blog banner
Metrics for Hypergrowth — The Startup Tapes #037

Metrics for Hypergrowth — The Startup Tapes #037


– For people who don’t know Algolia, like, what do you guys do? – Sure, so we are a hosted search engine. So, we will help websites, mobile apps like Twitch,
Medium, Periscope, deliver a great search experience. So let’s say, you know, Google educated all of us, you, me, to expect such a great experience
when you are searching, super fast, relevant. But when you are developer
in your own web app, people expect that from
you, but how to do that? I mean, super complex,
it’s very difficult. So, that’s where we can
help them become successful, become the new heroes there, so that they can deliver on that promise. – Right, and so, it’s like for anybody who has a website, like an
ecommerce site, an application, and they wanna have that search box, and they’ll do a search, they
just do it well, oh yeah, and they get to keep the
data, and all that, right? And they get to control all that? – We’re really an API, so the
data belongs to our customers. We’re not holding anything,
it’s really an API. Whereas I will push the data,
and just tend the queries. – Makes sense. And so you guys have been
growing like crazy, right? And so what data points can you give us? You know, it seems like you’re everywhere, and I see you in more and more places on more and more websites and apps, but what does that look
like from the inside? – Yes, sure. So you know, we are like a
bit more than four years old, we actually launched the
product in September ’13, and since they we have grown
to more than 2,300 customers in 100 countries. We are serving 20 billion
queries every month today. So yeah, so we’ve been super
successful in these months, so that was cool. – And on a very small staff too, right? – Yeah, we are reaching 85 people now, so we are pretty efficient, I would say, compared to like the market. So yeah, we are already
doubled from last year, 78. – Yeah, so that’s kinda crazy,
so how do you track growth? I mean, of course, everybody looks at ARR, and wants ARR to grow, but
what deeper metrics do you use to know whether or not
you’re growing well, or you’re growing as fast
as you could be growing? – So ARR, still, of course. More precisely, I would say
new ARR is more important, actually, when doing just ARR. It’s just the delta that’s important here, because you want that delta to grow too, so that you get that exponential curve. And so what we would track very
closely is that new business that we can add up every
month, every quarter. And then you get like
the traditional suspects, like the churn, like all
about retention of sales, of course, when you have
this negative churn, it’s a super good position to be, because, well, it stacks, it compounds, and helps you to get successful, even if you cannot loan
that many new deals. – Right. And you know, you mentioned earlier that you also look at
stuff like NPS, right, to see how fast you’re growing? And I thought that was
such a good reflection of how organic, in some ways,
your business is, right? How much kind of the bottom of the funnel almost informs a lot about
how you’re gonna grow, and how much your product is gonna spread inside a company, or to other companies, through referrals and everything else. As opposed to trying to attract just completely new business, which I know a lot of startups do. So let’s talk about that a little bit. Like how did your product grow, and why is so much of your growth coming from those bottom
of the funnel metrics, as opposed to top of the funnel? – So you know, I would say
that was our strength at first, really, the product, the
quality of, not only the product like the software, but
also the interaction that our customers would
have with the team. We really cared about
making them successful, and that’s really how we got
to get successful at first. And there, NPS is kind of
the super good indicator. You want leading indicator, like indicator that’s going to
help you predict the future. And NPS is a very good one, we
are between 60, 70 NPS today. That means that we have a
lot of second-hand revenues, coming from people recommending us, coming back when they change company, and that’s really helpful to
grow a community, a user base, that’s going to follow
you year after year. – And that’s also to be very
capital-efficient, yeah. It’s also about that
capital-efficiency, right? So if you have so much of the referral, and to some extent, the
product selling itself, which is never true, but
if a lot of that happens, then that explains why you
can stay at just 85 people, and still have the success
and the reach you have. – But then, you want to
increase that growth even more. And at some point, this
organic growth is super nice, but how can you go faster? And that’s where you want to
experiment with new things. Well, we have very good self-sufficiency, but our company already,
so to get it a bit lower, for the sake of new
experiments, to grow the legion. So we started to do more
outbound, for example, lately, but keeping that only for
enterprise strategic accounts. You know, developers,
they don’t like outbound, so let’s avoid that, but
that could work there. We’re opening new offices. So a lot of new experiments like that, that hopefully will help us
to grow the top of the funnel, and that will then increase our revenue. – So you want to start from
that organic base, right, and find more ways to grow faster, right? And I think it’s so important to have that mindset of experimenting, and looking at different things
that may or may not work, tracking all the indicators you can, and having that mindset that
it’s not just about mistakes and stuff you could improve,
but like brand-new things, that you’re not doing, right? And growth that you’re
robbing yourself of, by not doing outbound,
by not doing expansion. – You don’t know what you don’t know. And yeah, of course you need
to keep that agility, too, to be able to explore new things, and you need to be able to fail. But it’s because you are
going do 10 experiments, that one of them is going to work so well, you are going to develop that one, and then it’s going to become like the key of your future success. – Right. That makes a lot of sense. But so what are some of those failures? What are some things that didn’t work, like stuff you were doing
and you had to stop, or stuff that maybe you
thought was gonna be great, but didn’t end up working so well? – Ah, so many things. I would say like the usual
one is always around people. We should have hired faster. That’s always the… I mean, better be safe than sorry. But we could have probably grown faster if we had been more aggressive at building some of the key functions. I was thinking with
marketing, for example, it took us a long long
time to really staff up our marketing team, which
was like a shame, in a sense, that it was a missed opportunity there. But on the other hand, it actually may have had
some positive impact. Because we didn’t have any marketing, the nice thing is that
everyone had to do marketing. Given our audience, our
community, are developers, and we are developers, we
could count on our team to explore, to do some hacks, to really do marketing to
developers much better. Today, we are going upmarket, much more enterprise customers, and that approach may not work as well. So it’s kind of like
finding the right balance, and like more typical marketing is super important for us now. – Right. Yeah, I think we in startups,
we tend to, you know, underestimate the value of experience. We’re like “Oh, we’re doing
things very different, “we’re brand-new, we’re changing things.” But, you know, there’s
a lot of value in stuff, and a lot of stuff, not
all of it, but a lot of it, does work, even if it’s been the same for 10 years, 20 years. And so I think it’s always something you have to be careful
of, not cargo-culting, not just doing stuff
that’s been done before, but also not neglecting the value of following standard models. – And to learn from your peers. I mean, so many of the
people have done that before, it’s never the same. And so you need to be able to
adapt to your own situation, because simply copy-pasting
what’s worked elsewhere is not going to work, it’s never working. But however, ignoring what’s
happened next is always worse. And you were speaking of failures. Actually, failures are
even more important here. Because if you do something and
nobody else did that before, maybe that was because
someone tried and failed. – You gotta be careful.
But it makes sense, right? Like that focus, again, on
NPS and lowering the churn, and kind of having that
base of organic business, and then that view to kind of experiment, try things that have been tried before, try things that’s never been tried before, but having that framework
of measuring whether or not they’re impacting your
leaning-in-together growth, I think that’s kind of
the gold way, really, to get fast growth in fast. So that makes a ton of sense to me. Thank you so much. – Thank you too. Bye-bye.
– Bye-bye.

Leave a Reply

Your email address will not be published. Required fields are marked *