By Eduardo Yi
CourseMinded is a comprehensive blueprint on how to create an online course. I helped launch it last month, and managed to get 565 email addresses in 2 weeks leading up to the launch. Here’s how I did it.
For the last 8 months, I’ve been working with Tommy Griffith, SEO at Airbnb. I’ve been running operations for ClickMinded, a digital marketing course and an SEO training course for startups. and CourseMinded – an online course on how to create online courses. After successfully launching CourseMinded and creating a strategy from the ground up, I decided to document how we did it. Specifically – how we developed our first email list.
Tommy and I had been going back and forth about how to launch CourseMinded, and I was convinced that a giveaway seemed like a great idea since it would potentially allow us to rally up a big number of people interested in the product. We decided to use the KingSumo plugin for WordPress, a great solution that allows users to set up a full-fledged giveaway in a matter of minutes. Neither of us had ever run this kind of initiative, so we had no idea what to expect of it. When I asked Tommy about what our goal should be, he said:
“Getting 500 emails would be a nuts goal. Let’s shoot for that”
And so we did – 500 email addresses might not seem like a lot, but considering how niche this market is and the fact that CourseMinded is an entirely new brand and product, we thought it would be a great way to ramp up the launch.
As I saw it, there were three critical stages for a successful giveaway execution:
1. The launch.
2. The sign up.
3. The in-contest follow-up(s).
The launch: Reaching the niche
There are people who like online learning. Others who like teaching. CourseMinded is meant for the intersection between the two. This group was going to be hard – but not impossible – to reach. We stuck to what we knew and tried to take advantage of all of the obvious options out there:
– Sending the giveaway to ClickMinded’s existing email list. It could be considered kind of a stretch to think that people who are interested in learning about SEO would want to learn how to create an online course. However, the whole premise for the existence of CourseMinded is that anyone can monetize their expertise. We received some validation from ClickMinded’s students through various emails:
“I’d love to start a few courses… I’m overwhelmed.”
So Tommy wrote an email based on that insight and we sent it to our list.
– Sharing the giveaway with our own network of people via Facebook, LinkedIn, and Twitter.
– Publishing the giveaway in Subreddits, Facebook and LinkedIn groups related to edtech and online learning. We were careful to not be spammy and try to provide value to these communities. Our posts were well received because of this.
– Submitting a link to the giveaway to sweepstakes blogs and Subreddits. The messaging for these was straightforward. People in these communities are actually looking for deals so we could be way more direct.
Tommy and I talked about how this would affect the results of the giveaway. We didn’t want to attract the wrong audience – people who cared more about other prizes in the bundle than the actual featured prize: the CourseMinded course. To prevent this, it was important to think very carefully about what to include in the bundle. We would rather provide products and services needed to create and launch an online course that would add up to $1,000 in value than to give away a gift card for that amount.
– Paid search. We identified two main searcher intents around what CourseMinded offers, “create an online course” and “teach online”, so we created AdWords ad groups to match them.
PROTIP: Since giveaways have a short and finite lifespan, you can include dynamic parameters in your ads to add a sense of urgency. Just remember to always have at least one ad without these parameters in each ad group, or Google won’t approve your campaign – happened to me.
– Facebook ads. In this case, we took advantage of the advanced targeting options available and created ads for two segments: educators (professors, tutors, faculty, teachers) and people interested online learning, MOOCs, and platforms such as Coursera and EdX.
PROTIP: If you can, split test video vs static ads and stick with the version that performs better. In our case, static performed 4-5 times better than video in terms of CTR.
– LinkedIn ads. We ran a similar segmentation as in Facebook, we used people’s job titles to target educators and LinkedIn groups to target people interested in online learning. However, average CTR turned out to be really low (0.073%) so we decided against continuing with this channels after just two days, to focus efforts in others.
Finally, we wanted to test our hypothesis that people who are already monetizing their knowledge and skills were more likely to want to create an online course. We reached out to 200+ leads via email. Only 2 of these signed up to the giveaway, so we decided to discard this acquisition channel too – at least for the giveaway.
Most of these tactics were executed in the first two days of the launch, others we tried once the giveaway had started. The key here is to try to launch covering as much ground as possible, but also staying nimble: identify what’s working and improve on it; check what isn’t working and decide if it’s worth your time to try to fix it, if not, just stop doing it; and if new opportunities arise, evaluate and test them.
