WTF Is Wrong with Webflow? | Hacker News Reader

WTF Is Wrong with Webflow?

twitter.com

25 pointslawgimenez48 comments

Looks clear enough to me right? They took a plan, but are using more than that plan offers, so they're being offered a plan that does offer what they're using.

It does show the reason why you shouldn't go for a vendor lock in like this, especially when your entire business depends on it.

Yeah idk why he is throwing a tantrum on twitter, he spent way more than what his plan offered and he got shocked when Webflow wasn't okay with it...

It’s alllll fun and games until the SaaS you 100% rely on decides to become profitable

I've been speculating for a while that SaaS is a zero interest rate phenomenon.

It's not that the SaaS model can't be economical, it's that all these companies are suddenly expected to beat a much higher interest rate bar and must enshittify rapidly to squeeze their customer base. Since SaaS is absolute lock-in, they can... for a while.

The PC revolution and local software exploded when interest rates were high and in addition to the novelty and flexibility a major driver was cost reduction and having local control over cost. If you can run software locally on your own stuff you can decide whether to invest in an upgrade. If everything is a subscription to a remote system you have zero control, which seems like less of a problem when money is free. Lots of economically irrational things like variable-rate outsourcing with hard lock-in make sense when money is free.

Maybe those of us interested in decentralization and local-first just have to wait for the ZIRP era to unwind.

Look, SaaS, or plain classic Software isn't an inherent enterprise stuff. SaaS model isn't a zero interest rate phenomenon. VCs selling successful software is a ZIRP phenomenon. We will survive without them.

Hear, hear. There are plenty of profitable SaaS companies with sensible unit economics.

This is amplified by the fact that many of those SaaS are startup and privately financed. They are living on borrowed money, and if the money dry up, they will need to move to a viable business model, that is, charging the consumer the real cost.

>The PC revolution and local software exploded when interest rates were high and in addition to the novelty and flexibility a major driver was cost reduction and having local control over cost.

At that time (80s to 90s), computer networks weren't really a thing outside of institutional settings, so it wasn't like there was anything to compete with "local software".

Software subscription is basically a form of loan. Obviously interest rates influence them a lot.

What is different from the ZIRP inflating unsustainable startups. SaaS is a loan even when it's bootstrapped or financed in a normal way.

totally unrelated comment, he spent 3x-5x times is allocated bandwidth of 400GB, it is his fault and should have been more cautious about it

You missed the part that this bandwidth limitation didn't exist when he signed up. Not sure how he can prove it or if he missed reading it.

Exactly. You sign up when you're just playing around, and the product is free or cheap. It's easy to ignore the details of the enterprise pricing at that point, even if it exists and is prominently displayed. By the time you need enterprise pricing, you're locked in and they've got you.

I'm not anti-SaaS by any means, but it's a cautionary tale about over-reliance on services that are way outside your control.

Not unrelated at all. Webflow is suddenly using the excuse of 3x bandwidth usage to justify a 3x increase in subscription price, when we know that this isn't proportionate to the underlying hosting costs.

I guess he can write about this failure to anticipate in his failory.com

Looks like he was paying $40/m for 2TB of bandwidth. No wonder he drew their attention.

And before you give me the BW is cheap speech realise that this is about segmentation not IP packets.

Moving the heavier assets to a more specialised domain would have been trivial.

2TB of bandwidth is $7/mo on AWS Lightsail. You can argue market segmentation but surely the bar for "enterprise" pricing should be something with enterprise costs. I as the sole user of my VPS can hit 2TB of bandwidth a month easily, it's just not that much.

When AWS, the company known for price gouging for bandwidth, sells for 1/1000 the cost the problem might you.

If lightsail was comparable Webflow wouldn't have a business. It's like saying a Bentley and a lawnmover are the same because they both require fuel.

The segmentation is, and has always been, based not on underlying costs, but on an understanding of what the segment can bear.

But the person you're responding to is comparing the fuel (bandwidth) and not the vehicles.

>2TB of bandwidth is $7/mo on AWS Lightsail. [...]

>When AWS, the company known for price gouging for bandwidth, sells for 1/1000 the cost the problem might you.

That's a slightly misleading thing to say when lightsail is specifically their low-end offering, designed to compete with other low cost VPS providers like linode or digital ocean. When people are complaining about AWS "price gouging" for bandwidth, they're talking about the $0.09/GB[1], or $180 for 2TB.

[1] https://aws.amazon.com/ec2/pricing/on-demand/

He said "up from $468" not $40.

He said:

>They're forcing me to pay $15,000/year, up from $468!

$468/year = 39/month

> Your current [plan] allots 400GB monthly ... your usage is 3-5x

What I've never understood about Webflow is why do people host their site with Webflow (for exactly the reasons above)

You can create/update your site with Webflow, but "publish" it to your own server (which might be a cheap $10/mo hosting provider that given large or unlimited bandwidth)

The website will likely be even faster hosting it elsewhere as well, it's radically cheaper, and you have more control yet don't lose any of the benefits.

https://university.webflow.com/lesson/code-export?topics=hos...

Webflow offers features like capturing form data, a CMS, and content translation, which some find convenient but make it harder to self-host the HTML/CSS that is generated by Webflow. But, granted, a static site can be designed in Webflow and hosted elsewhere easily.

Because it's a one-click process to publish, which makes it easy for designers. It's not really more complicated than that, "just host it yourself" sounds an awful lot like "Dropbox is just rsync".

Webflow's core user base are users who don't know words like "server", "hosting", "bandwidth" and such. They're just users who want to drag and drop some stuff together, click on "publish" and now have a website.

This is what the price reflects as well.

The "publish to your own server" is a cruel joke. Anything remotely dynamic, even things like CMS pages that can be generated at build time, don't work. To be fair, webflow itself is a joke.

Do a Twitter search for Webflow, and you'll see why. They aren't targeting mature companies at all, it's all solopreneur hustle culture sites.

Really getting tired of SaaS-makers relying on serveless, not monitoring or controlling ANYTHING, and then trying to put the service they use on blast publicly to try to negotiate themselves a special deal.

his app was a blog and newsletter, hardly costs $15k a year to run

Still, it looks like from the email that the website creator didn't read the ToS, and consistently consumed significantly more resources than allowed for a few months.

Really tired of hustlers coming to defend dark patterns use.

What's the "dark pattern" here? Having bandwidth caps being hidden? It's pretty visible on their pricing page[1]. Having bandwidth caps at all?

[1] https://webflow.com/pricing

It is pretty visible now. Was it pretty visible before? Was it the same amount when OP signed up? That's what he is arguing, it wasn't a limitation when he started to pay.

They have ALWAYS been very visible. They may have changed just like the cost of the VPS or whatever infrastructure you rent may change. I'm not arguing that serverless platforms have fair pricing—I don't think they do—but I'm arguing that if you subscribe to a platform that bills on usage, it's 100% your fault if you fail to monitor your usage and get insane bills.

It's not even SaaS-makers sometimes. Last time I remember it was a casino trying to get sentiment and attention so that they could continue abusing Cloudflare for $250/mo [0]. This honestly needs to stop.

[0]: https://news.ycombinator.com/item?id=40481808

Isn't this trivially fixable by putting a cdn/cache in front of the webflow site? Does webflow prevent this somehow?

Yes, I have put Fastly in front of it to drastically reduce costs.

Putting cloudflare etc in front of webflow breaks how webflow does their lets encrypt renewal (they use HTTP, not DNS), so after ~90 days the site goes offline unless you allow un-encrypted origin connections.

If you were 3x-5x over the bandwidth limits for your pricing tier for multiple months in a row you were living on borrowed time…

Is this on Webflow to continue to allow usage wayyyy out of accord with their pricing or the site owner to monitor and understand their own bandwidth consumption?

This is vendor lock-in 101 unfortunately, and why I've always steered clear of VC-backed companies that offer "all-in-one" solutions that can't be migrated easily. Another company that does something similar (builds WYSIWYG landing pages with marketing analytics) is Unbounce, and they jacked up their prices too, but not to this level.

It was clear from Day 1 that Webflow's webpage building software with HTML export was a trial run for a hosted CMS. That's how they get ya.

>This is vendor lock-in 101 unfortunately, and why I've always steered clear of VC-backed companies that offer "all-in-one" solutions that can't be migrated easily

Webflow allows you to export your site and host somewhere else.

https://university.webflow.com/lesson/code-export?topics=hos...

That's effectively useless if you're using Webflow CMS. All the export will provide is a bunch of blank template pages. The site data is on their DB, and that cannot be migrated. That's vendor lock in pure and simple.

Webflow is not portable the way WordPress is. You can host WP on anything that supports PHP, even a local development server on your computer. Webflow can only be hosted on Webflow's servers.

First rule of a dealer: do not bankrupt your customer. Their pricing needs to be more transparent and more gradual.

I see two different issues here, one is that allegedly Webflow added the traffic limits later. If they substantially change the rules, this is something I'd mostly blame the company for as you can only work around that by not using any SaaS products.

The other issue is that it looks like they have a bunch of nice, well-defined plans for the low end. They have all their limits and prices spelled out on the pricing page. And then they have an enterprise plan that seems to start quite early in terms of traffic, and you can't see those prices beforehand. And given the report here, those prices seem to jump enormously compared to the public plans. That is a very deceptive practice as they're not telling you the rules and prices beforehand.

Why can't SaaS implement a proper throttle after quota is exhausted? If customer sees value, customer can pay to raise quota.

At least this used to be possible in most hosting plans in old days.

It doesn't look like they're sending a huge bill for overages. They're sending a warning letter to get them to switch plans or risk being terminated as a customer. Adding a throttle/quota wouldn't have really made a difference here, as the guy seems to be more upset at how much the new plan would cost.

It's not the only one doing this.

It almost feels like a final squeeze before some kind of collapse.

Let's see if his startup can help him overcome this risk of failure

Reply on news.ycombinator.com