50% More Leads with the Same Ad Spend in Google Ads through Conversion Consolidation and Smart Bidding

Time: 5 months

Team: Project Manager, PPC Specialist, Analyst

Tech Stack: Google Ads, GTM, GA4, Looker studio

Key Results of the Project:

1. Achieved a 50% reduction in CPA
2. Developed a strategy resulting in a consistent CPA below the KPI
3. Established an Evergreen setup that is easy to maintain without requiring profound knowledge of the Google Ads system

Background 

The Client

Catering company is a U.S.-based business operating in various regions throughout California. The client's business comprises 12 branches, each with its dedicated website catering to a specific region.

The company's mission is to provide high-quality service, excellent cuisine, and a streamlined process for ordering catering for events and parties. Their services cover a spectrum, from straightforward drop-offs for events of various sizes to extravagant weddings and corporate gatherings accommodating up to 2000 guests. The company offers comprehensive packages that include service staff, bartenders, rentals, and catered food.

The Objective

Achieving the right targets for the campaigns was a vital component of this project. Our goals ranged from an initial objective of increasing the number of leads by at least 15% to ultimately meeting the specific acquisition cost per lead by the end of our collaboration. 

We began with a straightforward approach and adapted our strategies as we worked collaboratively with the client, reaching a point that effectively met their needs.

Our Starting Point

Our collaboration started in mid-June 2023 when the client's typical Google Ads engagement model proved ineffective in generating the desired volume of leads at a reasonable cost. This was attributed to a combination of global factors and specific challenges within the state of the paid campaigns. The PPC campaigns were initially set up by the client several years ago with the primary goal of generating leads for later processing by the sales team.

When we examined the data prior to kick-off our changes, we came up with the following hypotheses regarding the reasons why the performance had deteriorated:

• Outdated Campaigns Bidding Setup: All of the client’s campaigns used a manual bidding strategy with bids set on the keyword level, rarely reviewed, and no utilization of smart bidding.Best Practice: To focus on conversions, it’s crucial to employ Smart Bidding, which automates much of the heavy lifting and guesswork in setting and updating bids. Smart Bidding considers a wide range of auction-time signals, aiding in achieving the desired CPA.
• Using Separate Universal Analytics Conversions Actions for Each Location as Primary Goals: This posed a challenge, as campaigns had too few conversions per month to successfully employ automated bidding. Due to this, the client hesitated to switch to automated bidding because of the poor performance observed in tests with automated bidding in the past.Best Practice: Since Universal Analytics has been deprecated, it's essential to either upload GA4 conversions to the Google Ads account or utilize native Google Ads conversion tracking. This ensures that all conversions are accurately recorded and attributed.

• Avoiding proper keyword segmentation: There was no separation of high-competition and consequently more expensive keywords (such as wedding or corporate catering) into their own campaigns with distinct target CPA.Best Practice: We recommend splitting keywords into campaigns and ad groups based on the theme and tailoring the ads in the ad groups to cover that specific theme. It's advised to have about 10-20 keywords in a single ad group.

Step Zero - Setting up proper conversion tracking (June 2023)

When NNPARTNERS was approached, the client had been managing Google campaigns independently for several years. Despite accumulating valuable data on ad performance in their Universal Analytics and Google Ads account, the existing analytics setup needed to be replaced with a newer version of Google Analytics (GA4). Additionally, there was no way to consolidate data and assess the collective performance of all websites, as the data was scattered among 12 individual Universal Analytics accounts. 

We began our work by creating a 3-step plan to implement an effective tracking system for the client’s business:

1. Created a new set of 12 GA4 properties, each dedicated to a specific business branch. And most importantly, set up a cross-domain property to consolidate all necessary data in one place for the client’s convenience.
2. Enhanced the GA4 setup by adding more conversion actions to track micro conversions that were absent in the old analytics. This addition aimed to provide a comprehensive view of user engagement and account success
3. According to Google Ads best practices, using GA4 imported conversions is not an ideal solution due to the conversion attribution time lag and inability to configure Enhanced conversion tracking. Hence, the third step of the plan was to create a native cross-domain Google Ads conversion actions. These actions consolidate all primary conversions of the same type into one conversion action that can be applied to all campaigns in the account. This approach ensures that all campaigns benefit from shared conversion data, promoting synergy among them. And, of course, in the new conversion actions, the Data-driven attribution model was selected.

After completing all these tasks, we went an extra mile by offering our client a user-friendly Looker Studio report containing all essential metrics. This was particularly helpful since navigating through the GA4 interface might pose challenges, especially for less experienced user

Step One - Hypothesis testing (July 2023)

After strategically setting up a unified Google Ads conversion action, we embarked on a 10-day waiting period to unlock the full potential of our data-driven approach. This deliberate pause aimed to outsmart algorithms, ensuring they learn from a more insightful canvas. By sidestepping the conventional Maximize conversion bidding strategy for the new campaigns, we envisioned achieving a remarkable reduction in the cost per conversion (CPA), fueled by our tailored and precise bidding goal.

And after collecting the necessary conversion volume, we promptly elevated our new conversion action to be the account's primary goal, ensuring that all campaigns, both new and old, are utilizing it instead of imported Universal Analytics conversions.

We were initially allocated a small portion of the campaigns' budgets to test our hypotheses and refine campaign setup and bidding strategies. 

Existing search campaigns, relying solely on manual bids, featured a multitude of user intent keywords, including broad match or broad modifier match types that constituted a significant portion of the campaign spend. Additionally, we identified and removed keywords deemed "too expensive" given the client's business specifics and budget constraints.

Our approach involved:

• creating new campaigns with smart bidding settings, 
• implementing refined keyword segmentation, leveraging Google Keyword Planner for new keyword identification, 
• enhancing ad groups with extensions such as callouts, sitelinks, prices, and structured snippets. 
• Furthermore, we introduced new Responsive Search Ads to each ad group.

By the end of July, our efforts resulted in a nearly twofold reduction in CPA compared to May (pre-intervention). Conversion rates nearly quadrupled, and the number of conversions increased by 50%.

Step Two - Scaling up (August 2023)

After successfully achieving a CPA that was almost twice lower than the previous benchmark, we expanded our strategy to encompass other branches of the company situated in diverse geographical locations. The overarching goal was to standardize the CPA across the entire account.

During this phase, our primary emphasis was on consolidating conversion data across all campaigns. This strategic move aimed to guarantee that even the smallest campaigns received sufficient conversion signals, thereby sustaining the achieved CPA.

To address this requirement, we conceptualized the creation of several portfolio bidding strategies. These strategies were designed to incorporate various geo-specific campaigns, aligning with the user's intent and ensuring a cohesive approach to maintaining the desired CPA.

Several combinations of portfolio bidding strategies were tested during this period, and despite the experimental nature of these tests, we successfully achieved an average CPA that was only 1.9% higher than the set KPI.

Step Three - Developing Self-sustained LeadGen Focused Structure Across the Account  (September 2023)

As a result of the previous testing process, we have developed an account structure that can yield consistent and predictable results with little to no need for constant adjustments, which the client can manage independently. 

Over the course of two months after ending the collaboration with the client, this setup has proven to be effective, bringing consistent lead volume with a stable cost per conversion.

Overall result  

In a span of only 5 months, we have successfully established a steady flow of high-quality leads within a self-sustaining account structure, achieving a consistent cost per acquisition within client’s KPIs. Moreover, we've witnessed an impressive growth of over 50% in the number of leads, showcasing the effectiveness and impact of our strategies.