Once we identified channels that were generating solid leads, we tried to figure out how to create network effects within them.
Once we got people to the giveaway page, it was important to convince them to sign up. Our copy structure was really simple:
The description of the course
The description of the bundle
We realized that there was a lot of information in this page. So we added a TL;DR explanation at the top of the page with a link to an anchor at the bottom of the page to sign up. This would let users join the giveaway faster if they didn’t want to read all of the content.
Once signed up, the participants had the option to increase their chances of winning by sharing the giveaway and receiving 3 extra entries per referral.
Here was V1:
Here was V2:
We added the number of entries to the giveaway to the page and updated it regularly. We did this to give contestants visibility of where they stood in the giveaway and encourage them to get more entries.
Several contestants emailed us because they didn’t understand how to get additional entries. We realized that this was not clear enough from the default copy in the plugin, so we added our own.
The KingSumo plugin provides contestants with the option to directly share their referral links through social media. This is great since it multiplied the reach of our own efforts. However, the default share message just included the name of the giveaway and a link. Our hypothesis was that if people shared that, it would look spammy and not too engaging, so people wouldn’t click through. To fix this, I hacked my own message into the share button link, that would make it seem more authentic and interesting.
After making this change, I realized that might not be enough. Considering the audience was so specific, the chances that the participants were part of the same online communities was higher than usual. To avoid too much message repetition and people seeing right through our tactic, I created 3 different messages and randomized its display. This was as simple as including a short script in the giveaways page template code:
There was something that was missing from the KingSumo plugin. The meta descriptions to create Twitter cards from the shared links were not a part of the original code. As the Twitter card validator showed:
Yet, these were really easy to add in the footer of the page through KingSumo’s advance settings:
And voilá! A really nice Twitter card:
A nice touch was to let people know that we liked that they were sharing our giveaway on Twitter. To do that, we did a search within Twitter for the giveaway URL and ran a script to like/favorite everyone who had tweeted it. You can find that script in this awesome post on growth hacking a Twitter follower base:
Finally, Tommy had an idea to repurpose one of his most successful stories on “how ClickMinded became a six-figure side project” since it tells the story of what encouraged him to create CourseMinded in the first place. The post was updated, retitled “How To Create an Online Course From Scratch”, posted under the new CourseMinded domain and 301 redirected from its old location. We included this post in the share links for the giveaway, which we figured would work to drive quality traffic to the site.
To avoid missing an opportunity to drive people to the giveaway. We installed a persistent header in the blog post page using Smart Bar from the SumoMe suite:
PROTIP: SumoMe allows HTML in the text to the header. This meant that we could add JS script for a countdown timer to the end of the giveaway.
The in-contest follow-ups: (Growth hacking)n
Now that we had maximized the impact of our initial interaction with the participants it was time to keep the momentum going. We leveraged on email automation to achieve this: the contestants received a follow-up email the next weekday (to increases open rates) after signing up to the giveaway.
Here’s how I planned the follow-up.
I put myself in the position of a contestant. When I participate in a giveaway, I want more entries (invite more people) to have higher chances of winning. In this case, our interests and our customers’ perfectly aligned. But a contestant might not know how to get referrals. So I came up with an idea to teach people how to growth hack their own number of entries by doing exactly what we were doing = (Growth hacking)n
The automated email essentially explained contestants how to leverage Facebook, LinkedIn, and Reddit in a very simple way to get referrals. We used clickbait/Buzzfeed-format subject line and copy to try to engage our contestants with this email.
Before we launched this giveaway, I spend several weeks gathering information on how other people ran their giveaways. Something that I found some people do and I thought was particularly interesting was to “sweeten the deal” in the middle of the giveaway period to re-engage contestants. We decided to do this by:
1. Adding another big prize to the giveaway bundle, which we actually had already planned to include, but held back on its announcement.
2. Offering a guaranteed prize to contestants who achieved a number of referrals. We chose 10 referrals since it was both attainable (not discouraging to the contestant) but not so much that every contestant would be able to get it. The guaranteed prize was our own product: CourseMinded, valued at $1,000.
