April 28, 2008
CR Newsmaker Interview: Travis J.I. Corcoran of HeavyInk.com
This interview was pitched to me rather than something I came up with on my own. As I explained briefly last week, I try not to do a lot of CEO/businessman-type interviews. First, I'm more interested in art, and when I apply the questions I find interesting about art to businessmen, they tend to have a lot of very slick-sounding answers. Second, while I like exact figures, and I sometimes think business folk are genetically vague on those kinds of thing, I also understand why some information needs to remain proprietary. Third, I lack the kind of background that allows me to easily check a lot of the claims made or to automatically catch a leap in logic that maybe shouldn't be made.
All that said, I liked talking to Travis Corcoran about his social networking sales site, HeavyInk.com
, and I think this interview can at the very least be valuable in establishing a baseline for comparison with what Corcoran might say in future years, and in putting his thoughts out there for others to scrutinize. – Tom Spurgeon
TOM SPURGEON: First of all, can I get some background on you in terms of your relationship to comics?
I grew up reading "real" books, and never
comic books. If I thought of them at all, I thought of them as sad cousins to "real" reading.
... and then in college (this is around 1992), an apartment-mate loaned me Watchmen, and I realized just how serious, and exciting, and artistic comics could be. Pretty soon I had borrowed Give Me Liberty
, and then I moved on to Ronin
, and before too long I was going to a comic book store to pick up new issues of Concrete
, Cheval Noir
, whatever I could find by Moebius, etc.
... and, speaking of comic art, probably my favorite possession is a framed signed print by Moebius (so what if I had trouble buying food that month?)
SPURGEON: Of what kind of comics and which cartoonists might you be a particular fan?
Even after getting involved in comics, I looked down on superhero comics...until I discovered The Authority
. Now I'd say that there's no kind of comics that I totally avoid, but my interest is still a bit outside of the spandex-and-superpowers stuff: right now I'm subscribed to two dozen or so comics, including my absolute favorites:
* Ex Machina
* Walking Dead
* Pax Romana
* Atomic Robo
* Mouse Guard
* Nearly Infamous Zango
* The Flight
* pretty much anything by Warren Ellis
You can see the whole list at my profile page at HeavyInk
SPURGEON: What was the impetus for starting your original on-line company SmartFlix?
I've done a bit of woodworking, and was interested in metalworking (lathes, mills, that sort of thing), but didn't have any way to learn those skills. I found out that there were some instructional videos available, but (a) they cost $70 each, and (b) there were no reviews, to tell me if they were even worth it. I decided that there should be some way for folks to rent these videos for a lot less. I started SmartFlix on sort of a whim, with no huge expectations for it. The first month it was up, I rented out one video. A few months later, I was renting perhaps 10 or 15 videos per month...and now, it's an 11 person company, with customer support, a fulfillment team, and a four person engineering team... all of which put me in a perfect position to do something really interesting, like HeavyInk.
SPURGEON: Where in that particular market did you see a need and for an outsider, how would you describe your basic strategy in servicing that market?
SmartFlix launched a little while after Netflix, and I was inspired by their model, and copied the customer support ethos from acknowledged great companies like LL Bean. From a customer's perspective, our SmartFlix strategy is pretty simple:
* grow the inventory
* provide great service
From our own perspective, though, there are two more bits of strategy:
* continuously do A/B tests of the service and the website
* relentlessly mine our data to figure out how to serve customers better
Let me explain each of these points in turn (these may seem a little boring, but bear with me, because each is highly relevant to the comic book world!)
A/B tests: If you come to the SmartFlix website, you see a certain graphic design, a certain navigation scheme, certain language, etc. ...but another customer coming to the site might see a slightly different version of this: we might have a different navigation scheme, or the pictures of the videos might be smaller, or whatever. We keep track of all of these semi-random changes, and then we see which ones made it easier for customers to find the videos they were interested in. We thus continuously tune the website to make it easier to use.
Datamining: we look for correlations which allow us to make recommendations, or suggest better choices.
You can already see how this applies to HeavyInk: first, we can tune our website over time to make it easier to use, and second, using information about how you rate comics, and authors and artists (yes, you can rate and leave reviews on lots of things besides just comics!), we can start to make predictions about new things that you might like, and recommend them to you.
The recommendations aspect was actually one of the core features that I wanted to deliver in HeavyInk: because I'm a comic book fan with slightly out-of-the-mainstream tastes, I often find it hard to get good recommendations for new things to read (the owner of the local comic book store that I used to go to had "The Fantastic Four" as an
answer to every
question)...so I really wanted to deliver this feature.
