Growing from 7,000 to 28,000 Signups in 5 Weeks

Hey folks!

We launched EntryLevel on January 29th and within 7 days we ended up getting 7000+ sign ups purely organically.

Our hypothesis was that students needed work experience to break out of the no experience - no job cycle. So, we created a mentor-driven learning program that gives students real work experience and a portfolio to leave with.

In all honesty, we launched without much confidence in our early growth prospects. We set a goal of 600 signups by Friday and didn’t really know what to expect.

This is how it happened…

As you can see, a lot of our assumptions were mistaken. However, this data has given us great insight into the problem we are working on, and what our customer’s motivations are. Now to start building the product!

After our initial spurt of growth, we canned all growth activities in order to focus on operations and running the first cohort. Transparently, we had had two natural growth spurts, during and right after the program from referral volume. By the end of the program, we were sitting on 13,000 total sign-ups. With our first educational program out of the way, we could now go back to growth. Realistically, it looks like we just had a cool referral loop and everything worked but actually, we just had to think about where our users lived. Right now, people are hungry for jobs and experiences so they are looking for communities and other sources for help. After a quick brainstorm, we came up with the following ideas to test.

The first time we did growth, we only did facebook before our viral growth loop kicked in so we just stopped going through the list.

This time, we posted in a list of groups we found on FB, Linkedin and Slack. We had the following results from just post 2-3 weeks over the course of a few weeks.

We also had a hypothesis that we could make a job post and disguise our virtual experience as a job to get more eyeballs on it but in the end, we couldn’t find a good way to do it that wouldn’t make the user a little annoyed that they clicked on a supposed job ad which ended up being a program. (We advertised it under volunteering which naturally gets much fewer eyeballs than an actual job)

Distribution partners were interesting. I did this by collating a hit list of people I wanted to talk to (in Australia there are only 43 universities) but there are also many public lists after google searching on universities across Asia, Europe and the USA. As a quick pilot, I started with Australia.

Now, you can go on Linkedin and just add the careers folks and send them a message that is somewhat interesting and solves a pain point for them. I tend to flex the number of students we have and that we’re a free program. You can also use a number of tools on the internet to extract their email address from LinkedIn, in this case, I used ContactOut.

From there, I crafted targeted emails to each person with an offering asking for distribution help. In the end, I managed to talk with 3 universities and get some minor support on advertising to their student base.

Paid is an interesting beast and there’s a reason we only used this after we had initial traction. It’s because paid is not a bandaid for poor products, it is more like fuel for a rocket that is already moving. Naturally, because we had success with organic, we wanted to trial paid acquisition strategies.

We started with Google Ads, which yielded really poor results. We received a total of 23 sign-ups.

Though it says we received 110 conversions, the actual conversion point was a user hitting the success page so it’s 23. The average cost of acquisition (CAC) was $5.00 which wasn’t absolutely terrible but I knew we could do a lot better. So I paused google ads and went to explore Facebook.

Facebook Ads was interesting and I worked on it, in collaboration with Michelle Lia. For those who don’t use it much, the way my team and I like to approach Facebook ads is to start with relatively small ad set budgets ($10-20 per day) and then split any interesting demographics we get and then scaling ad spend up proportionate to the level of success that we’re seeing. 

For example, after a good amount of sign-ups, we can break down their demographic to see if they are from Gender X, Country X, Age group X or some other variable. Then we can split that into a separate ad set and see if that gathers more results.

We used our existing userbase as a lookalike, set up some ad sets and also a retargeting campaign (basically hitting people who came to our website with more ads to convert them). 

The traditional strategy is to target people then drive home conversions with a remarketing campaign. However, what we found was that people generally signed up on the first interaction with the website (our website had a relatively high conversion rate) which rendered the remarketing campaign obsolete.

We also made use of dynamic creative which allowed Facebook to do the work by spinning up multiple variations of image and headline combinations and testing personalised ads on our behalf. This not only kept our costs down but maximised the number of sign-ups we were getting.

As you can see our CAC decreased to between $0.33 and $0.46 which is more than 10x better than Google Ads. With marketing channels, I believe in tripling down on the ones that work the best rather than diversifying. 

We’re also not afraid to quickly kill off any underperforming ad sets or strategies that we may be testing out.  So we started pushing more and more money into this channel. You’ll also notice that I haven’t included the remarketing channel since the CAC was over $1, so it was not worth the spend even if we lost people in the funnel.

For the growth that occurred between 13,000 to 28,000, this was the breakdown:"

As for google organic, we don’t have any blogs but our word of mouth traffic is really high so people are googling us to find our programs from an in-person referral.

As a final note to the reader - growth shouldn’t be difficult.

If you’re truly solving a need and an issue, people will flock to your product.

If you’re trying all of these tactics and nothing is working, maybe it’s time to reflect on the product.

Until next time,

Ajay

A few tools and collections I’ve built (or obsessively curated) over the years:

These are tools and services I use personally and professionally — and recommend without hesitation: