ntalbott

That's certainly one perspective; prescience is something I wish we were better at. Another perspective is that a startup is a search for a viable business model, and each of the subscription services in turn has realized that their entry price in to the market was decidedly unviable. We were able to sustain ours for longer than most because we (a) never offered a free plan, and (b) have kept our overhead ridiculously low. But (b) has kept us from improving the service like we need to, which in turn hampers growth, which is a nasty feedback cycle.

So it was time, and it was painful, but now we get to focus in on the fun stuff: making the service more awesome, and growing.

Once you hit sufficient transactions/month your per transaction fee will drop back to 20¢/transaction.

Trust me, we weren't happy about having to raise the price. It was a painful, hard-fought decision to change it internally, and to not permanently grandfather existing customers. But in the end it's about having a sustainable business - the worst possible outcome for customers is for the service they were powering their billing with to disappear altogether.

One learns a lot in a year. I said some things then that I wish I'd had more experience to back up; if I had, I probably would have tempered them more. That said, I also said this:

"Even if you absolutely have to raise prices across the board, I think three months of warning is the absolute minimum amount of time to give a customer base before you hit them with the increase."

We're doing that, with a unilateral offer to grandfather customers for 90 days.

In general, I still think grandfathering is best when possible, but when you realize your business isn't growable (and barely sustainable) at current prices then something's gotta give. We've put this price raise off too long, hoping to soften it with new features, etc., but it's a chicken and egg problem. We can't do what needs to be done without the resources to do it. So prices have gone up, and it's the next day and we're in a better spot to help all our (remaining) customers rock.

Reading the bill, I'm not seeing where it bans community fiber; rather it seems to require local government to not use tax monies to subsidize the access, and to not abuse their position to hamper private competition. Both seem pretty reasonable to me. What am I missing?

We (at Spreedly) have talked to several QSA's about this question, and their take is that using a redirect removes the application from PCI scope. It's a really good illustration of how PCI != security.

I run Terralien, and I can only speak from my own experiences doing so, since it seems most businesses that have been built mostly by a freelancer or a devshop don't go out of their way to publicize that fact.

Terralien does ongoing development work for six businesses for which we built the original technology. Their degree of success varies, but they all have in common that they're successful enough to have persisted 1+ years and are continuing to make progress. All of them rely on software for their core business to some varying degree, but in none of their cases is the software itself rocket science. It may be moderately complicated and there may be a lot of it, but a good hacker could pick it up without completely derailing the project in the meantime.

On the flipside, Terralien had one very tech-heavy business that we got off the ground, after which they brought development in-house in the form of a developer working for equity. It made total sense for them, since they were doing algorithmic work in the energy space so their technology was in a way "rocket science" and they needed to have that close.

I think anyone considering outsourcing the software build for their startup should look for a freelancer or devshop that is ready and willing to serve as their CTO for the long haul. You want a shop or an individual that understands technology as a tool that should serve the larger business purposes, and not an end in itself. That's the way we try to approach it at Terralien, and it seems to be working really well for our clients.

That's correct; we'll rapidly (and I mean rapidly) add any of the gateways on that list to Core as beta customers request them. And we can probably add gateways that aren't on that list - just drop us a line to double-check. In the meantime you can kick the tires with the Test Gateway today.

There's a lot of content still to get up, and a lot of design polish still to be applied. We're not quite ready for a public launch yet, thus the "these guys are just getting started" ghetto look. That said, our Subscriptions product (https://spreedly.com) has been around for 3+ years and is still going strong, so we're in this for the long haul.

It doesn't work/help yet, but improving the 3D-Secure process is in our sights. And it could definitely use improving.

Our copy needs so much work. Wasn't really counting on hitting the front page of Hacker News today...

Good call on having the pithy 5-10 word description - that's on my short list now.

There are a few key value props:

* Your PCI compliance burden is significantly reduced. * Card storage is a seamless process, and at the end of the day you own the card data. * You can easily switch gateways (or balance across multiple gateways). * Our API doesn't suck, which doesn't sound like much until you've had the misfortune to implement against a few gateway API's.

And there are other value props we'll be rolling out as time progresses - there's a lot of cool things we can do to improve the life of anyone collecting money online.

We do indeed have the process in place for sensitive data exports - it's a bit involved, as it has to be, but we've done it more than once for our Subscriptions product.

Our understanding after talking to multiple QSA's about the transparent redirect is that it allows the merchant to legitimately claim that they do not "store, process, or transmit" credit card data, which means they only have to fill out SAQ-A.

And trust me, you definitely want to fill out SAQ-A.

Our philosophy is one of "you own the data". We've done dumps for customers of our Subscriptions product, and we'll gladly do them for Core customers as well.

