How to Advertise a SaaS Product: Case Study in the Crypto Niche using YouTube Ads

This case study details the process and results that I used to create successful YouTube ads for a SaaS app that sits halfway between B2B and B2C in the Cryptocurrency trading space. I've had to censor some of the commercially sensitive data, but it should still provide useful insights if you're considering YouTube ads as a way to generate leads and signups for your SaaS.

The company

A bootstrapped SaaS Tool in the Cryptocurrency trading niche, with a team of 30.

Situation before

Google search ads worked sporadiclly due to heavy handed algo bans, organic SEO was good, but they wanted a way to be able to grow quicker and capture more of the market in an expanding niche, before well funded competitors got there first.

What we worked on

YouTube ads for a particular geographic region of focus

What would success look like?

Generating sign ups for a similar or lower price to what they are able to get with Google Search Ads, but with a cold audience

Process

Script Writing

I wrote three scripts, with versions of each excluding the terms "crypto" and similar, in case the algorithm picked up on those and rejected the ads (even though the company was actually fully compliant with Google Ads policy on cryptocurrency, the algo often has false positives).

It's important to have at least three variations so that you can start to test different ideas.

On a larger budget and with more time, I would also test out more variables such as presenter (male or female, accent, etc), filming background and lighting (indoors, outdoors, plain background vs office setup, shaky iPhone video vs pro filming), music, and many others.

Hiring a Presenter

I've found that having an on camera presenter typically gets better results than only having a voice over on top of motion graphics or stock footage.

There's a number of presenters I've worked with in the past who have their own home studio. In this case I needed to find a 'millenial' looking presenter who had an Australian accent.

There's several freelance marketplaces one can use to find preseneter talent, in this case I used Fiverr.

Editing & Motion Graphics

In any kind of filmmaking, the principle "show, don't tell" rules supreme.

In this case, I decided to show specific flows from the app in a simplified, very clear way that communicates the value of the tool quickly.

The process of adding motion graphics and working to a final edit took a couple of days, but the results were really worth it.

Ads Campaign Setup

We wanted to test different audiences, and with input from the company's Head of Growth, I added in some audiences that he wanted to test.

When starting out with YouTube ads, my preference is to go with Placement targeting first of all. That means manually selecting a list of videos and channels that your ideal customers are highly likely to be watching.

My criteria when starting out is:

  • Channels: are 80% or more of the videos on the channel likely to be viewed by our ideal customers?
  • Videos: same criteria, but try to select whole channels if possible. Check to see if the video appears to still be getting a reasonable number of views (otherwise there won't be enough people to show ads to).

I spent several hours researching the cryptocurrency landscape on YouTube, and built a placement list which I felt was well targeted.

I also selected a 3 day site visitor audience for retargeting, and another one based off YouTube keywords which are very specific to the problem that this product solves within the crypto space (again, I wish I could be more specific here but I don't want to reveal anything commercially sensitive).

Many of these channels are ones where the company has tried to get in video sponserships but the fees asked for by the influencers were deemed to be too high by the company. This way, we get the benefits of these influencers specific audiences, but without negotiations, wait time, and high sponsership fees.

Results

In the inital 7 day campaign (based around an event happening in the Crypto space), we were able to generate 118 direct sign ups, plus an additional 39 'view through conversions', which is people who didn't click 'sign up' in the YouTube ad but instead visited the website directly or through Google later and then signed up.

The cost per conversion was 30% lower than the benchmark for what the business was happy to pay to aquire a signup (based on their back end LTV calcs), for the campaigns that 'won' in testing. Including the testing budget, the cost per signup was in line with the benchmark.

The cost per sign up was 37% lower compared to Google Search Ads over the same time period.

Conclusion

After the initial 7 day campaign, the results are very encouraging. Typically YouTube ads take more experimentation and time to get good ROI, but in this case they worked well right out of the gate.

The key here is that the product already has good product market fit, and has an offer which is suitable for paid traffic. I write more about this in my guide to YouTube ads for B2B.

If you'd like some help figuring out if YouTube ads make sense for your business, reach out using the contact details in the footer.