SPURGEON: What does "inspired by their model" mean? To my uninformed ear, that just sounds like you ripped them off, Travis.
After Netflix pioneered the model, a lot of folks followed in their footsteps: Blockbuster, GameZnFlix, Redbox, GreenCine, CinFlix, etc., etc., etc., all the way down to a bunch of niche folks renting out, say, just martial arts DVDs. So, we were one of this crowd, but more successful than most, and I think the reasons that we were more successful were:
* excellent customer support
* good bargains for customers -- rent a $90 dvd for $9.99.
* willingness to be bold and really take bets on expanding into the space (mortgaging my house and buying hundreds of thousands of dollars of inventory)
* doing something somewhat different (the whole how-to angle -- not just another clone of Netflix offering nothing but mainstream movies)
* good online advertising
* a blog and newsletter with interesting stuff -- customer interviews, vendor interviews, a column by humor writer James Lileks, etc.
* lots of experimenting with the website to see what works best for people
* fiscal discipline in ramping up the company (although this sort of contradicts the "willingness to be bold" up above!)
SPURGEON: Travis, can you provide some actual figures as to what your goals were that you claim to have met?
We wanted to have 100 customers in November, and then get 70 percent growth in December, then 30 percent growth in January, February, March, and April (so that by the end of April we'd have a bit less than 500 customers).
In fact, we've got just short of 2000 customers.
We're exceeding our sales goals as well -- this month's sales annualize to $120k/year -- but by a smaller margin. All in all, this is great news. Those customers who are signed up, but aren't (yet) buying are deep in the sales funnel: we've got their interest, we've got their names, they've opted into our newsletters and our social network, they stop by to read author interviews, etc.
SPURGEON: Are you in the black on the project overall at this point?
The plan from the get-go was to break even on sales (sale price minus cost of goods, shipping labor, envelopes and postage) for the first two years, and consider requiring minimum shipments size for free shipping in 2010 or so.
In fact, we're slightly exceeding that break-even goal.
Of course, there's a significant engineering effort involved, and 2000 customers don't let us pay the salaries of our engineers ... so we're deep in the red on that score (which was the plan all along). This is a long-term investment, and we'll earn back the cost to build the site after we've grown a fair bit. Two years from today we expect to have 18 times the volume we have right now, and at that volume, if we can tweak shipping expenses or negotiate a slightly better deal on purchasing our inventory, so as to increase our profit from roughly 1 percent to roughly 10 percent, then suddenly we're not just paying our engineers, we're making a profit, and paying back our initial investment.
Of course, that's an audacious goal -- not many comic book stores are doing $2 mil/year in revenues... but given our experience, our customers' enthusiasm, and our growth to date, we're quite confident we'll hit that goal.
SPURGEON: How much volume do you have to do in a sales month to pay overhead and operational expenses while offering 20 percent discounts and free shipping?
Because of our volume, we're getting a pretty nice discount tier on our inventory and on our shipping supplies (50 percent off of list for boxes, etc.), and because of our experience in shipping and our facility with navigating USPS regulations (having a USPS permit imprint account, a USPS business reply account, knowing about pre-sort discounts, etc.) our postage costs and labor costs are lower than those of most folks in this business.
Because HeavyInk is the second venture of an existing firm, we don't have fixed overhead -- if HI uses 10 percent of the floor space, and uses 10 percent of the labor of our fulfillment staff, we just pay for what we need. This is a nice situation -- if we had to go out and rent space, and/or hire someone for 40 hours/week, then we'd have a fixed overhead that would really suck the money down. Instead, we're in the nice position of being able to ask folks for a few more hours each week, and paying for just what we need.
Anyway, the direct answer to your question is, "As long as we're selling a couple of hundred comic books and graphic novels each month, we can break even -- and we're well beyond that."
SPURGEON: When coming up with your basic set of strategies for HeavyInk, what was valuable about the SmartFlix experience that could be applied and what did you know was very different?
There are a lot of different things going on in the online comics world, from discussion boards to online sales and subscriptions, and more.
The core idea of HeavyInk is "as you deliver each of these features in turn, the next feature is simultaneously easier to implement than the last one, and delivers more value
because it ties into a network of existing features."
If you've already got a comic book store, then you add a social networking feature to that, suddenly I can learn what my friends are subscribed to.
