Algolia (YC W14) raises a $18.3M Series A | Hacker News Reader

Algolia (YC W14) raises a $18.3M Series A

techcrunch.com

123 pointsredox_47 comments

I'd love some insight from the Algolia folks on what metrics it took to put this massive round together (esp considering the 2nd seed). Obviously, it means that you had middling numbers for the 2nd seed that you've managed to turn into fairly substantial growth (congratulations!).

$2m ARR with 25% MoM growth?

Thanks! Pretty good guess, you're close :)

Can you give more specifics, esp around the monthly growth. The difference between 20% and 30% is pretty massive there!

30% on average in 2014, lower today. Sorry we don't disclose specifics (at least for now). Happy to share experience offline however.

No worries. Those are fantastic numbers, congrats!

How did you come up with $2M ARR, my usual assumption is a startup will raise what they expect to make the next 18 months. In this case that would be close to $1M Monthly Recurring.

Algolia is mind blowing. We are replacing mysql backend with algolia and its bringing down latencies from 300-400 ms down to 10-20 ms. Things like out of the box multi location support and full text search etc makes algolia compelling choice.

All this at $49/ month is actually a good deal!

I wish them well.

Did you try Solr or ElasticSearch before deciding to go with Algolia? Comparing them to MySQL is a bit of an apples/orange comparison, I would like to know how they compare to other search platforms.

Well. Setting up your own Solr/ elastic search would cost lot more than $49. Here you have a benefit of having platform as a service too

Have you checked out Crate.IO? https://crate.io/overview as an "SQL for Elasticsearch" option.

Congratulations!

"Originally from France, the company is now headquartered in San Francisco."

Hmm, is it really advantage to move engineering to Bay Area from France. On LinkedIn it looks they have more than half of company employees are in Paris. Given large cost of living in Bay Area and how hard is to hire engineers there, it may make sense to grow their Paris office.

Thanks :) Actually, we intend to continue growing the company in Paris too. The product is developed there.

Good luck raising funds in France :(

Docker also moved from France to San Francisco and if they hadn't done so, they would be dead by now.

Congrats to Algolia. They have a killer product, backed by a really talented team.

The 'dailymotion' story certainly didn't help the French tech scene.

Yes, it's a huge advantage. It's unfortunate but that's the way it is at the moment, especially if you plan to get much bigger. Europe has very few large tech success stories. Of course if all the successful companies move their headquarters to the USA that will continue to be the case but there are certain very hard to replicate properties of both the Valley ecosystem and the USA in general (access to both capital and a very large audience with one dominant language) that make it an option you ignore at your peril.

Note how it was Accel partners and not some French VC for this round.

There are a number of examples of foreign owned successful companies that appear to the public as a US success story (Logitech for instance, which in fact is a Swiss company).

I've never had a problem implementing realtime search. Can somebody give me an example use case where algolia is a better solution than doing it yourself?

Clearly it's a popular, in-demand service -- I'm genuinely curious as to the types of apps/websites this is best suited for.

Companies that don't employ people as talented as you probably are. There are lots of them, and they have plenty of dollars to spend on services that fulfill their needs. Often, technology is not the core strength of these companies.

I really want to use Algolia, but their pricing model just sucks. What's the point of a 14 day trial? If I'm going to spend a couple days hacking something using Algolia, I wouldn't want it to end in 14 days. And it's extremely expensive..$49 a month for the starter package? No thanks.

Their free plan can only store 1000 rows..that's really small. What they should do is at least 100k row for free (I'd even go to 1 million), but charge me the moment i surpass X amount of requests. That way I would be comfortable using Algolia from day 1 of my startup and stick with them forever.

If you're building a startup, you should consider $49/mo to be a trivial expense and just pay it. I'm aware that you'll need to use dozens of services and the cost may well achieve $1000/mo, even in the early days. Doesn't matter, just pay it.

Startups are about growth. Paying for SaaS allows you to focus on your product and to grow much more quickly and aggressively than you would otherwise. It's silly and wasteful to even think about not spending $49/mo for the kind of leverage you'd get from Algolia.

Nobody said my startup is going to make it.. I need to test it out on the side, let it run a couple of months before I make a real move. And if it's limiting me on 1000 entries my startup won't be able to work on it's full potential

With anything you're trading money for time. If you don't want to pay their more than fair asking price, then use your time to roll your own solution. It probably won't be as good but, as you say, your business may not succeed so why allocate resources towards the best solution before you've even proven your business model?

Because their solution is KEY to my early stage startup. Fast to iterate, and easy to scale and debug.

If it's key to your startup, pay them the money they deserve.

Well, if you just need a couple of months to prove your business, I'll invest the $49/month on your behalf. What are you prepared to offer in return for my investment?

As someone who bootstrapped a "startup", there are plenty -- PLENTY -- of non-SaaS solutions that:

1) are perpetually licensed. You don't have to worry that they'll disappear on you, or jack up their prices.

2) are cheaper. far, far cheaper.

3) are just as effective.

