An Interview with Paul Kafasis, Rogue Amoeba

This interview with Paul Kafasis of Rogue Amoeba is the tenth in a series of interviews I've held with indie software developers about marketing Mac software. Previous interviews: Gedeon Maheux, Justin WilliamsGus MuellerDaniel Jalkut, Rich Siegel, Oliver Breidenbach, Jacob Gorban, Jean MacDonald, and Kevin Hoctor.

Rogue Amoeba makes cool applications like the home audio streamer Airfoil, Audio Hijack Pro, the Radioshift audio recording tool, and the audio editor Fission. Paul Kafasis, whose title is CEO/Lackey of Rogue Amoeba, was kind enough to talk about marketing from his perspective.

While not in front of a computer, which he admits happens too infrequently, Paul enjoys/suffers through distance running, reads voraciously, and attempts writing humor. He can be found on Twitter as @PBones.

DW: First off, can you explain what you do at Rogue Amoeba?

PK: I handle the myriad tasks that let our top-notch programmers keep programming. That's a litany of things, from marketing and PR to interviews to all sorts of boring paperwork. Probably what I enjoy most is product design and development, working with our coders and designer to determine what we'll include in a new app, or an update. This sort of process takes knowledge gained from talking to customers, reviewers, reporters, and more, and distills it down, to shape what the future will be. I find that fascinating, and rewarding.

DW: Tell us about the marketing activities you do.

PK: Well, one thing we DON'T do is a lot of ads. Ads are expensive, and the ROI [Return On Investment] is very difficult to measure.

We exhibit at Macworld every other year, and we've done that since 2004 (so, 2004, 2006, 2008, and now 2010). That's the single biggest expense, as far as the marketing budget goes, costing $10-$20,000 in a single go. We also do some sponsorships and traditional ads, though not very often. That's mostly it, as far as "traditional" marketing goes.

So, what else do we do that I would consider marketing? Plenty, really. We provide free software to folks looking to review it, as well as for use as raffle prizes for sites big and small. We have a popular weblog with a mix of company news, industry analysis, and controversy. We've got a Twitter stream where we publish most updates, and engage at random with users, though I can't say we've really captured the potential Twitter might have.

A quiet but important piece of our marketing is our newsletter. We send our most important updates to that list, which contains many, many, MANY users. We're VERY careful about how often we mail to this list (1-3 times a year), and we try to make sure there's something useful in it for everyone, every issue. Minor updates aren't included. We send this out only for major updates and new products, with perhaps a mention of smaller updates included as well.

You've covered newsletters before, and I would only add that this has been a secret weapon of ours for 7 years now. The sales bump we see when we put out a newsletter a few days after a major update is usually as large as the one we see from all the press coverage the day of a release. The key thing is the quality of our list - it consists almost entirely of users who've tried and purchased at least one of our applications. These are thousands of Mac users who pay for software, and like OUR stuff. That's golden for us.

One other thing worth noting is that we strive to have top-notch technical support, for current and prospective customers alike. That is, most assuredly, a form of marketing. When you impress users, they remember you, they remember your product, and they'll help spread the word. Word of mouth has always been our very best form of advertising, and top-notch support is a big part of getting positive word of mouth. With the explosion of blogs and Twitter, word of mouth is becoming ever more effective.

DW: How does exhibiting at Macworld work for you guys?  That's a big expense, so do you think you are recouping the costs?

PK: I had a four part series of posts about MWSF a couple years back which I think are still a good overview.

As far as recouping the costs go, it depends on how broad a view one takes. From straight sales, to people we talk to? No, we don't recoup our costs. We get back anywhere from 10-50% of costs there (it varies, and the "exclusive show special" coupons leak online). But if you step back, and look at the exposure it gets us? It's absolutely worth it. Macworld draws the eyes of, well, the Mac world - hundreds of press folks, MUG leaders, and general tech folk. These are all the types of people who pass on the word about our software to many others. Being at the show puts us in front of these people. The goal is definitely long term, in getting mindshare.

DW: Any reason why it's biannual?

PK: The biggest reason is that it's a whole lot of work. The show floor itself can be exhausting, and the setup beforehand is a lot of work as well. 

DW: What is the "low hanging fruit" that other developers ought to try in marketing?

PK: I think providing excellent support is something which is all too-often overlooked. I haven't seen it recently, but I recall seeing companies that provided support only to licensed users. This is just a terrible idea. When a prospective user emails support, they haven't given you any money yet, no. But this is someone implicitly saying "I am trying to use your product. I am having a problem, but I'm interested enough to take the time to email you. If you fix the problem, I will buy the software". That last bit is so crucial - they've got a problem and they want you to fix it. Once you do, you're likely to get a sale.

Even if developers are providing support to all customers, I think there can still be a lack of focus there, particularly in one-man shops. For one-man shops, there are a million things to do, and it can be easy to let support fall low on the last. This is a tough problem, there's no doubt about it. Rogue Amoeba was fortunate to have three people from the get-go, which made splitting up tasks much easier, and let me focus a good portion of my energy on front-line support. I think this helped our reputation in the market immensely.