Once you've got a history of my orders and subscriptions all in one place, then with a bit of work, you can help me manage my collection.
Once you've got social networking and collection management, then I can see what graphic novels my friends have that they can loan to me.
Once you've helped me organize my collection, it's easy for me to rate everything I own, which gives my friends information they can use.
And so on and so on.
That's what we intend to deliver -- and we've already rolled half of it out, and we're still ramping up hiring to help us deliver the rest -- but you asked "What's the basic set of strategies?"
What we did at SmartFlix resulted in a very enthusiastic user base, and we want to replicate that.
* Customer support is our first priority. We don't care if we lose money on a given transaction -- our first goal is to make sure that the customer is satisfied with his or her experience. If we deliver a good experience, the customer will be back, and will tell friends.
* Continuously roll out new features, and perfect the old features.
* Listen to what the data tells us. Why send a one-size-fits-all newsletter, if we can tell from your subscriptions, or your ratings, or your circle of friends that you're more interested in this
kind of interview or news item than that
* Work with new artists and authors, small publishers, and independents (including those without Diamond distribution) to get their products out where comics fans can see them ... and can buy them. Diamond often serves as a bottleneck -- at one point, the demand for Red5's Atomic Robo
was growing and growing, and most comic book stores were sold out, and Diamond had no inventory, and -- it seems -- no answers. We got on the phone with Red5 and ordered several hundred copies of all the issues. We had them in hand two days later, and they went out to hundreds of Atomic Robo
fans later that day. That's the level of service that I, as a comic book reader, want, and it's the level of service that all comic book readers deserve.
SPURGEON: Can you provide information as to which copies of Atomic Robo you ordered and at what time so that I may potentially double-check that it was not available from Diamond at the time you claim? With whom did you work on this matter at Red5? At Diamond?
We started having problems with Diamond having stock of Atomic Robo around January 3rd. Our software automatically checks the in-stock status of things at Diamond and tries place orders daily. From January 3rd through mid March, Diamond didn't have copies in stock.
We placed several orders with red5.
January 5th, 2008:
Robo Issue #2 -- 50 Copies
Robo Issue #3 -- 50 Copies
Robo Issue #4 -- 25 copies
January 16th, 2008:
Robo Issue #2 -- 100 Copies
Robo Issue #3 -- 100 Copies
Robo Issue #4 -- 100 copies
January 24th, 2008:
Robo Issue #1 second printing -- 100 Copies
February 4th, 2008:
Robo Issue #1 second printing -- 20 Copies
Robo Issue #3 -- 50 Copies
Robo Issue #4 -- 50 copies
Robo Issue #5 -- 80 copies
March 13th, 2008:
Robo Issue #1 third printing- 100 Copies
Robo Issue #2 second printing -- 100 Copies
Robo Issue #3 second printing -- 100 Copies
Robo Issue #4 second printing -- 100 Copies
Robo Issue #5 second printing -- 100 Copies
Robo Issue #6 second printing -- 100 Copies
We also started ordering other stuff from Red 5:
March 27th, 2008
Neozoic #1 -- 20 copies
Neozoic #2 -- 20 copies
Neozoic #3 -- 20 copies
Abyss #1 -- 15 copies
Abyss #2 -- 10 copies
Abyss #3 -- 10 copies
Our data on Diamond stock levels came from their retailer website.
Our customer support rep at Diamond is Kathy Flemming, but I don't think we spoke to her on the topic -- at that point we were already fully up to speed on using Diamond's retailer website, which is how we continue to do reorders daily.
SPURGEON: Are you aware of the failure of Next Planet Over, a failed on-line retailer from the previous Internet business age?
Before I made the decision to launch HeavyInk I did extensive investigation of the marketplace. We did competitive research -- purchasing comic books from the top 10 or so online stores, checking out both the strengths and weaknesses of local comic book shops, talking to comic book creators, doing an online survey of 500 + comic book fans, working up lots of spreadsheets, etc.
I can't speak to the details of Next Planet Over, but I will say that we went into HeavyInk with
* our eyes wide open
* a first class team that can do more with less
* a boatload of experience hard-won from the process of launching SmartFlix and growing it to a very nice size without a single dollar of venture capital
* a good knowledge of what our technical and other strengths are (and what they are not)
* a fiscally prudent attitude that "web 1.0" startups were not exactly known for.
I can say that when I look back at old press releases, I don't quite understand why a webstore selling comic books needed to have a CEO, a vice president of merchandising, a general counsel, a vice president of business development, and a vice president of information systems.
