The last stall in the market never sells, unless?

AI is adding a new layer to a world where software has been packaged and sold as a commodity for a long time. The old-school players have formed a complex ecosystem with deep distribution networks and user bases — something that would take years to build and is extremely difficult to achieve in the short term unless it’s truly an idea that cannot be duplicated.
Press enter or click to view image in full size

However, the growing concern among these players is that every day, a new product or clone pops up that does and offers exactly the same thing to the same user base, but at a dirt-cheap cost — many a times free. And mind you, there is absolutely no quality degradation in these offerings. Most of the times there are highly concentrated, superior products being offered.

I started with same idea, i.e., if I build a superior product and make it free, people will come to use it in flocks. I was in for a suprise, entering a brutal cutthroat world where before I stepped in, there were already hundreds with same idea. Now this was a fairly simple idea based on something I personally struggled with while looking for a job change. At that point — and this was 3–4 years ago — there was hardly any site that would let you build and export your resume for free unless they were scraping your data or locking you behind a paywall on the verylast screen after promising a free resume. The free part, later that my sorry soul would realise, included just building it.

It took me 7 days to create my own resume builder with frontier model and host it on a fancy site. Another 7 days went into SEO, Reddit posts, and Product Hunt to generate some noise around it, and users started tickling in. For months, I continued adding bits and pieces. The stream of users remained consistent, averaging 4–5 per day, and that was wonderful for me — someone who never had the time or patience to build something that would take AI hours today, but in the old coding days would have taken months to build and debug, and even then you would still be far from perfect.

I am now over 6 months in, and there has been no increase in the number of users — it has remained stable at 4–5 users a day, mostly from recommendations by the very AI models I used to build it, and a few others from the crumbs I dropped around the internet. The site was mostly complete, with everything available for free, along with a bunch of tools and supporting pages designed to attract traffic through SEO.

Write on Medium
Just one problem, though — Google refused to index me.

It did index me first when the site has just 3 pages — A landing page, an editor and FAQ. It probably felt there were already loads of resume builders indexed and that there was no distinct advantage my tool was giving users except being another clone. So now I am in the waiting game. It is almost certain that Google will never index it, and I acknowledge that fact. Still, for those 4–5 users who keep coming every day, I will continue to keep it running for as long as I can. Not that I didn’t try to source other channels for users but that is probably a topic for another article.

What it did teach me were few valuable lessons in my attempt to build something for the audience:

1. You need to have an audience or atleast base of it when you are starting.
2. Make sure the market is not overloaded with what you are building, your offering should be unique enough to not become another product one stumbled upon.
3. The last stall in a vast market gets the fewest customers — unless you have very strong distribution. ( Reiterates on my first point) In the world of today and tomorrow, what will sell is the ability to reach many people with whatever you have to offer. Building and selling the product is no longer and will never again be the bottleneck.