100,000 Users in a Day: An App Marketing Case Study

INTRODUCTION

This is a real life app marketing case study that follows the launch of our popular music app, Jam for iPhone (Now Jam:Instant Song Creator).

Hopefully this case study will provide readers with some useful app marketing tips and strategies. More importantly though, I hope it demonstrates just how important a marketing strategy is and how much thought and effort needs to go into one.

I’ve included the following details in the case study:

• Cost breakdowns
• Marketing plans
• Press releases
• Real results

QUICK SUMMARY

Jam has been downloaded by almost a million users since its launch in 2013. This case study will focus on the first four months. During this time, Jam was downloaded by approximately 290,000 users.

As far as benchmarks go, this number certainly puts it in the ‘successful launch’ category but isn’t so big as to push it into ‘unicorn’ territory. This makes it a very useful case study, as the strategies we used are universal and the results are attainable for many app startups.

We worked with an app marketing agency, who were integral in the planning and execution of the Jam marketing strategy.

COSTS

We had a marketing budget of approximately US$170,000. But, before you dismiss this as not relevant to startups bootstrapping it, consider these three things:

• The majority of those 290,000 users were acquired with just $8,000 of this budget.
• Over 100,000 of these users were acquired overnight
• There is no ‘right’ amount to spend on marketing. Apps with lower budgets have been more successful than Jam and apps with much bigger budgets have failed completely.

$170,000 is simply the budget we had to launch Jam and our strategy was developed around this number.

USER ACQUISITION COSTS

The effective user acquisition cost we achieved was approximately 58 cents per user. The average paid media CPI (cost-per-install) in the US in 2013 was $2.96. If you do the math, our app marketing campaign effectively earned Jam around $688,400 in free installs.

Of course user acquisition cost is not the only metric we use to measure the success of a marketing campaign. Here were our goals of the launch campaign:

• 1 million installs within 12 months
• An engaged and active user base
• Self-sustaining growth (enough revenue and virality to continue to grow the user base on an ongoing basis without spending more money on marketing)

KEY FOCUS AREAS

The Jam launch campaign focused on 4 key areas:

• Press Coverage

• Events

• Paid Media

• Social Media

Social media marketing advice is easy to come by, so we won’t focus on that in this case study. It was also the least successful aspect of our campaign and I don’t believe it contributed much to the app’s success, largely due to us starting it too late. Where social media did play a role, I will be sure to mention it.

I’m using a timeline style format for this case study, so you can see when things happened as they happened.

NOVEMBER – Planning Begins

Jam app development took twelve months with a team of three internal mobile app developers and it was ready for beta testing in November of 2012. This is when our Jam marketing efforts officially kicked off. 

We had our hearts set on a Christmas day release.

Most professional marketers will tell you marketing should begin as soon as you decide to build the app.

In my experience though, most app startups operate in stealth mode and the thought of disclosing their precious idea to anyone before the app is finished makes them very nervous. This was certainly the case for us with Jam.

So, less than two months before the public launch date, we reached out to a handful of reputable app marketing agencies and selected one.

DECEMBER – A Strategy Forms

December was spent strategizing, preparing collateral and content, making bookings and getting ready for launch.

We identified early on that Jam was one of those apps that really needed to be seen to be believed.

Jam uses artificial intelligence to create original music based on any vocal input. The user just picks a music style, hits record and sings. Then, within seconds, Jam generates original backing music and allows you to share your masterpiece via SMS, email, social media or to the internal community and charts. 

Explaining all this takes a long time but demonstrating Jam in person only takes a few seconds and the results are quite impressive. So, we knew we needed people to see Jam in action and that idea became central to our marketing strategy.

We also knew that virality was crucial. We couldn’t afford to pay $2.96 to buy users in CPI campaigns. Therefore, we needed a strategy that encouraged organic growth of our user base.

A few weeks was spent with our agency on the app marketing strategy and this is a very brief summary of what we initially settled on, with costings.

WEEK 1 – Burst Campaigns Begin

We missed our Christmas release.

There were a few reasons. Last-minute bugs meant we missed Apple’s Christmas submissions cutoff. We also missed the exhibitor registration cutoff for CES.