In the Marine Corps, every marine is a rifleman.
At SmartFlix / HeavyInk, every employee is a do-er, not a manager.
SPURGEON: How has it gone so far? Has the bottom line met your expectations for this early date?
Our plan has always been to discount nicely (20 percent off all orders) and ship for free (yes, even if you order just one comic book)...and use our recommendations engine and social features to show folks some of the great stuff that they really want to be reading, but they don't know about yet. Our online survey of comic book readers showed us that a lot
of people are ready, able, and willing to spend more on comics, if only they could find more stuff that they like.
That being said, our plan was aggressive: we had certain targets for the first month, and then 70 percent / month growth for a while, and then 30 percent/month growth for a while after that.
So far, we're six months in, and we've matched or exceeded all of our goals.
...and, frankly, that astounds me. Eisenhower once said "plans are useless, but planning is indispensable." With that in mind, I fully expected the plan to be garbage...but, amazingly enough, it's been a perfect road-map.
SPURGEON: Is there anything that's been particularly surprising about your first few months out of the gate in terms of fan reaction?
I feel a bit like I'm in a job interview: "What's your biggest weakness?" "Either that I work too hard, or I care too much..."
I really hate answers like that, but -- seriously -- the biggest surprise has been how much our customers are pulling for us. When we roll out a new feature, we get customers stopping by our personal profile pages, or leaving posts in the forum thanking individual
developers by name.
Before SmartFlix and HeavyInk, I was a software engineer writing big server applications, and I never heard of someone sending an email "Hey, guys, the new caching code in the NFS bridge is teh r0x0r!"
So...the great fan reaction was a surprise, but of the very best kind.
SPURGEON: What's the greatest issue facing the site right now? Is it just getting your name out there? Working out the kinks? Improving content? Finding out which features work and what don't?
One thing I've learned about working at a small, self-funded company -- there are always three dozen greatest issues!
We've got all the usual bug fixes -- we've coded up handling for alternative covers, and incentive covers, but then we find out that there are also rare high-list-price incentive covers, and our code doesn't handle that.
We've got a list of features as long as your arm -- we want to ship worldwide, we want to deploy collection management tools, we want to give authors and artists more tools to manage their talent pages, we want to host previews of issues, etc.
We pride ourselves on doing data-backed decision making, but we haven't (yet) deployed the full set of tools that we developed for SmartFlix -- the corporate dashboard (easy insight to all our data), the A/B testing, the data mining, the statistical significance and correlation tools, etc.
We want to make available related products that customers are asking for: figures, card games, posters, etc.
We've got people begging us to do an affiliate program.
Will we get it all done?
Coolidge once said:
"Nothing in the world can take the place of persistence. Talent will not; nothing is more common than unsuccessful men with talent. Genius will not; unrewarded genius is almost a proverb. Education will not; the world is full of educated derelicts. Persistence and determination are omnipotent. The slogan press on has solved and always will solve the problems of the human race. No person was ever honored for what he received. Honor has been the reward for what he gave."
I hope it doesn't disappoint Coolidge, but we're not setting out to solve "the problems of the human race"...but we are setting out to solve a lot of the problems of the retail comic book world.
We've got years of hard work ahead of us, but we're up for it.
SPURGEON: At what point do you think you'll be settled on a model for the long haul, or are you already there?
Our front page gives our five word mission statement: "HeavyInk is learning, shopping, sharing." There's a ton of work to be done there, and a ton of value to be delivered to our customers.
That mission statement is our guiding star for the long haul.
SPURGEON: In theory, how does the social networking aspect of the site lead to more sales?
It's not that we want to just make more sales -- we want to make better sales. If I could either browbeat you into taking on one extra comic book subscription this month, or tempt you into taking on the right
comic book subscription one year from now, I'd far prefer the latter. You're going to be enthusiastic about the right comic. You're going to love that comic, and recommend it to your friends, and maybe buy the graphic novel when it comes out.
The social networking, then, isn't about selling you more stuff -- it's about exposing you to the right stuff.
SPURGEON: Let me rephrase the question about social networking and sales, then: if it is not intended to lead to more sales, how does social networking lead = to better recommendations? My friends are aesthetic morons with terrible taste in art.
Ha! Yeah, I've got a few of those myself.
SPURGEON: More importantly, how is social networking as a vehicle superior to all the other ways of having stuff recommended, or simply looking at stuff either on-line or in a comic book store.
OK, good questions.
