Streamlining Google Ads for B2B E-Commerce Performance: A Case Study in Simplification and Consolidation

If you are reading this, you will already understand that the world of PPC rarely stands still. Matt Beswick, a 15-year Google Ads and B2B E-commerce professional shares his journey and insights on completely revamping a reasonably well-performing account through consolidation and simplicity.

Background

I recently transitioned from the world of agency PPC to an in-house role and have tried to use the experience gained from all that previous experience to improve the performance of an overly complex account. Upon arrival, the account was flooded with over 100 standard shopping campaigns and 20-30 search campaigns, all sharing sub-200 conversions.

A Platform For Success Through Robust Tracking

This is where anyone should start when taking any account, so the first step is ensuring that conversions are tracked accurately.

  • Set up server-side tagging with Taggrs.IO – This enabled us to ensure the data we received was reliable, and fewer conversions were slipping through the gaps.
  • Enabled Enhanced Conversions – Passing back hashed first party data to Google Ads to increase the accuracy of conversion tracking.
  • Offline Conversion Tracking (OCT) & CRM Integration—B2B e-commerce drives many offline conversions. OCT and CRM give us a holistic view of what our customers are doing and what drives valuable actions.
  • Enabled Consent Mode V2 (you can read my article about this here).
  • Correctly Configured Remarketing Datasets. We ensured remarketing lists were fit for purpose, ensuring dynamic shopping remarketing can work by triggering the ID (or dynx_id) parameters where consent was given.
 

Embracing Simplicity 

Lots of LinkedIn gurus will call this the Hagakure method. I call it simplicity. If I can’t see it, I can’t measure it, and if I can’t measure it, I can’t improve it.

Overly complex accounts lead to the PPC specialist only focusing on what’s great or bad. This results in the loss of average performers and opportunities for growth.

Initially, this account had fewer than 200 conversions across over 100 campaigns. All campaigns were on Target ROAS bidding, giving the bidding algorithms almost no chance of success.

I consolidated these initially into just ten campaigns.

  • 4 Pmax
  • 2 Dynamic Search (DSA)
  • 2 Brand
  • 1 Remarketing
  • 1 Standard Shopping
 

Account Restructure 

Step 1 – Protect Brand

If you are going to have a Performance Max-led structure such as this, it’s essential to protect branded search.

  • Create a branded campaign for each country we sell to.
  • Two ad groups per campaign – one targeting pure brand, the other for anything PMax likes to consume (reviews, contact us, telephone, etc). This gives control over how much you want to spend on those search terms and where you want to send them.
  • Account level negatives for business info type searches like Ltd, Limited, Company, Directors, etc

Step 2 – Pmax Led Shopping

Initially, we created just two campaigns per country, one each for our top-level categories, to allow Google enough conversions in each campaign to work out precisely who was likely to convert.

The steps we took to ensure success with these campaigns were as follows:

  • Create all asset PMax campaigns with URL expansion on – we wanted to give Google complete control at the early stages to gather as many conversions as possible.
  • Asset groups for each manufacturer/brand supplied – this ensured assets were more relevant than generic all-encompassing ads, and measurement was more straightforward.
  • tROAS bidding – My go to tactic is to use tROAS with a relatively low budget initially. I prefer to have PMax campaigns limited by budget to ensure spend is spent only in the areas of maximum performance.
  • As performance dictates, move products/brands/product types that hog budget/conversions into their own campaign to allow others to shine.

Step 3 – DSA with URL Feed

As I prefer PMax to be limited by budget, we created a catch-all DSA campaign for each targeted country to capture data and conversions across the search network.

  • Create a URL feed for products only. This was to ensure dynamic search ads were not directing people to non-transactional pages such as about us, contact us etc.
  • Created ad groups for each manufacturer/brand supplied to ensure messaging was better than a catch-all ad.
  • Add all derivatives of branded search to negatives to ensure the campaign only targets incremental traffic.

Step 4 – Remarketing

  • Created brand-led remarketing campaigns targeting people at each stage of the funnel: product browse, add to basket, view cart, and abandoned checkout.
  • Dynamic remarketing shows customers who have triggered the ID (or dynx_id) parameter.

Step 5 – Standard Shopping

This may be the more subjective or controversial aspect of our decision. Many people suggest that in e-commerce, you should run PMax campaigns at all hours and on all days. I beg to differ.

I don’t believe you are starving the algorithm of data if you know you convert less or acquire customers more expensively during certain hours or on weekends, particularly in B2B.

  • Use standard shopping with bids set at less than £0.10
  • Actively add negatives
  • Adjust bids accordingly

Step 6 – Feed Optimisation

Now that we had the foundations of a well-thought-out structure and strategy, our efforts turned to optimising the feed.

This is an area that is, in my opinion, criminally undervalued by many in the PPC community.

  • Use a feed management tool to manage the feed effectively. In this case, we used Shoptimised.
  • Rebuild the feed from the ground up. Out of the box feeds from E-commerce platforms can be sketchy at best. We rebuilt this completely to ensure the data provided was as accurate and up-to-date as possible.
  • Improve product type depth – Google can use this to help match searches to products, and it’s the easiest way to help structure campaigns for your segment.
  • Used all relevant attributes 

Step 7 – Constantly Evolving

As data came in, we constantly evolved the structure and strategy to ensure campaigns kept improving performance.

Examples of this evolution are:

  • Conversion Rate Optimisation – Performance improvements are often driven outside of platforms. Starting on the most valuable pages where revenue is lost, we conducted multiple A/B tests. This started at the checkout, as revenue is leaving your site here and at the homepage, as returning users are your most valuable and often land here.
  • An automated process for moving zero-click, zero-impression products into a new campaign in an attempt to find new hero products.
  • Move high-ticket and hard-to-match items to standard shopping to give complete control over which products were being shown for which items. 
  • Test different bidding strategies to see their impact on online and offline performance.
  • Move brands/manufacturers/product types that were hogging budget and conversions to new campaigns to allow other products room to spend.
  • Create new search campaigns as trends emerge from DSA. 
  • Consolidate search campaigns across countries where conversion volume was low.
  • Use scripts provided by Mike Rhodes & Jack Felstead to monitor precisely what is going on within PMax campaigns.
 

The Results

When comparing Q1 of 2024 with Q1 of 2023

  • 76% increase in cost
  • 115% increase in conversions
  • 126% increase in conversion value
  • 11% decrease in CPA
  • Percentage of revenue from the brand fell from 48% to 19%
 

Conclusion

As always, this structure won’t work for everyone. Hopefully, you can find some insights from how this project was approached to rethink your accounts or provide inspiration to help gain incremental growth from your B2B accounts.

Leave a Reply

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

Our Sponsors

EVENT GALLERY

GALLERY FROM PREVIOUS EVENTS