How we DOUBLED Revenue for Our Client with Just 1/10 of Visitors

DISCLOSURE: This blog post may contain affiliate links and any purchase(s) made through such links will result in a small commission to HUSTLR (at no cost for you).

  • This article has been fact-checked for accuracy.

Table of Contents


A little over half a year ago, our Head of Growth, Jeremy, published a piece on his experience working with FMCG companies, and how they were literally lighting their ad budget on fire based on his experience working within such companies.

Well, that narrative came in full circle when HUSTLR recently began working with an FMCG client on its performance marketing campaigns after discovering that ad spend on the channel was not generating positive ROI. We had a look at their previous ad spend, and sure enough, they were literally burning away their ad budget on vanity metrics.

Based on our analysis, we’ve discovered large discrepancies between the number of visitors and sales revenue. It was also revealed to us by the client that their ad spending was not methodical, and they needed help to drive more relevant traffic to their site with paid advertising.

Our client decided that they needed to attribute sales revenue as their goal instead of the number of visitors. After all, the number of visitors is merely a vanity metric for any eCommerce company that ultimately needs to move products.

The core focus on our end in terms of performance marketing has been to drive more relevant visitors to their site, to yield a net positive ROI on ad spend.

To achieve our objectives, we worked closely with our client to revamp their entire marketing mix and media buy strategy.


Execution


In order to fine-tune the ad media for optimal digital performance, a minimal amount from the total budget was used for A/B Testing, where different types of ads with the same imagery were tested to identify the more lucrative option.

With that said, we only had a very limited time frame to run our proposed ad campaign, spanning just 2 weeks, which gave us little to no time to waste on testing. This also meant that we had no time to optimize our Facebook Pixel based on Events that occurred on the site.

To measure the campaign’s performance, tracking codes from both Google Analytics and Facebook Pixel were implemented on the target website. This enables data analysis to be performed on every step of the customer journey.


Results


Despite having no time to optimize our Facebook Pixel, our sophisticated demographics targeting still produced outstanding results for our client. Our campaign brought them more than double the revenue of the month of November in just 15 days, when our client had spent exorbitant ad budget to bring more than 10 times the number of visitors to their site than our campaign which had a relatively modest budget. 


On top of that, our results were produced with an extremely tight time constraint of just 15 days. This means that there is plenty of room to grow had we had enough time to launch retargeting ads, as well as allowing for our Facebook Pixel to be optimized based on Events that took place on the site.

Needless to say, our client was absolutely ecstatic for the results, especially considering we’ve been able to generate more sales in 2 weeks than one of the biggest months they had last year.

The key takeaway that we’d like to share based on this case study is not to be fixated on vanity metrics, as seen in the chart above, with the huge spike in the number of visitors in November that resulted in mediocre sales.

More importantly, this case study also highlights the importance of rapid A/B testing and finding a winning ad before doubling down on it. This is the sole reason that we were able to generate tangible results in just a span of 15 days, as we were able to identify the winning ad before allocating most of our budget to it.

While this may be hard to foresee the effectiveness of the ad campaign while it is running, especially with the huge surge in visitors, we found that conversion rates and bounce rates often give a clearer picture of analyzing the quality of traffic that is being driven. 


For our ad campaign, we can see that the Conversion Rate is 10x of the November, which subsequently resulted in the end goal of capturing more orders and revenue.

Bounce rates were also noticeably reduced, which means that more of the visitors on the site is spending time browsing products and potentially making a purchase.

For those of you that are running paid advertising campaigns, we hope that this case study shed some light on the importance of properly targeting your audience to bring relevant traffic to your site.

Want similar results for your business? HUSTLR is now providing services for performance marketing and more! Do check out our Services page to learn more about how we can help you grow your business. 

Leave a Comment

Your email address will not be published. Required fields are marked *

Posts Like This

The Ultimate Guide to Amazon PPC
eCommerce

The Ultimate Guide to Amazon PPC in 2020

With close to 120 million products listed on Amazon, it can appear very daunting to get your product to stand out from the endless sea of products, especially for new Amazon sellers. However, if you’ve done your product research and

Read More »
Share
Pin
Tweet
Share
Reddit