Obviously, if you've got access to a comic book store that stocks absolutely everything that's in print, and you've got the time to flip through everything there, then that's the absolute best way to find out what you like.
Or, falling short of that, if you're into comic books enough to know who your favorite authors and artists are, then you can make pretty good guesses (although, even there, HeavyInk can help you -- we give each community member a personalized RSS feed, and let them add and subtract things from their feed -- for example, you can add Paul Pope alerts to your feed, or Warren Ellis, or Walking Dead
, and then get a feed that tells you of new things in each of these areas).
Next, we can do a traditional recommendation engine -- based on how you rate things, we can say "people who liked X and Y also liked Z." And, in fact, we've got such a system in place, and we're updating it with a newer and better version this week or next.
Using social networking data gives a few things above and beyond the traditional reco engine:
* for folks who don't know where to start, getting some information on what their friends are reading at least gives them a pointer, saying "you don't have to buy it, but at the very least, be aware that person X thinks that there's something good about Y." This may not be that interesting for someone like yourself who knows the comic world inside and out, but our surveys show that there are a lot of comic neophytes who are interested in reading more comics but want better ways to learn about titles.
* there are studies showing that adding independent data sources to a recommendations engine can increase the accuracy of the prediction results much better than merely increasing the quantity of the first type of data. Which is to say that we can use the social networking aspect to augment the other data, and do an even better job.
And, of course, your point about "I have friends with terrible taste" is well taken, and at some point, we want to allow you to mark a friend as "a friend, yes, but not someone who I'm interested in hearing the opinions of"...but it may be that we can do this algorithmically (if you have a friend with horrible taste, and we know that you've disagrees about a few things, we can actually use his recommendations in a constructive
And, regarding your point about "looking at stuff online or in stores," we have plans to host online previews. We've already locked down permissions from a few publishers, and plan to deploy this later this year.
SPURGEON: Doesn't your model depend on a system over which you have no quality control? Isn't social networking an unwieldy mechanism for a simple transaction?
If by "unwieldy" you mean "computationally expensive," then yes...but from the user's point of view, how hard we work to dig up recommendations and such isn't really an item of concern -- what the user cares about is the validity of the suggestions. If you look at how Google determines page quality (the page rank algorithm, as modified by 10 years of research, machine learning, etc.) it's also hugely baroque...but it gives excellent results. The fact that Google has to spider the entire web and do an almost incomprehensible amount of work to generate those results isn't relevant.
SPURGEON: Is the editorial content simply there to draw people to the site or is there a PR aspect to it as well?
It's part of an integrated strategy:
* the interviews make our site more interesting, which results in folks mailing around links to the interviews.
* the interviews make authors and artists more aware of us, and often lead to inquiries as to how we can work together
* and, finally, that's the kind of thing I want at the store where I buy my comic books. It's like those horrible "Hair Club for Men" commercials -- I'm not only a HeavyInk employee -- I'm a HeavyInk customer!
SPURGEON: How are you supplying your comics -- through your own warehouse? Diamond? Amazon?
As of today, we buy most of our inventory from Diamond.
SPURGEON: What's the most hopeful endgame for you in terms of how your company develops from here on out? What might be the next sign you're on the right track?
Getting back to the distinction I drew between "web1.0" startups and us: we didn't get into this thinking "we'll acquire two hundred customers, sell $1,000 of comic books, and then IPO for a bazillion dollars." There's one sure-fire way to build a company, and that's to provide real value to your customers, and improve a little bit each day. If Amazon or Barnes and Noble offers us a private island, a pile of gold, and a gift certificate good for being Best Friends Forever with all of our favorite comic book authors, yeah, it's not impossible that we'd sell...but that's not going to happen, and that's not what we're building the company for.
We're in this for the long haul. Ten years from now -- heck, five years from now -- HeavyInk is (still) going to be the best place to buy comics online, or off. We'll have all the features we have now, all the features that are already on the wish-list, and tons of other features that we haven't even thought of yet (but our customers are busy dreaming up).
You asked about the endgame -- there's no fixed place we want to be. Right now, we can see what "learning, shopping, sharing" means a few years out. Five years from now, we'll be able to see what "learning, shopping, sharing" means a bit further out.
Constant improvement, and constantly delivering value to our customers -- that's what we'll be doing.
We'll know we're on the right track next month, and next year, the same way we know it today: people write emails or forum posts saying how much they love us... and then they tell their friends the same thing.
posted 4:00 pm PST
Daily Blog Archives