A short version of this post was originally posted on TechCrunch.

This post is the result of a long struggle, learning how to market a mobile app –  the main struggle most mobile app entrepreneurs experience.  I’m sharing the lessons I’ve learned, in the hope of shortening other entrepreneurs’ marketing learning curve.

I never planned to become a marketer. I was on the product/ Techy/ UX side of things and was quite happy there. In the (first) startup I co-founded, VisualTao (acquired by Autodesk, the product is now AutoCAD WS) we developed a web and mobile CAD app. The technology was innovative, the product was great and the need was clear. We started marketing it by the book – crafting our positioning and working with a PR agency to approach bloggers. This didn’t work. We went at it a second and third time – tweaking our positioning and website once again, adding more product features and writing to more bloggers. Didn’t work – again.
I gradually came to an understanding that when competing with hundreds of thousands of other apps for attention, marketing is not just another ingredient in an app’s success. It’s the main one. We were a small team with a very limited marketing budget, so we declared war the only way we knew how – as engineers and UX designers. The following months we left what we knew about the traditional marketing behind and started exploring new and creative ways to reach new users, and like the engineers we were – we measured each step along the way, down to the last click. These days the product is celebrating 10M downloads worldwide with customers from dozens of Fortune 500 companies.

AutoCAD WS Google Play

Here’s how we reached our first 1000 users, our first 100K, and our first million users.

The app graveyard and how to avoid it. 
When I prepared our app launch I looked up to TechCrunch/ Mashable/ LifeHacker/ other tech blogs with lots and lots of traffic – If only I could just nail it and get their coverage, I thought,  everything will be alright. It wasn’t. The good news is that if you don’t make it to the big blogs your app can still make it big-time. The bad news is that even if you will, you’re still up for a long fight. Most apps get a nice download peak during the three first days, then things start to slow down. At this point, the real struggle begins – how to increase your daily download rate. To reach an engaged user base and for your app to become viral you will have to increase your daily downloads, by just a bit every day. Your daily downloads graph should look something like this, the peaks are usually caused by press coverage.

Mobile app daily downloads graph

This is how a typical daily downloads graph of a successful app looks like

To get there you’ll need to scratch your way up from the bottom, step by step, using one different marketing technique after the other. Our first step was to write down all the different marketing moves we could think of (your list should include at least 20 items). For each item on the list we evaluated:

1). How long will it take until we’ll be able to measure it 2). Cost 3). Expected results. I I already warned you in the first paragraph – our marketing was an engineers’ one. Once you have this list – start following it and adding more items as you go. Within two-three months you should be able to learn which marketing moves work for your app and which don’t. The most important thing about this list is its diversity. You don’t want to put all your eggs in one basket and you never know what will bring the traffic and what won’t until you’ve tested it. Measure each item and if you feel it’s not working, remove it from the list.

1. Which blogs do matter and how to get there. When I talk with app developers I usually ask what their #1 referral traffic source is (not including social networks). I was surprised at first but kept getting the same answer again and again – usually, a mid-sized blog, definitely not one that was on their original targets shortlist. Other than being more accessible, the second advantage of small and medium sized blogs is that they drive traffic for a much longer period, unlike the large tech blogs where you’re yesterday’s news, even before yesterday, SM blogs bring traffic even a month or two after the post was published. An interesting point I’ve experienced myself as well as hearing about it from many other app developers is that blogs that cover mobile apps drive the best traffic. When I first targeted small- medium sized blogs I tried to reach design blogs, thinking that that’s where our users are at. It didn’t bring the results I was hoping for. My guess is that when users read a professional blog they’re not in the right state of mind to download an app. I heard the same feedback from developers whose app was covered by sports, family and health blogs, etc. Apparently, when users read blogs that cover mobile apps they not only tend to download apps but also to share apps they came across with their relevant friends – aka your users. Having said that, I do advise approaching the large tech blogs mainly because it’s easier to get covered by smaller blogs once you have that reference. But, yet again, AutoCAD WS was never featured on TechCrunch or on Mashable and still made it to 10M downloads. Let’s move on, how can you convince these bloggers to write about your app?

2. Making a vertical app horizontal (or in other words, making a boring app interesting) – We were convinced our product was the most exciting app – changing the way engineers and designers work together. Sadly, not many writers shared our excitement.  We were classified by the press and media as a ‘niche’ app and were having a hard time getting coverage.
The first big marketing step for a vertical app to admit that it is one. Your killer aquarium manager, classical music SongPop or wood chopping app won’t interest the average person, but to get to your vertical users you’ll have to use mainstream channels. Some of you are probably smiling to yourself right now – “I don’t have this problem, my app interests everyone!”. Those of you who run a horizontal app have a different problem to deal with- you’ll have to stand out among other apps that are similar to yours and were already covered.
We managed to get out of our ‘niche’ by working hard on creating funny, quirky and even touching content. Instead of trying to pitch our product and new features, we tried to make people laugh and feel something about it.
Here’s one example where we worked hard to make our app more interesting:  when launching our Android version we decided to use the one thing we knew Android users love best – Android. We created ‘Andy the Engineer’ as our mascot, and the video we created for the app showed an architectural version of Andy with plenty of ‘Andy’ jokes. That video got over 1M views, an amazing number considering we’re talking about a video for a CAD app. It also helped us to get covered by blogs that write about Android apps, the kind of blogs I’ve mentioned before.
Bear in mind that the way you present the product to your users doesn’t have to be identical to the way you present it to bloggers. In order to get bloggers to write about your product, it should be simple and catchy. Stop listing all the features and the future plans – it’s enough (and even better) to have one great feature when pitching to the press.

