You may have come across the semi-popular side gig of uploading stock

6 min readJan 17, 2021

I believe that one of the hardest parts about programming is finding the problem, not the solution. We go around, trying to find a cool project for our portfolios, maybe even something that makes us a lot of money — but it’s not easy.

You can learn coding more or less easily using online resources, but I have yet to see a course on “problem finding” out there in the wild.

So let me show you a problem I discovered recently where I see common problem, a common solution and a gap where nobody has yet jumped on the chance.

First off: Why am I not building it myself?

I asked myself this question before I set out to write this post: Is this something I actually want to build? The answer is both yes and no. Why?

I can build the scripts but it is unlikely that I can whip up the billing, subscription side of things needed to turn this into a full-blown service. I wish there were boilerplates for webservices.
There is maintenance involved to keep the scripts current that would probably require multiple coders once more people than just me start relying on it. However for a hobby project and personal use you can keep it much simpler, even hard-coding your logins.
It requires user authentification for multiple sites as well as storing billing information etc. and that is something I don’t want on a mere side project (or know enough about to keep reasonably safe)
I think the concept is sound and interesting to show not jus the solution but also how to find a problem.

The problem: Lots of overhead

You may have come across the semi-popular side gig of uploading stock photography. I say semi-popular because many know it exists but few start up and even fewer keep going.

Why is that, because it’s so hard to make good, buy-worthy photography? No, because it’s not. You can take pictures with your phone, certainly with any regular camera of decent quality that are well worth buying.

You can also sign up to all (?) stock photo sites with relatively little effort and with only a modicum of required tests. I know I signed up for depositphotos at some point and had to upload three or five pictures of decent quality — all of them taken with my phone because I no longer use a real camera.

I can also take some interesting shots of city and country, cars and water, birds and prey. I can also upload them, tag them, name them, find the right category. I can do that just fine on one site, two sites, all the sites.

Do you see the problem? Taking a picture is very easy compared to the time, effort and mindnumbing work that goes into uploading them to the various stock photography sites.

The solution: within reach

If you look closer that is a job that is perfectly automizable, in fact that is something you can automate with a decently complex Selenium script.

That is something that they probably all offer API access for, if you like living on the edge it is something you could automate using AutoHotkey with mouseclicks and button presses.

The market: existent

So we have something here, I know I would pay five bucks a month for someone to take this overhead away from me. Couple it with AI picture analysis to give suggestions on tags and categories and you have a really good feature and selling point at your hands.

The point to understand about those five-dollar-a-month subscriptions is this: If you save just a single hour by using that service you are already well into the profit zone even if your time is worth minimum wage to you.

So now look at all those photographers out there who are better than me, probably better than you and who don’t yet have accounts and aren’t already making heaps of money.

Now think further, look at all those people who make reasonably good money on those sites but spend hours of painstaking work to upload them — time they could spend walking through the city and taking more pictures.

Someone out there likely hires virtual assistants on Fiverr to handle the uploading, someone else abuses their interns by clicking and silently crying their futures away.

I would build this program for exactly one customer: That last guy there, the person who thought they were getting to learn something about art and photography and instead they slave away at a desk all day. I built this tool so that they can sit at home, pay five bucks a month and get paid 200, 500 by someone else. All the others who profit from this tool are just extra.

What techniques would you need / need to learn?

Let us take a look at (one possible) tech stack that could be used to realize this project. It’s not that much, actually if you leave out the billing portion that would probably consume a lot of time dealing with stripe, paypal, intermediary payment processors.

(also there’s something for you, an easy turn-key boilerplate that lets people like me quickly set up projects like this)

I would go with Python because it’s what I know to work with web automation. You can use either Selenium or direct API access depending on which sites offer that.
For the front-end you could go with Wordpress (I have seen this work!) with multi-user-logins and file uploads. This is not how Wordpress is intended to be used and it has lots of downfalls, but you can make it work.
More likely you’d go with proper front-end choices like React or maybe one of the python-based ones like Flask or Django if you already went with python for the main functionality.
Database usage.

As I said the biggest issue here is not in the actual functionality but rather everything around it.

If you were to go with a simple GUI tool with a bit of Python, a bit of configuration files and promote your work through sites like BlackhatWorld you could turn a pretty sizable profit with a whole lot less effort.

Summary: A real world problem with a realistic solution

Some people are much better at front-end stuff than I am, maybe I have also been missing an easy turnkey solution that’s been known to people for a long time — if so please let me know because I have plenty of little scripts that I could turn into subscription services if only I knew how.

I hope this post achieved the main goal in showcasing some real world problems that a programmer could work with, solve and monetize.