Case Studies | September 23, 2020
How We Used Business Data To Transform PPC Bidding And Win An Award
How a new approach to ecommerce bidding brought our furniture client significant growth before the pandemic and high ROI during it, whilst also bringing Liberty a PPC award and a Google case study.
This is a story about questioning everything you possibly can about the value of paid traffic and applying as much data as possible to your pay-per-click (PPC) decision making, and how in doing so, budgets can be stretched further and irrelevant traffic almost eradicated.
Thinking differently about how to approach Google Ads for one ecommerce client has lead to PPC performing twice as well as expected. It has gained us recognition from one of the top awards in our industry and also from Google, who featured our project in an advice article telling other retailers how to get ahead.
A brief introduction to the Happy Beds and Liberty relationship
Way back in early 2016 Happy Beds took on Liberty to help improve their SEO, increase organic revenues and create content that could get them positive attention in the hugely competitive beds and mattresses markets. Thankfully we proved ourselves quickly, so the next year they asked us to review their PPC account, with the hope of having all their digital marketing activity with one supplier.
It was obvious in the first review of the Google Ads account that the previous agency had run out of ideas – the account just wasn’t suitable for such a dynamic and ambitious business. Very basic Google Shopping and Search campaigns were in place but half the time they were failing to hit the Return on Ad Spend (ROAS) needed for Happy Beds to break-even, let alone make a profit.
When we dug deeper, it became apparent that they simply hadn’t taken the time to understand this client properly. As a business with hundreds of products that can be bought in thousands of combinations, with dozens of suppliers and a range of price points and delivery options, it was clear that a bespoke approach was the only way forward. We needed to consider all of these factors when determining how much to spend per click and how best to allocate their media budget.
Using business data to improve PPC performance
Thanks to an already solid relationship with Happy Beds they willingly shared a lot of sensitive business information with our PPC team from the start. And if it wasn’t for such openness and treating this relationship like a true partnership then I wouldn’t be able to write this article. It made all the difference.
Initially, this data sharing involved business costs, margins and profit. Giving your agency this data can be a scary prospect, especially for an SME like Happy Beds, and over the years I’ve heard many reasons as to why we can’t have access to this level of business knowledge. But the reality is this – if you are in a very competitive market, want to outperform your rivals and make every penny count, then this is the minimum level of company data you should be sharing with your agency.
When we knew the average costs and margins and understood them at product level, we could optimise toward a return on investment (ROI) that generated the most profit. Giving us this data allowed us to mix different bidding strategies and work out the numbers in various ways so that the budget could be used to meet the goal. As we had a strong understanding of the business and its aims, we were able to not only improve sales but also raise profit margins by controlling click spend and making sure that we were never spending more than the profit buffer for any given product.
The Liberty PPC team control what we spend every day based on what our central spreadsheet shares. This is a live document and is the result of one of our data specialists working closely with a couple of the Happy Bedders and their web developers. Using APIs to gather a number of data sources, it allows real-time monitoring of not only the entire product suite, but also the profit level and stock level. It automatically subtracts Happy Beds’ running costs, click spend and margins to reveal profit levels. With this data, we know what ROI we need to achieve at different levels of spend, which is why we’ll drop the budget on underperforming days and push it up when we know we need to push harder and have the room to do so.
We continually use this data to inform our decisions. It enables us to effectively manage our campaigns in-line with their internal targets and make sure we never stray from their goals as a business. Happy Beds remains a fast growth yet profitable business (a rare thing in the furniture world), whilst maximising the profit of their PPC accounts without needlessly increasing their ad spend. Using real-time business data and considering the real-world issues that retailers face, we created a brand-new bidding strategy. This approach to budget allocation lead to us winning ‘Best Retail Campaign’ at the Biddable Media Awards 2020. Judges said:
“We felt this entry was an excellent example of using biddable media to drive performance and business transformation. We loved the approach to budget management taking a live feed on profit”.