Together with AutoCAD WS Andy

3. Use great graphic materials. I’m a big fan of imagery, videos and pretty much everything you can think of other than text. The text is great if you have news – for example, your app has reached a significant milestone or has a super interesting story, but in order to get there you’ll have to stand out among thousands of other apps. We saw a huge difference in our success rate when trying to convince bloggers to cover us once we had great imagery and a video. The minimum I think you need in order to launch your app is  –

  • At least one beautiful image, this image should not be a plain screenshot but rather an image that explains what your app is all about. Try to visualize your one liner.
  • Great screenshots, craft every one of them. The content you show on your app store page should be interesting and funny (please, no more screenshots with ‘John Smith’s in the contact list or ‘buy milk’ items). Example: Disqus
  • Great image of the team. Bloggers always like to use pictures of people and show the faces behind the app. Don’t use your LinkedIn profile picture! Be original. Here are two team pictures which I used when sending our story to bloggers and they worked well.
  • It’s not a must but I would advise having one video. My rule when creating a video is that people should enjoy watching it even if they’re not your target audience, in other words – it should be fun. Don’t start with a long explanation of the problem. If there is one people are aware of it and no one likes hearing why his life is a mess. Please, no more “John is an engineer. He has to send 100 emails a day. He doesn’t have time to…”. Once you have this video, don’t forget to put it on your Google Play app page.
AutoCAD WS Launch image

4. Localize the app. Then, more important –  localize the marketing. It doesn’t matter if it’s a free or paid app – when playing in the mobile arena it’s a numbers game. Every download counts, whether it’s coming from NY or from a village in China too small to be shown on Google Maps. Localizing the app was the first step, but the 2X increase we saw in our numbers came from localizing our marketing. We started by localizing every pixel of content on our app store page.
It’s not enough to translate the app description – we wanted a Russian user to see a screenshot with a Russian username in it, a Brazilian user to see drawings in Portuguese and a Chinese user to see the app’s contact list with Chinese service email addresses. We worked on different marketing kits for each country – sending local bloggers a summary in their own language, images of the app relevant to their readers and full download numbers for their own country.
I promise to write more about marketing localization in the near future, there were lots of small tweaks which took relatively little time and increased the daily downloads. You can find more localization tips here (the quick and dirty way).

AutoCAD WS localization Italian, Russian

Some images from our post localization marketing efforts. Rotating Pisa tower in the Italian version and the Russian version promotion image.

5. How to get customer stories the guerilla way? From day one we heard great stories from our users about how they were using our app – from building theme parks to oil rigs. But every time we approached companies asking them to write their story –  Legal and ‘what’s in it for me’ got in the way. I decided that instead of contacting users I’ll try going the other way around and work with those who contacted us.
When we received email support requests from users from interesting companies, I actually picked up the phone and called them. Yes, imagine sending your feedback and having the founder of the app call you 10 minutes later. After talking to users about their request and learning how they were using our app and how it helped them be more productive, I asked their permission to write about it.
In less than two weeks we had amazing stories about designing mines in Brazil, a new children’s hospital, musical theater hall and much more. Small tip: the sooner you call the more likely you are to get good cooperation. I practiced the “one-hour” rule, calling users no later than an hour after receiving their email.

6. The app store is where users find your app. SEO it. If there’s one thing I could have changed about the App Store/ Google Play it was adding transparency to where your users are coming from. Not just from the app developer’s perspective. Websites are optimized based on the way users reach them and search. Once this transparency doesn’t exist, app developers describe their app instead of making it easy to find. It took us a few months to understand that what’s good for our website also works in the App store. That would be using as many words that people would search for – like file formats we support and accurate feature names. No, no one has searched for ‘reinventing collaboration’ or ‘sharing is now easy than ever’. Look for the keywords users use when googling your website, now use them in the app store description, as many times as possible. If you haven’t picked a name for your app yet, I would add a word which describes what you do to it – ‘GPS’, ‘CRM’, ‘Alarm’ and so on to the name itself. It can add 10-20% more downloads.

SEO the app store

7. Create direct channels to your users. Facebook and Twitter are not enough. Every successful startup has its downtimes, broken versions and awful bugs, and that’s exactly when you’ll want to communicate directly with your users. Requesting users to sign-up using their email was one of the hardest product decisions I ever made. We lost about 10% of users during sign-up. It paid off, though – we had one long downtime following a new release and another one when Amazon had an extended outage. A lot of users don’t check your blog or Twitter account at that point but instead go directly to the App Store – to rank your app with one star. Using email and in-app messages we were able to share the problem with our users directly. Unexpectedly, not only were most users supportive, we even saw a pickup in usage after notifying users that everything was back to normal. Watching your 4.5/5 star rank you’ve worked so hard for the sink in just a few days due to a tech problem is every app developer’s nightmare. Newsletters also served as a great tool for us. After releasing a new version, we sent newsletters to all active users (which entered at least 5 times this month for example). The newsletters were sent in 8 different languages based on the users’ settings. About 30% of the recipients accessed the app store/ Google Play directly from the newsletter. We had a chance to tell them about the new features with videos and images. The bonus – more downloads and much better reviews.

Marketing has changed significantly over the last couple of years. It’s no longer just about copywriting or positioning. You can be a great marketer by producing witty videos, mastering SEO, drawing funny cartoons or writing good professional tech content. All possibilities are open, start testing them and expect results.

We hope you enjoyed reading this post. You are welcome to try Oribi Analytics for Free, to improve your marketing results.