DW: I'd like to get your current thinking on the state of the "delicious generation" which you had described as apps that "are heavy on the marketing buzz and light on substance". I take it from this that you think it's more important to have substance?  :-)   How  have things changed since you coined that famous phrase?

PK: Ah, there's a term I haven't heard in awhile. Looking back, I think the idea of what I was trying to express is important, but the phrase alone was too vague. It wasn't an insult, nor a badge of honor, but it became both. The simple idea was that style wasn't enough - applications need substance as well, and that substance should be the first thing added, not an after-thought.

I certainly think substance is what needs to come first, yes. You can sell the sizzle, not the steak, but you still need to have a steak. If you're selling sizzle, and you've got chuck beef, that's not going to end well.

I think one thing to notice is that many of these applications flamed out, and haven't seen any of the promised updates or improvements. They sold very well for a short time, which is obviously great. But while focusing so heavily on the sizzle worked to sell a product, few people built a company around these products.

Perhaps that's ok, but as a user, I want to buy a product from a company who intends to stick around and keep developing the software. Using an application represents an investment of time by the user, the time spent learning it. If the user has to keep switching applications, a whole lot of time is wasted.

What I think is really interesting is how well the Delicious Generation-type of apps fit on the iPhone. The platform is perfect for relatively narrow apps that do one thing beautifully and well. Personally, I prefer to make deeper applications that encompass a bit more than the iPhone allows, and as such, the iPhone hasn't been a very appealing platform for us. For others, however, it's been great. There's some very cool stuff being done on the iPhone, things like TapBots' apps or an app like Classics.

We've long worked to add more flair, in addition to more substance, in our apps. This hasn't been our strong suit, but we've gradually improved. Software doesn't have to be functional yet boring - it can be both functional and fun, and that's our goal.

The trick is finding the right balance, and personally, it's always seemed to make sense to err on the side of substance, not style.

DW: You talked about measuring ROI. How do you measure success of your various marketing activities?

PK: The best measurement tool we have is coupon codes. We include a unique coupon with most every type of activity we can, and then track the sales which used that coupon. So at Macworld, we gave away a Macworld-specific coupon, and then watched sales related to that.

The biggest problem is that these coupons leak out - ultimately, that's ok. It's likely to encourage new sales we might have otherwise missed. Further, we never make a coupon so big it's actively hurting us, so if someone who was going to buy anyway finds it, that's still OK.

I wish it were easier to measure the effectiveness of various things. Far too much of it comes down to gut feelings and anecdotal evidence. That said, when something really works well, it's usually pretty clear, even with our limited measurement tools.

DW: So have you found any marketing activities that aren't beneficial?

PK: Hmm. Most online image-based advertising (banner ads, badge ads, etc) aren't very beneficial, I'm afraid.

DW: Since you are all about measuring, can you give us an example of how you would judge that?

PK: We're always interested in the cost per click. If ads are $100, and we get 1000 clicks, that's 10 cents per click - that's not terrible. With an average sale of around $28, we need one in 280 people to create a sale (usually by buying, but perhaps by telling a friend), for the ad to be paying for itself. The lower the cost per click, the more effective the ad is likely to be.

For a couple years, I measured this, using unique URLs. I calculated the cost-per-click and then we shifted advertising to sites with the lowest cost and highest traffic. The problem is that for most sites, the cost per click wound up being quite high, $0.75, $1.00, $2.00. Maybe that's effective, maybe it's not, but it was more than we wanted to pay. We've stopped almost all of our banner ads because of this.

Where does the blame lie for this? Is it with our own ads? Maybe. But I question how many readers actually click ads, period. I know I sure don't, or I do so very rarely.]

There's an argument to be made for branding - Coke doesn't have ads because people have never heard of Coke, they have ads to keep their brand awareness high, to keep Coke on people's minds. I think there are much more effective ways to do that however. Work to have the best products. Update your products frequently with new functionality. Treat your customers well, and provide top-notch support. Support the community as a whole in various ways. All of these help word spread about what a great company you are.

DW: One last question is just your thoughts about your website, since that's the "face" of a company to the world (unless people meet you guys at Macworld or something).

PK: Oh man, I could probably give you 3 pages on that alone. I worked extensively with Dan Cederholm on that design, he was great to work with. He listened to me, looked at some example sites I gave, and focused where we wanted to focus.

In short, we knew the home page needed to have our major products and their associated logos. That was THE key. Beyond that, having links to the rest of the site, as well as news items, was something we've done since day 1. 

There are other options for showing news. I've always liked the giant graphic found on or Panic's "What's the latest?" bit, but at the time, we didn't have a full-time designer, so text was our best bet.

I think it's important to have a company site, if you plan to make a business. It's likely you're starting with just one product. Even so, establishing yourself as a company and not just a product, provides a better impression for visitors.

DW: Thanks for sharing your ideas and opinions with our readers!

Copyright © 2005-2014 Karelia Software. All rights reserved. · Terms & Conditions · Privacy Policy · Site Map