Back
Post
Should I pull the plug on Tailstream?
Tailstream is the most ambitious project I have started building so far, but I am also considering on pulling the plug on it. Why?
- - The initial trigger was this post by Yongfook
- - Pretty much zero traction
- - Visualisation alone is not a painkiller people want to spend $$$ on
- - Log management is useful, but there are other players in the market that can likely offer better tooling for less
In the past couple of months, in my free time, I've built the main product, an open source Go client, an open source Go ingestion agent.
I still think that the onboarding process I've built is a lot easier than most of the other players in the market, but without users, this is all moot.
Then, at the same time, there's this voice in my head that says "keep trying" because I might be close. That's the hard part. Maybe there is a magical sequence of buttons that I can push that will suddenly turn this into a product that people want to use, but it's a guessing game. The marketing site is also not really reflective of the product.
The process so far is building something you think people need, convincing yourself that people will actually spend money on it. But when push comes to shove, people don't use it and it's super hard to figure out if this is because of the product, the marketing or something else.
Anyway, that's where I'm at right now. It feels like such an incredible waste to pull the plug, but if it's not going to work, I should do it at some point.
Would love to hear your honest advice.
👋 Join WIP to participate
Ok, since you've been working for a year on it, I would spend at least one or two weeks to decide and do the following:
Make the product demo not a video, but the real product with data.
Something like datafa.st
I need to be able to try your without having to plug my data, because otherwise it's too much friction.
Talk to your users. Reach out to your first customer, and all your signed up users. How did they find you ? What do they like about your product?
You need to find the ONE feature you do better or different than everyone else.
Your product seems fun, but you need to understand the real problem you can solve and yes, maybe some fun dataviz is not enough.
My intuition would be to add some AI. Something that can directly fix issues by reading log streams using Cursor Agent/Codex/whatever
What's your traffic ? How many signups do you get per month ?
Maybe I'm wrong, but it seems you focused a lot on the product, and not on marketing. If you know nothing about SEO, just buy Danny Postma's SEO course and watch it: www.dannypostma.com/course/se…
Then get a ahrefs subscription (I would do the $119 for one month, then downgrade to the $29 plan) and research keywords that your can easily target to get users.
Then build some pages on those keywords, and grow your DR a bit because 0.6 is way too low (Read and implement backlinks-list.com/domain-aut… + send me DM and I'll give you a free premium account)
You just need to find the ONE killer feature and focus on acquisition.
Good luck :)
Thanks for taking the time to reply!
Some more context:
My initial impression looking at your website is that you're selling a space invaders visualization of logs?
If that's the case, it's no surprise at least from my POV. Sure, it's fun to look at, but why would I pay for this?
It doesn't solve any problem that I'm aware of. I already have alerts set up to notify me when errors happen and a dashboard with a line chart for errors/warns/latency numbers, no need to look at some new type of visualization for this.
IMO it would not be any more effective to see errors/warns visualized on a space invaders animation instead of a standard line chart.
Thanks @ben, this echoes what I think is a large part of the problem for people who do end up on the website.
I started out building this tool as a way to build real time visualisations for event streams (I have support for multiple visualisation types and the Space Invaders theme is just one of them), basically a "web based logstalgia" with built-in TV-mode.
I then figured that this is "fun" but not really a painkiller, so then pivoted slightly and started building more tranditional log management features (see video) but that's not shown on the landing page at all so I understand how that gives people the wrong idea.
Apart from that I also built a CLI client that allows people to "tail" their logs via the API.
I think what I'm currently doubting most is whether I should continue down this path or abandon it because the log ingestion space is become more crowded recently with players like BetterStack and Sentry integrating log management into their product offering.
Right, I would start by asking: "What can I do better than BetterStack/Sentry beyond an animation?"
And if you don't know, you should probably build something else. I tend to only build things where I think I can provide a significantly better solution.
DId you link to the wrong post? That post that inspired you was from 2 days ago.