[CASE STUDY] From $10K to $723K Monthly Revenue for Manta Sleep

The Challenge

Before we started working with them, Manta Sleep had a strong product and brand and were a true innovator in their space, bringing to market the world’s first 100% guaranteed blackout sleep mask that doesn’t put pressure on the eyes and is totally adjustable for comfort. Before they started with Big Flare they’d done well with Facebook Ads, investing the time and effort it takes to really learn and master that skill. As they saw it, they had the Facebook Ads side of things down, and they had a good product in a category with plenty of searches per month on Google, but what they didn’t have was Google Ads expertise. They needed a true pro to come in and fully build out and manage for them a profitable Google Ads campaign to complement their already-great Facebook Ads efforts.

The Results

Fast forward a few years and we took Manta Sleep from $10k to $100k in monthly revenue on Google Ads after less than a year of working with them. We didn’t stop there though - by the end of March 2023 we took the account to over $723K in monthly revenue (and hit over $1m in December 2022)!

Beyond that, across the period we’ve worked with them we achieved a ROAS of 467%, meaning for every dollar spent we’re getting back $4.67 in revenue.

How We Did It

Performance Max

When initially scaling Manta Sleep from $10k to $100k, Google Shopping campaigns formed a large part of the activity we ran but as we moved into 2022 we wanted to ensure we were using all tools available to us. This is where Performance Max came in, Google’s powerful, semi-automated product that’s perfect for running E-Commerce campaigns at scale while supplementing our existing Shopping activity. As we gradually devised a formula over the following months for PMax, we saw them become so efficient that they eventually replaced our Smart Shopping activity entirely.

In order to set up our PMax campaigns for success, we used a tiered campaign structure. This meant having one campaign (our highest spending) for only the four or five highest performing products at any given time, meaning they had plenty of budget available to them for maximum delivery. Aside from that, we segmented campaigns by country and product type (Sleep Kits, New Products etc.) meaning not only were all assets hyper-relevant to the products they were associated with but also we were able to adjust budgets depending on the performance we were seeing from a particular market or product category. These smaller, more precise campaigns were perfect for a partner like Manta Sleep which has a lower number of SKUs but each of which drive significant sales volume.

Something which proved valuable in the early stages of setting up these PMax campaigns was making sure we provided Google with Audience Signals of what we expected the highest performing audience looking like, rather than suffering a period of low performance while Google figured the audience out for itself. This included obvious signals like users who’d searched for terms around ‘sleep’ recently, as well as targeting which we’d seen perform well for our older Shopping activity (including Dating and Fitness Affinity audiences).

Equally important as we set up these campaigns was an acute awareness of ROAS targets and when these should be adjusted over time. We settled on aiming for a ROAS of 500% but each campaign reacted to this goal differently - some were able to aim a little higher while other, higher spending campaigns needed a slightly lower target in order to attain the scale we were looking for. Ultimately, we saw that our existing philosophy of lower spending campaigns optimising towards a higher ROAS target with high spending campaigns doing the opposite served us well here.

As we exited those initial weeks of activity, our focus shifted away from audience signals and bid levels and more towards ad creative. We were working with high scale campaigns which was a gift and a curse when it came to the assets we ran: we had plenty of data coming in to show us what kind of asset performed the best for us but because they were being served so much, we had to make regular changes in order to battle creative fatigue. This is where the strength of our partnership with Manta Sleep proved invaluable: we were able to feed back on what kind of creative was engaging potential customers the best and the client were able to go and design more of it for us, meaning we had a fresh supply of high performing creative throughout the campaign. Regularly updated copy and static images drove the majority of volume for us but Manta Sleep also created strong video content for us which supplemented these other assets well. We find that to set up PMax for success, you really need to be hitting all three of those creative types and iterating them over time.

Looking at Performance Max since we launched, the results here were excellent and easily rivalled the efficiency we saw for Smart Shopping but at MUCH higher scale. In total we achieved a ROAS of 549% with this climbing as high as 860% some months. The fact we can achieve such efficiency even in months when we’re also getting over $1m in revenue is a testament to how well PMax can perform when you really take the time to nurture these campaigns.

YouTube

While, in retrospect, Performance Max was mostly to thank for the incredible ROAS results we saw, one other campaign type proved essential not only for gaining further sales volume but also for complementing the work that PMax was doing: YouTube.

When setting up YouTube for Manta Sleep we had to establish what we really wanted to get out of these campaigns: Awareness (making people familiar with the brand so when they’re served a more directive PMax ad they’re more likely to engage) and Retargeting (allowing us to clean up any potential sales from abandoned carts and repeat customers).

YouTube’s a really strong opening channel for brands with a little awareness work to do - CPMs are lower and, realistically, if you don’t run an awareness campaign of some kind then your conversion activity (in our case, PMax and Search) starts to take some of that load which reduces its effectiveness at driving sales.