Of the two upgrades, the latter seems to have been the most effective one. We received several emails regarding this as soon as we announced it. This was particularly satisfying since it was a qualitative measurement of how excited our contestants were about a not-yet-launched product.
Bringing it all together: Results
It took A LOT of work to plan, launch and maintain this seemingly simple tactic, but you get what you put into it. Tommy said that 500 would be a “nuts goal” and we achieved 115% of that goal. In my short career as a digital marketer, few things have been more satisfying than hitting this number.
I put a lot of work into optimizing the mechanism to get referrals as much as possible. As a result of it, referrals drove the most contestants of all of the channels mentioned above: 47% of the total number of contestants were referred by someone else!
Our existing email list was another important channel: 28% of the contestants were part of it. However, this represented less than 3% of the people from that list. The takeaway here is that unless you have a HUGE list or one that is mainly composed of people who are interested in your new product (which is kind of a catch-22 since this is the reason for running a giveaway in the first place), the email list will likely not make or break your giveaway. You will just need to work harder to acquire the first contestants that will get your referral engine going.
The combination of all of our other direct efforts to get contestants (AdWords, Facebook ads, and groups, LinkedIn ads and groups, Subreddits, etc) generated only 25% of the total number of contestants acquired. This just confirms that we were right to have put a lot of effort into referrals since our own contestants turned out to be the most effective channel for acquisition in the giveaway.
The chart below shows the sign ups to the giveaway per day (blue bars). The yellow bars represent sign ups that we got from the ClickMinded Students email list. The red bars represent sign ups that occurred as a result of referrals.
It could be that our whole referral engine was just starting to work when we stopped or that the urgency of the deadline prompted action from people. In any case, it would’ve been interesting to know what could have happened if we had continued for another week.
Lastly, data shows that people got truly motivated to share when we announced that there was a guaranteed prize (on 11/05) and that’s when we really started getting people through referrals. I guess the takeaway here is that (counter-intuitively) our target seems to prefer the smaller but certain prize, than the larger one that depends on chance. Our audience is definitely not composed of gamblers or the lottery-ticket buyer type. If I had to do this again – which I will most likely do – I would offer the guaranteed prize right from the start.
What we could have done better: Analytics
If there is one thing that I wish I would have invested more time in, it is analytics. It was difficult to track each acquisition channel separately and we could only get so far as click-through. The better way to do this would have included conversion tracking on sign-ups and event tracking for each of the many buttons available, as well as UTM links for each channel of acquisition, campaign and target segment. This would have allowed us to be more effective when it came to optimizing our efforts and do a deeper analysis of the results of the giveaway.
IMPORTANT NOTE IF YOU’RE PLANNING TO USE THE KINGSUMO PLUGIN: We host our site on WordPress and use Google Tag Manager to install tags on this site – this includes Google Analytics. However, the page generated by the plugin does not include the tags from GTM by default. We had to manually insert the snippets of code into the giveaways page template to be able to measure traffic to that page.
The deal-maker: KingSumo
The bigger part of this whole plan revolved around a single tool: the KingSumo giveaways WordPress plugin. Out of the box, I was VERY impressed with the plugin. The list of pros to this plugin is long, but these are some of the most important things for me:
– Installation and initial setup of a giveaway takes less than 2 minutes.
– The possibilities for customization of the default features were many – I made my own changes to add non-default features and it was very easy to get access to the .php files to do that too.
– The design of the pages generated by the plugin is both stunning and functional. It also looks great and works perfectly in mobile.
– It natively integrates with several email marketing services (Aweber, MailChimp, GetResponse, and Campaign Monitor)
– And the most important piece of all: KingSumo is designed specifically for virality. Not only does it include a referral scheme, but it leverages on social channels to do so.
It shouldn’t be so surprising that this tool is so well made considering that its creators used it to build a massive email list for the wildly successful AppSumo. I will definitely use it again when I launch another giveaway for our other products.
Solving a consumer problem: CourseMinded
This launch was intended to get our first customers for CourseMinded. No matter how good of a marketing plan we had come up with for our product, we wouldn’t have been able to make it work if we weren’t addressing a real need. CourseMinded compounds years of learnings about creating and commercializing online courses, a field that has been proven to be an excellent business opportunity and to have a growing demand, but one where there is little guidance if you’re just getting started.