'm most excited about.

- Comment to DM automation for Instagram and Facebook: someone comments your keyword, they get the DM, you see the response rate. Like Manychat but cheaper.

- Multi-brand publishing to 11 networks from one composer.

- Analytics on what you published, not just on what you found.

- A public methodology page, so you can check how we rank things instead of trusting us.

The unglamorous part nobody posts about: we spent whole days fixing 757 Bing 404s caused by relative internal links, rebuilding the video upload flow because large files were dying before our code even ran, and moving media storage off the app server. Zero of that is fun to post about. All of it was necessary.

One thing I want your honest opinion on: AppSumo.

I'm considering a lifetime deal, but a limited one. The lifetime tier would get you the trend radar and some limited daily queries, and if you later want unlimited scheduling and the full thing you can move to a subscription with a permanent discount as an early backer. The goal is more marketing reach than revenue, I know the economics of LTDs are ugly.

If you've actually run an AppSumo campaign I want the real story: did those users stick around? Did any of them ever upgrade? Or did they just triple your support tickets? Trying to decide if the exposure is worth it.

Building this in public, so I'll keep posting updates here. viralhunt.io if you want to poke around, we have a Free trial with limitations so you check the whole thing. Will be very grateful to read feedbacks and ideas to improve.

I went quiet here. I was building a lead generation tool, and then I stopped posting about it. Here's what actually happened, because the lesson is more useful than the update.

I didn't quit. I got honest about priorities.

For almost 15 years I've been building a media network in Spanish and English. It isn't a side project , it's my life's work, and it pays for itself through content.

The lead gen tool was the opposite: a product for a market I was still learning. Both needed full attention. Only one of them was the thing I'd regret not betting on. So I parked the lead gen SaaS, still alive, not deleted... and went all in on the network.

Here's the part that's relevant to this community.

Going all-in meant fixing my own workflow before anything else. Running pages at that scale with a team looked like this: ten tabs open every morning hunting for what's starting to go viral, ideas in Trello, assignments in one chat, feedback in another, and a scheduler at the end that never talked to any of it.

So I built something internal to fix my own pain. A live radar that tracks what's going viral across 8+ networks, feeding a kanban board where my team curates, fact-checks, rewrites and schedules to every network from one place. My AI agents work that same board through MCP, as members with their own access.

Somewhere in there I realized I'd built the app I spent 15 years wishing existed. I had paid Hootsuite, then Buffer, BuzzSumo, SocialBee, waiting for one of them to build it. None of them did. Now it runs my whole network daily, so I put a price on it. It's called ViralHunt.

The lesson I'd offer anyone juggling two projects: the question isn't which one has the bigger TAM. It's which one you'd still be doing in ten years. Answer that honestly and the pivot stops feeling like failure.

Mine stopped being a pivot at all. The product came out of the life project, so every improvement serves both. That's the part I didn't expect. I'm not splitting attention between two things anymore, I'm compounding it into one.

I'll share the journey here from zero: real numbers, first customers, what flops. Happy to answer anything about growing pages organically too, that part I actually know.

Question for the room: has anyone had a product fall out of your main thing, instead of the other way around? And once it did, how did you split building it against selling it?