So we decided to soft launch on January 11. This gave us a couple of weeks to iron out any potential issues and gain a bit of exposure before a hard launch at Macworld on Jan 31.

On the 14th we sent out press releases (read it here) and the marketing agency started reaching out to their media contacts. It didn’t take long to see the results and by the end of the week, we had dozens of articles and reviews around the internet.

Meanwhile, a few of our smaller ad campaigns kicked off and data started coming in. We initially ran ads with the following networks, spending approximately $8,000 in the first week of launch:

• Facebook Ads – $1,800
• AppsGratis – A site to discover apps that were free for just one day. – $3,000
• Millennial Media – Mobile ad banners – $1,700
• Meta Resolver – Mobile ad banners – $1,500

We were surprised to reach over 200,000 downloads in the first week and the number one spot on the App Store music app charts in twenty-four countries.

AppsGratis was a big contributing factor, driving 50,000 downloads in just one day (very cost effective at just $3,000!). The viral impact of the burst campaign led to around 148,000 organic installs and the paid ad banner campaigns contributed a couple of thousand more.

We also observed the following metrics:

• No. 1 music app in 26 countries on App Store
• No. 1 app overall in 2 countries
• Over 15,000 songs published (shared inside the Jam social network)
• Over 500,000 in-app song plays
• $2,600 in revenue

Facebook ads didn’t perform well at all, costing us over $4 per user, so we quickly pulled the plug on those and moved the rest of our Facebook budget to the other app marketing channels.

Running Tally

WEEK 2 – More Ad Tests

Following the success of the AppsGratis campaign, which achieved an unprecedented 6 cents per install, we were keen to run another free-for-a-day campaign.

We managed to secure a spot in the popular App-o-day, but the cost was much higher than AppsGratis. We were also keen to get our hands on as much mobile ad banner inventory as we could for a big second week burst campaign. So, when the AppsFire network guaranteed us at least 100,000 clicks within a week, we added them into the mix.

Here’s what we spent in week 2:

• App-o-day – $29,000
• AppsFire – $20,000
• Millennial Media – $1,700
• Meta Resolver – $1,500

Though we spent $52,200 on ads in week two (almost seven times more than week one) the results fell well short. App-o-day performed the best, bringing in just under 20,000 new users but the ad networks performed terribly achieving just a few thousand additional installs at a cost of $3-$5 per user.

In spite of the poor banner ad conversions, Jam still had some media buzz and good app store reviews. This, plus users sharing tens of thousands of Jam creations on social media, meant the app managed to stay at the top of the music app charts for another week. But, by the time MacWorld arrived, Jam had dropped to #201.

Running Tally

WEEK 3 – Macworld

Hard launch.

With the money we saved not attending CES, we wanted to pull some kind of stunt to get media exposure and help make the most of our Macworld investment. That stunt came in the form of a live chapulin monkey named Abu.

Monkeys are a big part of the app design, so bringing a live monkey to Macworld was an on-brand gimmick.

Macworld organisers said no, but we opted to bring him along anyway and ask for forgiveness later if required. 

The cost to hire the monkey was $3,500. We printed a little Jam T-Shirt for him and attendees went bananas!

Over the course of the conference, we serenaded attendees with live Jam demos, ran an iPad giveaway and a took part in the Macworld Media Preview press conference. But, it was Abu the monkey who attracted the most attention, with people queuing up for a photo with Abu on their shoulder. Watch Video.

The Guardian, Star-Telegraph, ABC7, TUAW, MacObserver and others all raved about Jam and our monkey friend. Even Macworld themselves named Jam and Abu one of the highlights of the conference. Needless to say, an apology to Macworld for breaking their ‘no monkey’ rule was never required.

Despite the hype, the attention we got at Macworld didn’t translate into downloads. In all, our user base only grew by about 8,000 and most could be credited to the banner ads we were still running. Macworld also ended up costing more than we anticipated.

Running Tally

WEEK 4 – What next?

Deflated by lacklustre Macworld results and more than half way through our budget, we went back to the drawing board. We’d seen how quickly we could eat through our money and didn’t want to waste what we had left.

At this point we switched off our ad campaigns and just let our social media marketing chug along. In doing so we were able to ascertain our organic baseline.