We see this as one of the significant value adds of Core - avoiding lock-in, including lock-in to Core itself.

Totally correct, and we need to re-iterate the SSL requirement in a few places. Thanks for pointing that out.

Spreedly CEO here...

So, we weren't quite ready to land on the front page of HN, but mission accomplished - we now have a whole list of smart people to invite into the beta, so thanks to bradleyjoyce for giving us that nudge.

I'm going to work through the questions that have come up in the discussion - good prep for the FAQ page I need to write - but feel free to email me directly if you have any other questions or just want to chat about Core (my email address is in my profile).

[Disclaimer: I run Terralien, and the author of the article works for me.]

Building custom software is nothing like building a cookie-cutter tract house - completely agreed. In that case you're getting the same house as someone else, the builder has done it all before, and there are only a few options that you get to pick with the costs for those options known.

Building custom software is a lot like building a custom house, though - while there's an estimate going in, it tends to change a lot, and the final price varies widely depending on the buyer's choices during the project. Try being the general contractor on your own house construction project and you'll quickly discover costs are very variable regardless of the estimates up front.

There are a lot of stereotypes about how construction projects are "fixed bid" and "everything runs off of well-defined blueprints"; my Dad is an architect working on large healthcare projects, and I can tell you that those stereotypes are largely wrong. The projects do not go according to plan and even things "in the blueprint" change regularly mid-project.

Quoting myself: "Capital infusion causes market distortion. Not saying that's a bad thing, but it's worth recognizing."

If this comes off as an attack on Grasshopper Group - my apologies. I think what ya'll have done is pretty impressive, and I'd like to duplicate it myself. At the same time - Chargify wouldn't have ever gotten into the situation of supporting so many unsustainable free users if it (as a stand-alone product) was bootstrapped. That's all I'm saying. Calling that a "sugar daddy" was probably overly inflammatory.

Perhaps - I have no internal knowledge of Chargify's numbers beyond what they've posted publicly. Even if they have 30000 global accounts, though, and it's 0.9% out of that - makes me think twice about freemium.

So lets do some math:

From http://chargify.com/blog/why-we-changed-our-pricing/:

"Grasshopper Group has been supporting Chargify for 15 months, and Chargify is now supporting 3,000 merchants - again, the large majority of whom are paying $0."

And from the linked article:

"We should have shared the data we collected for over a year that demonstrated quite clearly to us that only 0.9% of customers were paying us at all, and that there was a direct correlation between those that did not pay anything and a high volume of support requests."

3000 merchants total * .009 paying = 27 paying merchants

Now, these numbers could be off in several ways, but even if they're off by an order of magnitude (and I doubt they are) - ouch!

So many lessons to be learned from this:

- Freemium's applicability is so over-rated. I couldn't be happier that we've had a (cheap) pay-to-play structure at Spreedly since the beginning. Personally I think freemium should always be a strategy introduced post-profitability, never before.

- It's been said many times, but I'll say it again: the "overnight success" is a myth. Most businesses take 2-5 years to reach any kind of profitability. Selling to startups? Better keep that in mind.

- Capital infusion causes market distortion. Not saying that's a bad thing, but it's worth recognizing. Spreedly noticed a definite drop-off in signups when Chargify came out, and why not - wouldn't you take the free option vs. Spreedly's pay to play? And wasn't that some great marketing they poured a lot of money in to? And yet their free option and their marketing were only possible due to the Grasshopper Group's investment.

- Put that all together and you end up with this: you've got to keep your burn low until you find your product/market fit and your scalable sales model. Or you've got to have a sugar daddy. Or you've got to win the investment lottery. Spreedly is still alive because we've only allowed costs to grow as we're able to handle them - that's the nature of the bootstrapped startup, hard as it can be at times.

One more bit of math: if my numbers above are correct, Spreedly has 6x as many paying customers as Chargify. And we've had zero outside investment. And, painfully, we're still not ramen profitable. But: we have a pivot in the works that will blow the doors off of what we've done to date (and it doesn't involve raising prices on existing customers - promise!). So stay tuned - this space is just starting to get interesting.

Oh, and if you're an angel investor: we'd love to have some capital to use to distort the market in our favor for a change :-)

Spreedly is PCI Level 2 compliant, which while not cheap, is significantly lower cost than Level 1 compliance. There's not additional liability involved by us being Level 2 vs Level 1 - while Level 1 is certainly a nice marketing strategy, the truth is that it's not required until transaction volume reaches a pretty significant level. Once Spreedly reaches that level we'll get it, but by that time we'll also be able to afford it.

In terms of support, what we're doing at Spreedly is focusing on providing excellent email support, and occasionally supplementing that with IM/phone support for critical issues. It's a trade-off that seems to be working well for us and our customers at this point.