Tackling the Covid lockdown together
When the pandemic hit, Happy Beds saw uncertainty in a number of areas. One of the biggest challenges was how to continue operating when staff within suppliers, couriers and importers were being told to work from home. Happy Beds had to keep stock flowing through the warehouse and get it out to customers quickly.
The Directors of Happy Beds managed this by proactively organising and speaking to key partners on a daily basis. By being in control of stock levels and by purchasing items allocated to other retailers who no longer had the demand for them, Happy Beds was in a position of strength and able to advertise efficiently, continuing to sell more beds and mattresses as the demand grew.
By maintaining communication with couriers and knowing how to satisfy demand, the business was able to make adjustments to accommodate the effects of the pandemic, such as a shortage of staff in each party. Delivery times were longer than normal, so in order to “under promise and over deliver” for customers, these were updated across the site in real-time and altered automatically depending on the number of staff within each delivery company. This ensured Happy Beds did not struggle to meet customer expectations.
All of this data was added to our central hub, so it could be used to inform digital advertising decisions even further than normal so that each day our PPC team and the Happy Beds management could analyse it and make important decisions about the media spend and wider business issues.
One digital marketing challenge that Happy Beds and all retailers faced going into lockdown was the uncertainty of how competitors would react. Fortunately, there was more data to consider here and by analysing the Google Ads auction insights we were able to get a clearer picture of what others in the market were doing. Some larger retailers who have brick and mortar stores were seen to be pushing their ad spend up to balance out their store closures, while some smaller competitors were pausing ads and reducing their spend. We had to optimise the account bids and budgets in line with both increased search volumes and changes to the competitor activity.
With the help of real-time data and working intelligently with suppliers, Happy Beds had ample stock levels and access to products, so was in a very healthy position. Rather than pull back on PPC and other digital promotions, we decided to increase spend. This was done in a very calculated way, where bids and budgets were adjusted on campaigns across product sets and where opportunities arose, using the latest data and insights. This included:
- Changing budgets depending on product availability. As soon as items looked like they were going to run out, we pulled back and spent that money on other categories, allowing the best sellers to continue just from market growth.
- Considering profitability. With constantly changing stock costs, we had to make sure that profitability was considered at all times and that the items with the biggest margins had larger advertising allowances.
- Allocating budget to fill the gaps in the market. By monitoring the competitor activity and auction insights daily, to see which competitors had lowered their advertising activity or even pulled out of the auction altogether, we were able to spot opportunities for much more cost effective traffic.
- Monitoring delivery times. Retailers rely on couriers to complete the customer experience and customer satisfaction is often in their hands, so we redistributed budget when couriers couldn’t support the business with acceptable SLAs.
As lockdown progressed, Happy Beds saw a tremendous boost to conversion rates across channels and the PPC conversion rate more than doubled during the peak lockdown period. The business achieved record sales and revenues as a result of the increased search volume, higher conversion rate and smarter use of budget.
Daily sales are up over 100%, with one day peaking at 143% of what would normally be expected. This was achieved with a marginal 9% increase in cost.
“It was like Black Friday every day for over 3 months. Nobody could have predicted a sales jump this big”.Managing Director, Happy Beds.
This approach to lockdown and the success we saw was the reason we were featured in a joint case study on Think With Google this month, in their “How the Pandemic may Affect Holiday Shopping” special. The marketing team within Google approached us when they heard about this real-time approach to bidding and using automation to do some of the heavy lifting.
I’ll end by saying that Happy Beds has been a dream client for an agency like ours.
One of our core values is “Results” and another is “Innovation”, and having such an ambitious client – one who lets us test out new things, so long as we stay within overall profit parameters – is the perfect environment for digital marketers to do their best work. We are allowed to do whatever we want, test out whatever takes our fancy and experiment with different platforms, messages, audiences and bidding strategies. The outcome of this is some of the work that I am most proud of from the teams within Liberty. The results that they’ve achieved for this ecommerce brand are exceptional, but the way they went about it is truly astounding.