Without spending anything, Jam organically achieved around 2,000 new installs this week.

The marketing agency floated the idea of attending SXSW in Austin, Texas. A massive music, film and tech festival sounded perfect for our rock star app, so we jumped on it. We wanted to make a big splash and resolved to spend the rest of our app marketing budget on it. A hail Mary of sorts.

Here’s what we had learned so far:

• Free-for-a-day promotions worked really well but you could only run them every so often or get diminishing returns
• Mobile banner campaigns were expensive and getting enough inventory to run big bursts was difficult.
• Events were great for creating a buzz but the hype didn’t immediately convert into new users.
• Advertising in big bursts did help to push Jam to the top of the charts
• Once in the charts, the virality kicked in but it was relatively short lived

One thing we hadn’t tried yet was a celebrity endorsement. SXSW would be our chance.

Running Tally

WEEK 5 – SXSW Planning

With the festival less than a month away, it became our sole focus. We spent the next few weeks coming up with a killer strategy.

We would sponsor the lounge at the SXSW Music Tech Expo, turning it into the Jam lounge. Here we could run live Jam demos to attendees and offer them free drink cards to use at the Jam bar if they installed the app in front of us.

And finally, the highlight of our plan – we’d have a famous rock star make an appearance in the lounge to generate hype.

We threw around a bunch of names and the marketing agency started calling agents. We figured it wouldn’t be hard to find someone, with hundreds of bands and artists already in Austin to perform at the festival.

Our list of potentials included Dave Grohl, Alt J, Third Eye Blind, Prince, Vampire Weekend, The Rubens, Major Lazer and our top choice, Macklemore and Ryan Lewis. That Summer, the hip hop duo’s song ‘Thrift Shop’ was number one on the Billboard charts and they appealed to our young target demographic.

We were able to get in touch with a few agents. Some artists were available, some weren’t. Some weren’t interested. Some were quoting way more than we could afford.

Then we got the call we were hoping for. Macklemore and Ryan Lewis were available. The cost was $22,000 for a one hour appearance and a tweet!

We locked it in immediately.

Running Tally

WEEK 9 – SXSW

March 13 came along and Sam and I flew out to Austin, where we were met by the marketing agency team and we hit the ground running. 

They had sent out a new press release (read it here) and teed up some opportunities for us to wine and dine media reps at festival events.

The day before the expo we bought a few dozen wacky garments and a coat rack at a thrift shop (op shop in Aussie). The marketing agency had the great idea of setting up a photo booth in the Jam lounge and allowing guests to dress up in thrift shop attire for a photo with Macklemore.

The app marketing plan for the next three days included:

• Live Jam demos to lounge guests
• The Jam bar serving drinks in Jam branded cups with quick-install QR codes
• Free drink cards for anyone who downloaded the app
• Dinners and shows with media reps (paid for by us)
• An iSnap digital photo booth with costume rack
• Screens in the lounge running the Jam promo on loop
• Flyer handouts promoting the Macklemore appearance
• Social media posts about the Macklemore appearance
• The big Macklemore appearance 
• Mobile ad banner campaigns were switched back on

The total cost, including staffing, airfares and accommodation came to $53,600.

How it all went down:

We sang hundreds of Jam demos. Sometimes we blew people away and sometimes we failed hysterically but overall the response was very positive.

Attendees downloaded the Jam app for their free drink cards and our media outreach started bearing fruit. On day one of the expo (14th), reviews and articles started popping up around the internet and you could feel the buzz starting to grow.

On day two, Macklemore turned up and things got a bit crazy.

That morning, we arrived early at the expo venue to find a cue of mostly teenage girls stretching all the way around the block. Macklemore wasn’t set to arrive until 1:30PM but fans had camped out for prime queue positions. We took advantage of the situation and got everyone in the queue to download Jam while they waited.

Macklemore and Ryan Lewis turned up an hour late with an entourage of about twenty.

They were very professional and enthusiastic with their adoring fans, who had waited patiently for hours to see them.

Fans started moving through the photo booth, every one of them leaving with a smile and posting their Jam-branded photos to Facebook and Instagram. Here’s a video of them in the photo booth (Sorry for the low resolution).

The extra downloads we gained on the day were great but the real impact came in the form of media exposure. 