4) enable growth by:

4a) Being extensible (server-side code execution enables a lot).

4b) Being economical (SaaS chews up a runway like nobody's business).

4c) Building valuable (in the business agility sense) internal expertise.

5) don't incur the overhead of managing accounts for people across a different provider for every service. If you're growing fast, having a single account switch to flip makes a real difference to new-hire integration overhead and effectiveness.

Hard dependencies on costly software subscriptions is rarely a net win for a startup.

So in this case you favor installing and maintaining elastic search? I don't think so. It seems it would come with more than $50/month cost in terms of maintenance and hosting time, as well as the extra hassle.

There is also the masters of search: https://www.google.com/work/search/products/gss.html#pricing... but looks like the price goes higher when you have a lot of queries. Also you probably can't customize the query result as neatly as algolia

Yes, I'm in favor of installing and maintaing "elastic search".

It's not rocket science.

http://en.wikipedia.org/wiki/Elasticsearch#Users

One thing they all have in common is that they're no longer 2 person shops. When you're at a single digit pre-PMF size, spending time on setting up your infrastructure and tooling is irresponsible. At some point the cost of the SaaS becomes greater than the cost of paying someone to maintain the tool, at which point you can do the switch.

And that point can come very early. We are handling a lot of data/user. We are hosting our ES for ~$80/mo which would cost $650/mo at Algolia for the same volume. Same for logging data. We are hosting our own EHK (Heka instead of Logstash) stack for ~$50/mo (few hours of setup and ~10mins/month of maintenance) which would cost us about $500/mo if we would have stayed with Loggly (and ours will stay at about the same price well into the volume of Loggly's $1k/mo plan).

As a bootstraping startup you don't have the luxury to throw money at problems and a lot of SaaS are not well suited if you have a heavy data workload.

with that sort of volume, you guys must be making millions!

> spending time on setting up your infrastructure and tooling is irresponsible

Investing in our infrastructure and tooling early-on was probably the most responsible thing I did.

$49 per month is 'costly'? That's barely 1/2 an hour of developer time. That's an absolute bargain. If it took a developer 6 hours to build an in house solution, that just cost as much as a year of Algolia. And that dev time doesn't even include the servers or maintenance costs. $49 is nothing. It would actually be irresponsible to not spend that amount because you'd be focusing reasorces wastefully by building something you needn't build. It's like people who hire a Dev Ops guy when Heroku would cost them $2000 per month; instead they'll pay AWS $1500 and then a full time salary to manage it. Or, to put it into perspective; $49 per month is probably hundreds less than a typical ping-ping startup spends on microbrews for the beer fridge.

An in-house solution and the resulting acquired expertise may provide a great deal more inherent value.

And what's irresponsible is using software (languages, architecture, deployment methods) so inefficient that you need to spend $2000/mo to host a minimal deployment.

Tangent:

Three problems in practice with the business model you propose:

1. Most free plan users will never convert to paid, but the company is still on the hook for supporting them; every dollar spent supporting non-converting users is a dollar not spent serving paying customers, so you have to believe you don't have a productive way to spend each free-plan dollar on real customers in order to have a free plan.

2. The free plan users are disproportionately going to be pathological.

3. Having the free plan is going to make it hard for paying customers to immediately opt-in to paid, rather than trudging through the free plan. But trudging through the free plan also makes a purchasing decision more complicated. If you have a product that creates enough value for paying customers that they would be willing to pay for it without a free plan, then having the free plan can actually make it harder to acquire those customers. Paying for something is a forcing function to integrating it and getting it working, rather than having it molder on a stack of todo cards in Trello.

I don't see Asana having a problem with this. The moment the team needs to grow bigger than 10, they have to pay, and that's a good problem to have and I don't mind paying more than 49$ a month because it's too late to change and it'll cost even more if I did.

If I ever happen to have lots of requests, then I too do not mind dishing out way more than their plan to get there.

$49 is expensive? It's not worth an hour of your time each month to offload search to them?

you could implement them in a portion of your application and see how things work out. For instance, have them as the search provider for a couple of your more popular products categories or something and fall back to your current solution for everything else (a little A/B testing if you will). by doing so, you can avoid hitting the limits and see if your ROI increases before implementing or dumping them.

Congrats. I checked out Algolia a while back, and its great to see such progress. I wonder if Algolia and Swiftype will merge because they both offer great products.

how are algolia and swiftype different?

I love Algolia. Used it as a core part of a project I worked on. Very slick.

by the creators of exalead

some exalead employees actually :)

... and HN Search!

Great folks, great technology. Well done Nicolas, Julien and team!

fantastic ui.

we barely use elasticsearch for searches but for statistics and data analytics, will these features also be added?

We already provide analytics about the searches performed on your indices. However, we focus on user-facing search use-cases. We don't recommend using Algolia for processing your logs for example. See this quora answer for an Elasticsearch / Algolia comparison: http://www.quora.com/How-does-Elasticsearch-relate-and-or-co...

Congrats to the team!

Reply on news.ycombinator.com