Here’s a general guideline I’ve come up with:
If you aren’t getting enough downloads, something is wrong with your marketing. Improve exposure, presentation, screenshots and writing. Make sure the potential customers you are talking to are the right audience for your game. Work on this until you get at least 100 downloads per day because you can’t test without traffic.
If you have traffic but not enough sales, something is wrong with your sales pitch. The demo is broken, too hard or easy, or shows too much or little. The price is too high or low, or there isn’t enough value difference between demo and full versions. You have a good download flow for testing because of marketing improvements, now tweak things in the demo. Some demo players will give you feedback that helps. Work on this until you are converting at least 1%.
If you’ve done the above but can’t get conversions to 1%, something is wrong with the product. Talk to your demo traffic and find out what you need to fix. Release improvements back into the traffic then watch conversion rates while soliciting more feedback.
I believe any playable game is sellable using this basic guideline. Tweaking a playable game is much easier than building a new one, so take advantage of the asset you’ve built! Talking to players who like what you’ve created and have ideas to improve it is very rewarding, both emotionally and financially. And once you have this experience with an audience, you’ll get future products closer to correct at version 1.0.