I’d been interviewed by a few journalists already but on day two, the crowds kicked this into second gear. An Australian journalist from the AFP asked me what Jam was and why it was so popular. I’m not sure he even knew we had a major celebrity signing autographs five feet away. The app got the credit for the crowds and though we didn’t know it yet, his story would soon be syndicated to hundreds of publications around the world.

The final day was mostly spent with journalists who heard about the excitement the previous day and wanted the lowdown.

Among them was a young woman doing a video blog for Fuse TV. She sang a song into my phone and we grooved along to the result. I didn’t know it at the time, but she was mega-famous fashion model Alexa Chung. View the video here.

That day 50 Cent also showed up to sign autographs at the headphone booth next to our lounge. This generated a lot more buzz and we sang more demos.

In total, we gained around 40,000 new users during our time at SXSW, which equated to about $1.34 per user. We attributed about 5,000 of these to event attendees, 5,000 to mobile banner ads and the remainder to media publicity and organic growth.

Running Tally

WEEK 10 – Back to Australia

Our app marketing budget dried up and our engagement with the marketing agency drew to an end.

Alexa’s Fuse TV segment, the AFP syndication and hundreds of other articles and press mentions, turned Jam into an Australian export success story.

We’d only been home for a day when I got a call from the Today Show. We were on air the next morning telling our story and listening to Karl Stephanovic Jamming to a TV audience of almost half a million viewers. Watch Video.

Following our Today Show appearance we were interviewed by a handful of other local news outlets, then things slowed back down.

Our mobile ad campaigns ended and I took over social media management. It was time to assess the impact of the launch campaign. Had we met the goals we set out to achieve?

• 1 million installs within 12 months? – Not yet achieved and not likely within the next 9 months at the current growth rate
• An engaged and active user base? – Yes, our users were very active and engaged
• Self-sustaining growth? – No. The revenue we were bringing in was nowhere near enough to sustain the growth of the product. Nor were our organic downloads.

But, this isn’t a post mortem. The app lived on.

Running Tally

WEEK 11-16

We had just over 280,000 users in total and almost 10,000 were active on any given day. We spent the next few weeks tweaking the features of the app and monetisation model in an effort to improve their experience and generate more revenue.

Organic downloads continued at around 1,500 per week. 

Despite our optimisations, revenue didn’t get much better and we continued to pull in around $300 per week.

After being on the App Store for 4 months our app marketing launch campaign had achieved the following results:

Running Tally

We also achieved:

• 315,000 songs created
• 34,200 songs published
• 1.3 million song plays
• 2,925 in-app-purchases
• No. 1 Music App in 26 countries
• No. 1 Free App Overall in 2 countries
• Top grossing music app in 1 countries
• Featured by Apple in ‘New and Noteworthy’
• 2nd place in the Appy Awards (Runner up to Shazam)

What we learned:

• People love to create and share music. All kinds of people.
• App marketing plans need to cover a much longer period than just a four month launch window.
• It is not about how much money you spend but where you spend it. We had more success with $8,000 of our budget that we did with the remaining $162,000.
• Make sure you have the right app before spending money promoting it. Scale up a profitable app, not one that is costing you money.
• It’s better to get feedback from test users, make iterations and ensure you are build an app people will pay to use before you launch publically.
• Run smaller ad tests and carefully measure the data to ensure you are getting a return from a particular channel before increasing the spend.
• Get social right. We didn’t focus heavily enough on social and it was a wasted opportunity. Social influencers should have played a much bigger role in our strategy.
• Momentum is key, especially early on. Plan marketing activities to bounce off each other to maximise exposure and keep people talking about your app.
• Not all downloads are created equal. Focus your efforts and marketing spend on users that will be engaged in the platform and generate virality and revenue.
• Make the most out of every opportunity. One thing we did really well was media interactions. We said yes to every opportunity and went out of our way to create newsworthy stories.

Where is Jam now?

Jam is still alive and kicking. The user base isn’t as active as it once was and we don’t spend nearly enough time maintaining the platform but it still has a following and people still love making music with it.

It has been installed by just under a million users now and while we never gained much financially from the whole exercise, what we gained in experience and knowledge is priceless.