Author Archives: Aaron Zakowski

About Aaron Zakowski

Aaron is the CEO and founder of Zammo Digital. He has managed over $2 million in Facebook ads and generated more than 500,000 new user signups for his clients including Invision, DigitalOcean, Ebay, Treehouse, Webydo and more. Follow him on LinkedIn, Instagram and Facebook.

DigitalOcean: A Lesson in Scaling Facebook Ads for B2B SaaS

DigitalOcean Facebook Ads Case Study

Do Facebook ads work for B2B SaaS companies?

Heck ya they do!!!

And we’ve got the results to prove it.

Just ask our very happy clients at DigitalOcean (and InVision, and many others clients).

We started managing Facebook ads for DigitalOcean a few months ago and the results have been… well… keep reading…

TLDR; Zammo Digital succeeded in lowering DigitalOcean’s cost to acquire a new user by 15% and increased the volume of new customer signups from Facebook ads by 330%. We achieved these results through a systematic approach to continuously testing, optimizing and scaling our ads.

DigitalOcean’s Goals

DigitalOcean came to us a few months ago with the goal of making Facebook ads a bigger, more scalable part of their growth strategy.

Until then, DigitalOcean had focused growth on building their online community of software developers. Facebook ads simply hadn’t been a big priority and they had never achieved significant enough results to justify tackling it in-house.

Facebook ads wasn’t one of their core competencies that they wanted to invest in in-house.

That’s why they decided to hire Zammo Digital to take charge of turning Facebook ads into a viable, scalable customer acquisition channel.

The Results

Within the first 30 days of the engagement, we had applied testing, optimization and scaling processes proved that Facebook could in fact be a significant growth driver for large B2B SaaS companies like DigitalOcean.

The results speak for themselves… And the team at DigitalOcean (from the CEO down to the digital marketing team) were thrilled with the ROI.

In the first 30 days of the project, we succeeded in lowering DigitalOcean’s cost to acquire a new user (CPA) by 15% and increased the volume of new customer signups from Facebook ads by 330% compared to the previous month.

“Aaron and his team at Zammo Digital feel like an extension of our marketing team. They push us to look at our overall marketing strategies differently and their systematic, creative process to running Facebook ads makes us feel like our campaigns are in good hands. Zammo has succeeded big time in making Facebook a viable growth channel for DigitalOcean.” ~David Dorman, Director of Growth and Demand Generation at DigitalOcean

In fact, the decrease in CPA was so substantial that the savings covered our agencies entire fee. And now their marketing team was freed up to focus on other tasks.

By the second month, we scaled the volume of leads even more.

like-a-boss

Want to learn how we did it?

Read on…

Systems and Processes

One of the most important ways to achieve success with Facebook ads, especially if you want to scale up and increase budget of your campaigns, is to make sure you have repeatable processes in place.

At Zammo Digital, we have processes around testing new ad creatives, new audiences, optimization setting and everything else.

We’re conservative and protective of our clients’ ad budgets and we want to get profitable results as quickly as possible.

cat-money

That requires being as systematic and scientific as possible.

We run lots of small tests at the beginning of a campaign on every relevant variable that we can think of.

Then we kill the losing tests quickly and scale the winners quickly.

When we test audiences, we make sure all of our ad sets have the same ad creatives.

When we test images, we make sure all tests use the same text and audience.

When we test ad copy, we hold images and audiences constant.

When we find ads that work on a small audience (i.e. 1% Lookalike) we scale them up to larger audiences (i.e. 2% Lookalike).

There’s a process and a system to achieving quick results and to avoid significant losses while testing.

Testing New Ad Creatives Styles

Digital Ocean has a strong brand with a unique style of imagery.

VirtualMachines-Ad-1200x628

Historically, their Facebook ads had always conformed with their brand style. Unfortunately, they hadn’t had the results from Facebook ads that they wanted.

From the time we got started with them, we pushed DigitalOcean’s marketing team to let us experiment with new image styles.

In general, our process for success includes testing dozens of new ad creatives every month. This way, we make sure to have new winners available as soon as our previous winning ads start to burn out.