At Spreedly we rarely do any phone pre-sales or support, and that's a very intentional choice. As noted above, good phone support is expensive to provide, and one of the ways we've decided to keep costs down (at least for now) is by sticking to email. For those who want phone pre-sales and support, Chargify's offering looks like it could be a good option.

On the one hand, I agree with you - Spreedly's not as feature rich as some of the competition. On the other hand, we have a lot of very happy customers, so that doesn't mean we're not a good fit for a significant set of startups.

> Bait-and-switch by accident or by design makes no difference to your customers.

Not trying to justify, just inform.

> Do you guys do segregated merchant accounts (one merchant account per customer)?

We don't do merchant accounts at all - that's pretty much a constant across everyone in this space. Businesses plug in their own merchant accounts - Spreedly just handles all the billing logic.

Full disclosure: I run Spreedly, an earlier to market but much less funded competitor to Chargify.

Here's the dirty little secret that Spreedly discovered about 18 months ago, and that I'm sure Chargify - like Recurly before them - has now found out for themselves: there are plenty of people who want to start subscription businesses, and out of those, the vast majority will not succeed and will actually end up costing more than they ever bring in. That leaves businesses in this space two options: either focus on successful startups and actively filter out the "losers" (by charging a minimum of $99/month, for instance), or minimize costs for "low probability" businesses - and they're all low probability early on! - and use "cheap to try things out" as a star search for the few businesses that will end up getting big.

Note that Chargify paired this price increase announcement with two other significant announcements that have largely been overshadowed by the hullabaloo: PCI Level 1 compliance, and 24/7 phone support. This pairing is not coincidence - I'm pretty confident that the pricing change is very firmly tied internally to these two new "features". PCI Level 1 compliance is a hefty upfront cost plus a large ongoing price tag. Good 24/7 phone support is crazy expensive to provide, and means that free customers would eat their lunch since so much of the support needs for one of these businesses is on the front end.

The ridiculously huge mistake I think Chargify made here was something I thought was just a given these days: they should've unilaterally grandfathered all of their existing clients, and quietly given the grandfathered plan to anyone who was already integrating but not yet launched as well. When Spreedly made our last pricing change - from percent of transaction fee to flat per transaction fee - it was a price drop for most of our customers. But not all: anyone with super low prices would've ended up paying us more, so we explicitly made it the minimum of $0.20 or 2% of the transaction. There's just no excuse for ticking off existing customers - it makes you look like a cell phone company. Even if you absolutely have to raise prices across the board, I think three months of warning is the absolute minimum amount of time to give a customer base before you hit them with the increase.

So, that's my $0.02 - hope it helps folks understand why I think this is happening. Questions, feedback, etc., welcome.

P.S. I think in some areas Chargify's definitely ahead of Spreedly Subscriptions in terms of functionality - all of that extra capital definitely shows in the end product. But that's because we've been focused on "what's next" after you figure out the naive business model doesn't work in this space. I said above that there are two options. That's a lie - we're working on a third option. If there are any angels reading this that would like to invest in a team that's been thinking deeply about this space since before Chargify and Recurly were a twinkle in their founder's eye, drop me a line.

It's in process - goal is to have certification by year's end. We know it's important!

Hey, Spreedly guy here - we're definitely a full-featured solution, and we're already servicing dozens of businesses (we've sold 30 Kickstarts [http://spreedly.com/info/kickstart] in less than two weeks, so we must be doing something right!). Sorry if we've ever given off a different impression. What you may be referring to is that we're primarily focused on time-based subscriptions for now, vs. usage based subscriptions - that much is true. That said, very soon you'll be able to pile usage charges on top of any time-based subscription, which covers what a lot of businesses need in terms of usage fees.

Still not clear? Probably my fault! Hit me up at support@spreedly.com anytime and I'll gladly clarify.

I'd strongly echo this - it's the reason I grabbed some co-founders and started Spreedly. After spending weeks building a billing system for a client and not being at all pleased with the result (bad API, insufficient budget, etc.) I knew there had to be a better way. Billing is the red-headed stepchild of an application, never getting as much time as it needs and simultaneously taking up more time than it should - indicators that point towards outsourcing it ASAP. I think Spreedly's the best option, but even if you don't go with us, please don't do it yourself!

I'm biased, but I'd strongly recommend taking a look at Spreedly (http://spreedly.com). We can't help with the provisioning, but we can make the billing drop-dead simple so you can focus on getting your infrastructure right. The best part is: you can grab an account and give it a spin with no obligation, as test sites are completely free.

FYI, in less than two weeks we've sold 30 of our "Kickstart" plans (http://spreedly.com/info/kickstart) - a lot of businesses are hopping on the Spreedly platform, and we're cranking away to build the best billing platform around. Come on in, the water's fine!