Online Education & Training Company
This client is in the online certification space, offering specific technical courses in their niche.
In this case study, we will explore the challenges this client faced along with the solutions we implemented in order to improve performance while scaling their success.
What we did
- Analyzed data on the platform and backend to determine top performing segments.
- Optimized geo strategy to maximize regional performance
- Integrated back end data into Google Ads
- Utilized a fully fledged negative strategy to keep query quality high
- Restructured campaigns to improve consistency and increase scalability
The Proof
1 Month Progress
5 Months Progress
Solution Process
1. Account Foundation
When we first started working with the client, we performed a deep dive into the account and all the data that was available to us. The client offers several online certifications at varying costs. We originally worked on a few of their certifications before getting a chance to run the entire suite.
We knew the client had some limitations connecting data sets but were up for the challenge.
Challenges
The client was facing a myriad of challenges when we started working with them, but there were 3 particular challenges that proved to be the biggest obstacles to their success:
Performance
In the year before we began working with the client, they were seeing a consistent drop in performance metrics. Their cost per lead was increasing, their cost per enrollment was increasing, and they weren’t able to hit volume goals.
Their cost per enrollment hit its peak the month before we stepped in, so it was imperative for us to tackle this challenge as soon as possible and right the ship.
Clarity / Reporting
The second challenge the client had was getting good quality reporting from the Google Ads platform. The client was utilizing Search Ads 360 through another agency, which inflated lead numbers and made aligning goals and performance frustrating and frankly a waste of time.
Reporting must be completed with 100% accuracy in order to create an effective and appropriate strategy, so this was another challenge we needed to solve early on.
Scale
The final of the three main challenges was scale. The client noted that even though they had months of positive performance in the past, there was always a struggle to scale their success and increase lead and enrollment volume.
Tackling this challenge would mean ensuring a strong and scalable account foundation, so we kept that in mind when developing our strategy.
2. Account Structure
A good and optimized account structure is extremely important in driving and maintaining performance. We always look at account structure as the foundation to a building. A strong foundation will keep the house standing for years, but a few cracks will eventually cause a downfall.
The match types weren’t properly bid tiered, which was causing queries to come in all over the place for no apparent reason. Negatives were extremely lackluster and the geographic bidding had some overlaps that reduced the overall impact of the adjustments. There was also some automation testing that was significantly driving up the cost of Branded leads.
We decided to break out campaigns based on level of intent, match type, and device. Keywords that drove queries tied to lower funnel mindsets were separated from those that drove early research type queries. We took a look at enrollment data as well to incorporate them into our decision process for intent-based keywords.
This breakout ultimately allowed us to tier bids accordingly. Exact match and intent-based keywords would be given higher bids than Broad match, high funnel keywords. This kept CPL generally lower for low intent, and higher for high intent.
With Branded, we immediately cleaned up the keywords and removed automation. That simple switch, along with properly cross matched negatives, gave us much better control of the Branded queries and in turn, kept CPC’s to a minimum while maintaining top and absolute top of page positions.
Our negative strategy consisted of cross matching for query control and creating a master list for completely irrelevant terms coming in as well as a bad quality list of queries that drove leads that did not turn into enrollments. We also broke out master lists between Exact and Phrase match negatives in order to track down any potential negatives that cut down relevant traffic.
3. Optimizations
After redefining the foundation of the account through our structural changes, we took the account back to basics with manual CPC bidding. This lets us maintain maximum control of CPCs while we collect data on new campaigns.
We also began to implement geo-based adjustments based on the enrollment rates of certain cities and states.
We maintained manual CPCs for several months, choosing to optimize our structure and geo-based adjustments before moving forward. Our manual bidding and closely-monitored optimizations led to immediate and rapid CPL improvements; dropping the CPL by more than half in two months.
4. Data Integration
Integrating back-end data is extremely important for multi-step funnel education clients. Whether you do it directly (in-platform) or combine data after the fact through a CRM or Excel, integrating bck-end data will take your decision making to the next level.
For this particular client, we initially tracked back-end information on a campaign level using a custom-made Excel template. This is not the ideal setup, but it gave us a much-needed glimpse into information that was unavailable in the platforms. This initial setup also made it easier to track trends in performance and monitor the overall effect of the changes we were making.
Later on, we began to integrate additional funnel steps into the platform. This allowed us to make active optimizations on a keyword and even a query level.
Throughout the entire process of integrating data, we took what was available and changed our mindset regarding optimizations and changes we could track. Early on, we could track campaign trends and larger scale changes, whereas later on, we progressed to granular, keyword level information.
Eventually, we were able to integrate enrollments into the account, which was our initial goal when setting up the data and reporting systems. WIth enrollment level data, you can really weed out the lower quality keywords and campaigns, regardless of what the front end suggests. For example, if a keyword drove 100 leads with a CPL of $50 but no enrollments, that’s a lot worse than a keyword that drove 50 leads with a CPL of $100 but 5 enrollments. This really takes the account to another level in terms of looking at qualifying queries, keywords, and campaigns.
Aligning reporting and data
This might seem like a minor issue, but one of the main complaints the client had with their previous efforts was wasting a lot of time trying to match up numbers between platform and SA360 reports with their own CRM numbers. This causes a lot of headaches on both ends and reduces the capacity to innovate. Also, because the data being referred to never matched up properly, any decisions made on the management side were likely incorrect.
Our first step towards a solution was to do a full run-through of their data access capabilities. We removed the use of SA 360, as it clouded the numbers due to the previous setup. Instead, we used numbers from the platform as well as the client’s CRM to ensure we were looking at results from the same perspectives.
This ended up not only improving our optimizations in the future, but also increasing our ability to communicate with the client about shifts in performance.
5. Lead Quality
Lead quality was something we wanted to keep an eye on throughout the restructure and optimization process. In education marketing, there needs to be a solid balance between CPL and quality (CPA). We could easily have dropped CPL significantly lower, but this would run the risk of increasing the CPA beyond our goal.
Early on, we utilized a campaign-level data integration report that let us keep track of the CPA for the second step in the lead funnel. This was not the final end goal CPA, but it still enabled us to analyze how the changes we made in the platform were affecting the quality of the leads and their funnel progression. We also kept track of the percent rate of completion of the funnel step in order to attach a numerical indicator to the level of quality that each specific campaign would drive.
Later on, when we were able to integrate data into the platform a bit better, we set up custom columns in order to utilize CPAs directly into Google Ads. This improved our ability to manage quality on an increasingly granular level.
Although our lead quality did not necessarily increase month-to-month, we were consistently shifting the balance of CPL and quality in order to drive CPA down regardless. This balance is integral to navigating shifts in the market as well as seasonal trends.
6. Funnel Optimization
In order to improve the performance of our campaigns even further, we began implementing funnel optimizations, mostly in the form of website changes.
We first drove our keywords to one page in order to simplify the account and establish a testing baseline. After a month or two, we began to test out subsets of keywords based on their specific topic; driving them to pages specifically designed to increase lead conversion rates. Winning pages were built upon and implemented across larger keywords sets.
The Results
In 3 months, we were able to achieve the following results
- 119% increase in monthly leads
- 147% increase in monthly sales
- 42% decrease in CPL
- 48% decrease in CPA
The Proof
1 Month Progress
5 Months Progress
How can we help you achieve your enrollment goals?
