Hacker News new | past | comments | ask | show | jobs | submit login
Show HN: FastComments (fastcomments.com)
111 points by winrid on Feb 1, 2020 | hide | past | favorite | 67 comments

Okay, here's some feedback -

What I Like - The feature set and the clean look

What I cannot find - A demo on the home page itself. Not be required to click a "live demo" button

What turned me away - The pricing strategy. As a dev, I love the pricing. As a user, I dislike the pricing. Just give me a flat price and call it a day. Especially applicable for small blogs, newbies and people who don't yet have traffic. That breakdown just indicates to me that I will have no idea what my end bill would be like, especially if my blog starts gaining any traction.

That pricing page also probably limits your audience to a small subset. I showed your pricing page to someone non-technical and they pretty much lost any interest.

The pricing most definitely needs to be simpler. The calculations on there and numbers used will turn away a lot of people. Think simple simple simple when it comes to user facing messaging and pricing.

^couldnt agree more. both packaging and presentation of the subscription tiers being simplified would help imo. that said, certainly a/b test that!

Yep, fixed the presentation and pricing.

Suggestion regarding pricing: two or three flat-fee tiers, each with a cap on traffic (after which you can just drop requests, or do whatever you want) and then a tier with a minimum + variable charge by traffic for higher usages.

This is just what I was thinking in the shower....


Your pricing is currently out of order -- $5 for 1 million, then $350 for unlimited, then $75 for 5 million

Yep! Here's an example that describes that strategy:

"Different Levels of Pricing" https://www.helpscout.com/blog/pricing-strategies/


Pretty much everyone says the pricing model sucks. I agree. Will rework.

Edit - Don't want to devalue it but wondering if it's worth having a tier for the really small players? So like $4.99 up to 5m page loads a month or something. Then $99 for unlimited and you know enterprise is its own game.

Edit#2 - If you sign up now I will migrate your subscription to the new pricing model which I will implement tomorrow, no worries (unless you really really like the current pricing....)

Seeing the current $70/month for 250,000 page views definitely turned me away (did I read that right? weird decimal place in your copy: 250000.00), but I might not be your target audience. If you do lower the price, just make sure it's long-term sustainable. $5 for 5 million page loads (if that's what you mean in your comment above) would be great, and something I'd consider (since the alternative for me is setting up a $5 DO droplet with schnack[0] or commento[1]) but that might be a bit too cheap depending on the costs on your side (e.g. fighting spam). Might be worth looking at competitors' pricing for inspiration. A lot of developer types would most likely choose your service over disqus even if you're a bit more expensive, simply because of all the bloat and tracking and shady behavior of disqus. That would be too expensive for me, but like I said, I might not be your target audience.

I love the simple setup and that you're focused on performance and not spying on users! If the pricing gets adjusted, I would seriously consider adding it to some of my sites. Oh, if you're not going to make it open source, then an "export" feature is crucial for me, because I despise lock-in.

[0]: https://schnack.cool/

[1]: https://github.com/adtac/commento

Yes I will definitely do load tests to determine what won't break me. I really like the $4.99 number. Now have to pick a traffic number.

You can export all of your data. Made sure I had that from the get-go. When writing the importer from the other providers it's amazing how much they hide from you. Disqus doesn't even export avatar urls.

Decimal place - yeah I tend to automate things and all that is generated from some configuration. I trusted Number.toLocaleString too much :)

Schnack - thanks. One cool thing that has that I don't is localization. I have to work on that.

Edit - a fun story that should have shown me how bad the pricing model is. I was using Mapbox for a project and accidentally made 1m api calls. I still owe them a few thousand dollars.... So yeah, I am morally obligated to fix it. :D

Localization's been added.


> something I'd consider [...] since the alternative for me is [...] commento

It sounds as if there's something about Commento you're not totally happy with -- I'm curious about what that could be?

I think you might actually be devaluing it by coupling pricing to page loads. The value of a comment system is more in how many people use it and how many comments they make.

