Startup Juicer. Questions answered by the people making good things happen in startup land


Glen Maddern, Co-Founder and CEO of GoodFilms

Glen Maddern, Co-Founder and CEO of GoodFilmsGoodFilms

Staff: 3
Founded: Mid 2007
Location: Melbourne, Australia

What problem is GoodFilms solving? and why did you choose to pursue it?

At the very core, Goodfilms helps you movies you’re going to be interested in. ‘Interested in’ might mean that it’s a genre you like, right for the mood you’re in, available on your subscription movie service, etc, or likely some combination of several factors. While there’s no shortage of information about each movie online, it still feels like a lot of work to find out if it’s right for you. And that’s where Goodfilms steps in.

Goodfilms is social, in a small circle way. You turn down the aggregate crowd in favour of your immediate network. Path is doing something in the same spirit, is this how you see social going?

Absolutely, you see services like ThisIsMyJam springing up around a similar idea – there’s value in only viewing a small, trusted, curated subset of the information on a topic. To put it in terms of films, knowing which of your friends really loved something and which really hated it, can tell you right away whether you should see the movie. We always believed that kind of close social wisdom is far more important than the average of a hundred critics or a million strangers.

Who is your audience and how do you manage the community so it doesn’t become dominated by one bias?

We have a range, from people who enjoy showing the interesting or obscure films they see, to the people who just like to keep a record for themselves, to people who just want to get the most out of their Netflix subscription. Like any online community, the data can be dominated by our most vociferous members, but by presenting your friends’ data ahead of the masses’ makes that less of an issue.

Our rating system departs from the traditional 5-star system that so often gets abused online (we talked about how Youtube and Amazon ratings break down on our blog right back when we started), and as a result outliers have less impact on people’s judgement. Since there’s no ‘score’ to influence, there’s less reason to give a 100% or 0% rating.

And because the rating system can so accurately identify similarities between you and other people that you might not know, we can filter some of the ‘background’ information to only show you what’s relevant.

Do you have any concerns about using the FB / Twitter platforms? (thinking about the debate and big players cutting off the 3rd party ecosystem)

Well, my cofounder recently weighed in on the debate with an opinion piece for TheNextWeb, regarding, Twitter and Facebook. But we also connect into various video streaming services, which add their own complications, and have their own motivations. It’s a moving target, and it takes up a disproportionate amount of our engineering effort, but the end result is that we’ve been able to build something of value for an individual. We’ll jump through whatever hoops we have to so we can keep doing that.

After identifying the problem, how did you progress?

A couple of pieces of advice spring to mind – one is that if you’re not embarrassed about what you launch with, you’ve launched too late. And, a great phrase I came across just the other day: “If we have data, let’s look at data. If all we have are opinions, let’s go with mine.” (Quote source).

So, I built something that only just hung together enough for me to queue up some movies I wanted to watch, and looked like a backend programmer went crazy with CSS3 (that’s exactly what happened), and started showing people. It lurched through several iterations, but showed enough promise to get into a startup incubator here in Melbourne (AngelCube), and that’s how the whole thing got serious. It felt like I had no idea what I was doing, and in truth I didn’t. I’m sure it wasn’t the most efficient way of getting started, but not having any preconceptions about how something like this should be done helped me just get on with things, and that gave me momentum.

As we’ve grown, different things become important, and now I’m pleased to say we don’t operate quite so chaotically. We still ship as quickly as possible, but with a better vision for how things achieve an end goal. We don’t follow any rigid process (we don’t have a kanban board or anything), but we do use few Lean techniques. We just pick the most important thing that’s not working for us right now and focus on it.

Your growth model depends on a viral invite model that can be tricky. What techniques are you finding works in encouraging adoption?

It’s tricky, without a doubt. A lot of the tests we run are designed to increase the rate at which this happens, which works out to be a good metric for the health and usefulness of the application anyway. It’s a balance between making the site as useful as possible as soon as you sign up, and showing you how it could be better with your friends on board, and so far changes in copy at particular points in the flow have made a surprisingly big impact – tailoring our message to be friendly and encouraging to our users. Listening to user feedback is absolutely essential here.