Facebook ads are labor intensive.

Definitely not “set it and forget it.”

Our experience with other B2B SaaS companies had proved that images with human faces resulted in better conversions. Our testing with DigitalOcean proved the same was true for them.

DigitalOcean Facebook Ads

DigitalOcean has a fun mascot named Sammy the Shark.

We knew Sammy would work great on ads.

So we tested our hypothesis and proved that the shark was a big winner.

DigitalOcean Facebook Ad SammyShark

We’ve now expanded with a system to test more ads based on our previous findings.

And DigitalOcean has been able to take the findings from all our tests and successfully apply them to other digital marketing channels such as Adwords and Twitter.

Scaling Internationally

DigitalOcean is quickly becoming one of the largest and most popular cloud computing platforms in the world.

And they wanted their ads being promoted internationally as well.

The problem is that running a campaign for each country is generally too much work to manage.

And batching all countries together usually results in less valuable countries getting most of the budget and impressions.

We implemented a strategy to create different campaign each targeting a different bucket or tier of countries. The Tiers were selected based on the value that the countries in it provided to the company.

We had a unique Campaign for the United States. Then a Tier1 campaign for more valuable countries (i.e. Canada, United Kingdom, Germany, etc), Tier2 for less valuable countries and a Tier3 to target third world countries.

Within each tier, we have campaigns for Lookalikes, retargeting, interest targeting, etc.

With this strategy, we can also assign different conversion values to different geographic regions when we review the campaign results.

Scaling to Mobile Ads

Like many large SaaS, IaaS and PaaS companies, DigitalOcean’s platform and sales funnel is designed to work best on Desktop rather than mobile devices.

So historically, they had only run ads on desktop devices.

Unfortunately, that prevented benefiting from Facebook’s huge mobile ad inventory.

We decided to test some mobile traffic to see what would happen…

To our pleasant surprise, we were able to acquire tons of new conversions when advertising on mobile placements.

And mobile ads were even driving lower acquisition costs than desktop ads.

But the really interesting thing we found in the “Conversion Device” report in Ads Manager, was that most of the conversions from mobile ad placements were actually taking place on desktop devices.

That means that people were clicking on mobile ads and then converting when they got to their desktop computers. ( FYI – most of these conversions were based on 28-day click attribution. Not view-throughs.)

Once we discovered this opportunity, it didn’t take us long to start scaling up mobile ads to broader audiences and bigger budgets.

Scaling to Larger Lookalikes

One of the biggest successes we’ve had with DigitalOcean and other clients recently, is with scaling to larger and larger Lookalike audiences.

Facebook’s algorithm continues to get smarter and smarter and that results in more opportunities for advertisers to scale ads.

For example, once we prove that a specific ad can run profitably with a 1% Lookalike of our conversion pixel, we will test broadening that ad creative to a larger 2% Lookalike.

From there, we’ll try to get up to a 5% or 10% Lookalike.

Although a 5% Lookalike is approximately 10 million people in the United States (which might be a larger potential audience than most companies actually have), in reality Facebook won’t display your ad to all those people. The algorithm will still attempt to only display ads to the users who are most like to convert.

In some cases, we’ve even been able to profitably go from a 10% Lookalike to totally open targeting (i.e. No targeting at all. Just opening it up to all people in the United States).

(Shout out to Jimmy Kim, CEO of SendLane for teaching me the Open Targeting strategy.)

One more point to keep in mind is that scaling up to larger audiences will be more achievable once you have thousands of conversions on your pixel.

Takeaways…

A lot of larger SaaS companies struggle to make Facebook ads a reliable and scalable revenue channel.

But just because Facebook usually won’t drive the same amount of revenue as account-based, outbound sales doesn’t mean that it can’t be a significant growth channel.

As we explained above, the key to SaaS success on Facebook is having strong systems in place, running lots of ongoing tests and understanding and leveraging how Facebook’s algorithms work.

Interesting in learning about how Zammo Digital can help with your Facebook Ads? Get in touch via our contact form and let’s discuss what your options are.