> Pretty much everyone says the pricing model sucks. I agree. Will rework.

Why would you ever listen to people who are not your customers (pulled out a credit card, agreed to be charged) about what you should be doing?

Starbucks coffee is $2.45 + taxes. 100/mo is less than a cup of coffee a day.

Well, I already knew it wasn't great just compared to the competition and the value proposition.

Don't worry. I have pushed back a lot on a lot of feedback (quite a few people did not like the design @ Reddit). I'm sticking to my values. :)

Looks good but need to rethink pricing. For a small blog, it says you won't pay more than $71 a month. That means I COULD end up paying $71 for a small blog. That's very expensive for a commenting system. Yes I understand that it is ads free and all but this pricing overestimates how much people really dislike ads.

Yep, going to redo that. Would you pay $4.99 for a small blog, vs free Disqus and deal with ads?

I am not looking for a solution right now but if I did, Yes I think a small charge of few bucks a month may work. The problem is that most small blogs don't really need heavy commenting and when they do, they can get away with free comments from WordPress or even disqus.

You need to think about the value proposition of why someone wants to PAY for a commenting system. It may not always be "I hate ads". I would say find people who run blogs with comments and start talking to them.

Exactly. That's why I'm adding lots of features a default WP install won't have. I should probably also support export/import from WP...

I also plan to write a WP plugin.

Edit - yes the 1:1 reaching out will happen soon. I suck at sales, and it's super hard not to just focus on writing code. But I will do it!

I currently pay around that for Commento. I wouldn't pay more at the levels of traffic I get, which is pretty/very low.

Well this is a fun load test. Nice to see things aren't falling over while being on the homepage (Thanks HN!!!!).

It's also a fun load test for my other project, Watch.ly, which doesn't track you but I can see in realtime the traffic to the site and it's peaking at around 100 concurrent sessions.

I'm seeing lots of requests from fun things like go-http-client and such.

I wonder how this compares with Commento?

Differences I've found:

FastComments has a smaller Javascript file.

Commento is open source. Could it be more stable since been around for longer?

What would you say, @winrid? There's this blog post: "Migrating From Commento To FastComments" -- what are the main reasons someone would do that?

Hey, first of all thanks for asking the right questions! :) I have considered putting together a big "FastComments vs..." chart to answer these questions.

I'm trying to offer a larger feature set than Commento. This means things like: 1. Better ways to fight spam 2. Migration supports more than just Disqus. 3. Inline image support (which will also be able to be toggled in the future) 4. Customization on a per domain or url basis: https://blog.fastcomments.com/(1-24-2020)-how-to-make-a-comm... 5. No signup process, no passwords. Everything is email based, however we allow anonymous comments now (configurable).

Also, solutions for larger customers (integrations, developer support, etc). At the highest tier for $20/mo Commento gives you 100k monthly page views. FastComments gives you one million on the lowest tier (closer to what Disqus gives you on the lowest tier, but cheaper).

I'm not sure Commento has pagination? I'd have to check. But we just added that today too. Also, a cool blog post is coming on how you can optimize threaded pagination! :)

But yes Commento has been around longer. I think I'll catch up quickly though.

Thanks for the reply and your thoughts, best wishes.

I like the goal and the product seems good apart from some subjective styling nitpicks, but your blog is terrible. Why can’t I click on any of the titles or/and a “read more” link?

After figuring out how to get to a single post, I noticed your blog uses the product, perhaps that could serve as a nice low-effort demo? Firing up an editor and trying out the code snippet is too much work for something you haven’t even seen yet, imho.

Yes that's on the todo list. Thanks for bumping up the priority...

Fixed the blog :p

With Firefox night mode enabled, the comments can’t be seen.

When blocked images, the whole comments is hidden, it can pose problem for some readers.

Would be cool if you checked now. On FF mobile I installed "Dark Mode" which made the page black and the text white. I also disabled images.

Will check that out thanks

Copy-writing on the pricing page: "We'll charge you for your traffic to FastComments. For example if you have a small blog and you never get more than 250000.00 visitors a month you won't be charged more than $71.43 a month."