Can you tell me about your engagement model and how you are using that to retain users?

Email is actually a great tool here, and if you wonder why Spotify, Pinterest or, lately, Twitter send you so many of them, it’s because it works. But it’s a fine line, and if you send frequent emails of little value to your users, it’s just spam. I think Quora and Svpply do emails really well, and Spotify and Twitter really don’t.

Specifically to us, we’ll send our users an email when, say, a friend rates a film you want to see, or queues a film because you’ve just rated it. But we have hard limits in place to make sure we don’t send you more emails than is relevant – say, if a friend signs up and enqueues 50 films you’ve rated, you’ll just get one email.

I really like Ryan Singer’s take on this – email is an extension of your interface. The guidelines that help you show only the most relevant information on a page also work for sending email, so that’s how we roll.

What’s the underlying revenue model?

We drive revenue through affiliate sales to transactional streaming services such as iTunes, or by referring people to subscription services such as Netflix. That’s an ideal mid-stage revenue source, as it grows as we do, however we anticipate different sources becoming possible at a larger size.

How do you identify and prioritise next steps?

We’ve used UserVoice as a feedback forum right from the very beginning, and it’s remarkable how well it’s worked for us. We also email many of our users, at different levels of engagement, to ask them how they watch movies, how they choose what to watch, what Goodfilms is doing for them now and what they wish it would do.

We’re obviously incredibly resource constrained, so we break things down into pieces and try to ship something meaningful to someone every day. It’s the best way to keep momentum up, even if your velocity seems low.

What tools and apps do you use?

For monitoring and analytics we use NewRelic, Google Analytics, Gauges, Wormly and BugSnag, and I’d happily recommend them to anyone. Code’s hosted on Github, Sempahore runs our CI server, and Postmark sends our email. UserVoice for feedback, Disqus for blog comments. It amazes me how many problems we don’t have to solve because of how good those products are. High fives all around.

What is the Goodfilms technology stack?

We use Rails and Postgres, and a great little Ruby search enginge Picky. We also use AngularJS on our mobile site.

Who are some of the voices (bloggers/speakers…) you find most influential?

The one class of blog post I find the most influential and encouraging is people in the early stages of their own startup. The wisdom being passed down from successful founders is extremely valuable, of course, but the ones I most enjoy is the honest self-reflection of people in a similar situation to myself. Particularly those from people who have gone on to succeed – the one that springs to mind is Ben Pieratt from Svpply’s post after raising their seed round last year.

What’s the best piece of advice someone’s given you?

That it’s our f***ing company. In the context that it’s the only way you’ll weather an incubator program, but it applies more broadly – you need to listen to all the advice you can, but you can’t externalise decision-making. When you meet someone with the insight of having been there several times before there’s a temptation to defer to their advice, but at the end of the day it’s your company, not theirs.

What other startups are exciting you right now?

I think the world of travel startups continues to throw out innovative ideas, and I’m pretty pleased that Australia has produced 3 of my favourites – Adioso, Rome2Rio and FlightFox. But, in my mind, the most important startup of the last few years is Kickstarter. I admit I started as an extreme skeptic, but have been won over by the quality of the projects being backed. And I feel like it’s just getting started…

What book would you take to space?

John, my cofounder, put his hand up to answer this one, because he feels pretty strongly about it. He votes for Phil Rosenzweig’s The Halo Effect, since it sets a good framework for consuming all other business books, and helps you separate the really useful information from what they call ‘business fairy-tales’.

More info

Glen on Twitter @glenmaddern…on GoodFilms

Check out GoodFilms, their blog and twitter

^ GoodFilms out of AngelCube

Get the latest Juice direct to your inbox

For more Startup Juice sign up for the newsletter.

One response to “Glen Maddern, Co-Founder and CEO of GoodFilms

  1. Pretty! This was an extremely wonderful post.
    Thanks for providing these details.

Leave a Reply

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

9 − = three

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>