We did a huge amount of testing here (over 200 audiences!!) to work out the best kind of targeting for our Awareness YouTube campaigns and eventually settled on keyword targeting: it shows a large amount of intent and we’re getting in front of users at exactly the moment they’re looking for a product like Manta Sleep’s. Although it wasn’t our objective with these campaigns, we even saw some conversions come through purely off the back of these campaigns (which we’d say is down to the strength of both the targeting as well as Manta Sleep’s videos). Creative testing, just as on Performance Max, was also key here but luckily we did see trends of videos working well for both YouTube and PMax.

Aside from those lower spending Awareness campaigns, we devoted a significant amount of budget towards Retargeting campaigns on YouTube allowing us to get back in front of users who’d visited the site, started a purchase but hadn’t completed it or even just engaged with another ad of ours. While not as efficient as Performance Max, ROAS still proved strong for us at 315% and was perfect for driving action from users who just needed one last push in order to convert.

Search

We’ve covered how important it is for us to be running full-funnel activity and YouTube meant that the top of the funnel was now covered. But even with Retargeting campaigns we had doubts that we were cleaning up all possible conversions at the bottom of the funnel: this is where Search came in.

With all the E-Commerce focused campaigns that Google now offer, it’s easy to forget about the bread and butter of the platform and we still find these more manual campaigns crucial for supplementing what automated campaigns like PMax and Smart Shopping offer.

Our thinking behind Search was that we’d likely achieve less volume out of it (only a certain number of people can be searching for sleep products each day while PMax can serve all over Google’s inventory) BUT given how high intent is, it should provide us an even stronger ROAS.

Overall, we were able to achieve an excellent ROAS on Search of 647%! 

We’d attribute the success here to truly embracing the manual nature of Search as we followed our internal best practices for these campaigns: constantly testing new RSA copy, using smaller ad groups by theme and, crucially, staying on top of keyword management:

Search Terms Management

How often in your Google Ads account are you reviewing your Search Query Reports (SQRs) and using that information to improve your keyword targeting?

If you said “not very often” or, worse, “huh, what’s an SQR?” then you could be sitting on a pile of missed opportunities.

What we did with Manta Sleep’s account was implement a weekly process of going into the SQRs at least once per week, and scouring through the report like a detective, looking for new positive and negative keyword opportunities.

Your SQRs are found in your Keywords section, under Search Terms. To give you an idea of how powerful this process can be, imagine you go into your SQRs one day and you see that your broad match keyword for “sleep masks” has also been triggering ads on people who searched “weighted sleep mask”.

There are multiple ways you can use this insight. First, and most obviously, you can take this search term, turn it into a keyword, and set up an ad group for this keyword in your account. Then, you can write ads specifically for “weighted sleep masks” that speak directly to people looking for this kind of product.

You can take this even further if you want and if the term gets great search volume, you may even want to set up a custom landing page with “Weighted Sleep Masks” in the heading and copy that focuses on the features and benefits that these customers care about most.

This kind of research can also give you a bunch of ideas for new product lines e.g. if you didn’t already have a weighted sleep mask, this might be a strong vote to actually go ahead and create that product in the first place. If creating that product is not on the agenda for now, then seeing this term in your list may indicate that this is an opportunity to add “weighted” as a negative keyword so that in future, you don’t accidentally show ads to people looking for something you do not sell.

Attribution

At BigFlare it’s crucial for us to be as close to the data as possible in order to make sure we’re not just working with the most accurate numbers possible but also so we can react quickly to changes and use as much information as possible in order to guide our optimisations. Manta Sleep share these values with us and so, after some discussion with the team there, we decided we’d reached a point where it was time for us to onboard a third party attribution tool in order to make sure we were working with the most accurate data possible.

We’d had doubts around Google actually underreporting results when looking at in-platform numbers and, when looking for a more holistic option for us to see how we were REALLY performing, we settled on NorthBeam. We’d reached a point with the account where we needed to use more advanced attribution models than what Google could provide so once this had been onboarded, we had a lot more confidence in our data as well as having more levers we could pull when it came to getting super granular with our optimisations. Their offering was perfect for E-Commerce activity and access to their dashboards meant that both ourselves and the team at Manta Sleep both had exactly the same picture of performance.

Wrapping Up

One thing that I really want to highlight is that what made our work with Manta Sleep so effective is that both parties really treat this like a partnership: we have regular calls to communicate what is and isn’t working, we get a vast amount of assets to experiment with and we get to work with a great website with a truly innovative product offering. Experimenting with new campaign types and hours of optimisation will only go so far if your website and product aren’t in good shape. 

My team and I handled the Google Ads and after implementing our processes and skills there, we saw great growth, but this simply could not have been done without an awesome client who worked hard on their stuff as well and put in the time and effort to communicate with us effectively. It really was a team effort between our two companies to get this stuff running so well, so here’s a big shout out to Mark, James, and the rest of the gang at Manta Sleep who helped us make this all happen, you guys are awesome!