Who measures number of visitors down to their 100th part? Remove the ".00" to make it simpler. The price "$71.43" suggests the underlying pricing model is somewhat complicated, as others have identified, too.

Redid whole page, thanks.

I've gathered all the commenting systems I could find in my blog post [1]. I'm glad to see yet another alternative. I will add it soon to my list (maybe I should wait for your pricing adjustment).

[1] https://darekkay.com/blog/static-site-comments/

Pricing's been adjusted! https://fastcomments.com/traffic-pricing

I'd use this if we had complete customizability of the comment system's design through CSS.

You do! I made a page that looks like Hackaday to demonstrate that:


It's good that the CSS is customizable, but many of your clients won't customize it, and the default text size is too small. I did a quick test of increasing comment bodies from 13px to 16px, and comment headers from 12px to 15px, and it made everything much more pleasant.

Consider increasing the default font sizes? If comments are worth writing and reading, it's worthwhile making them eye-friendly and easy to read.


Interesting to see, I am curious how this goes especially with the pricing model. Have you been doing this very long?

I would want to see a docs page and a React, Vue, Angular etc client library before I would realistically purchase this and spend time trying to consume it.

Could you help me understand why would you need a client in each framework? It's in vanilla JS and attaches to a Dom element. :)

Edit - one thing to note is 99% of the code is actually just loaded in an iframe. So the actual code on your site is very small.


Edit #2 - no I've only spent a month on this so far.

Would you be happy with a wrapper in each framework around the native library, and how would you use it?

This looks really great, congrats on shipping. I think it'd be worth it to have a designer give the site a facelift, I think it would convey more trustworthiness. Also the pricing page doesn't work very well on mobile when selecting plans.

All good feedback thanks

Pricing and pricing infrastructure has been redone. Thanks everyone!


There are a LOT of blogs which get well under 1 million pageviews per year, let alone per month. I dont know your admin costs, but perhaps there's room for a Micro plan at a flat $15/year for something like 500k annual pageviews?

I could work that out if you're interested.

How long do I have to wait for my cat post to show up in the demo, after confirming both my email and my post?

EDIT: I think my first comment simply vanished, the second one went through instantly :(

That's weird and simply shouldn't happen. What's your username? I did do a deployment earlier, and since I don't have a queue in front of the this (and nginx isn't setup to replay POST requests yet) it is possible that is why.

meowmeowmeow. My first comment included a picture (rabbit showing tongue) and I received a pair of emails for verifying my account & comment at 23:30:33 EET, according to server logs. I promptly verified both, but that comment never showed up.

The spam filter thought it was spam :) It should have shown a message. Right now verifying your comment doesn't unmark it as spam. I'm still contemplating that.

Edit - it should show now. Just clicked "not spam" in the dashboard...

Oh, I actually saw that message. I thought it just meant all messages from new users would be flagged as spam until verified.

Ah, well after I deploy this next round of changes verified accounts bypass the spam filter (and verifying a comment marks it as not-spam).

I'll experiment with this. I bet it'll have to be configurable in the future.

The pricing model itself is one thing that people have covered here, but why the hell is it so expensive? Over $1000 / mo just for a comment widget???

Yeah that was something I should have thought about more before posting. I am thinking of having two tiers, up to $99/mo for the biggest sites. Most people would use the $4.99 package.


Just to mention, when I added the embedding I was getting a 404 error. Not sure if it is supposed to work or not without a subscription;)

Should be fixed now, except commenting won't work for the demo tenant id. I'll address that too. I was a little overzealus with my request validation.

Well. Took a little downtime there due to an issue I had to fix which is embarrassing. Back up now. Whoever just broke it - thanks. :)

I think you should rework your pricing model.

Is there a fallback for no-JS users?

On the TODO list. However, I think it'd have to be opt-in.

EDIT: Maybe not. You'd just see a scrollbar since the iframe can't auto expand:


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact