Paid Case Study: Amazon Advertising – Office Supplies

• Sponsored product campaign restructure
• Resolve keyword overlap
• Create coherent plan across several brands
• Create new storefronts
• Reduce ACoS to 10% with keyword research, negatives, bidding adjustments

The foremost issue with the accounts was the poor structure. Many products had their own campaigns and were not sharing keywords with similar products. Besides managing 600+ campaigns, this keyword overlap was causing issues because poorly performing keywords, when excluded, would simply start generating impressions in a different campaign for an almost identical product. Having the same keyword in a dozen or more campaigns dilutes the data, making it more difficult to make optimization decisions while also causing ad spend to occur on low ROI products. We needed to simplify the campaign structure to make it more manageable, but the very first thing we wanted to do was cut wasted ad spend.

We reduced wasted spend in all high value campaigns to help our pressing ACoS issue by lowering bids on some keywords and by pausing others. Amazon makes keyword bid recommendations, but it is important to have your own research and experience to help you set the correct bid. Amazon’s keyword bid recommendation is designed to help advertisers get more impressions and earn Amazon more money. Be wary of bid suggestions that are significantly larger than others in that same ad group. This client automatically set most of their bids to the Amazon suggestion, which in many cases was not the optimal bid level. Advertisers need to consider that if there is a recommendation of $10.50 in an ad group that spends about $1.50 per click, you need to not only change the bid, but maybe re-think that keyword entirely. It’s often the case that these stand out bids are tied to broad match keywords, and that possible keyword matches are not closely enough tied to your product. If you have “table saw” as a broad or phrase match keyword in a campaign, but you only sell the table saw blades, you can’t allow yourself to be fooled by the bid recommendation. Your product won’t match searcher intent in most cases, and table saw advertisers can afford a much higher bid due to the higher margin. Keyword modifications like these helped us reduce ad spend and increase our CTR without losing any revenue.

Campaign restructure

The goal of our new campaign structure was to group all similar items into campaigns, with ad groups to make slight distinctions between products of keyword groups. Before WTM Digital, most items had their own campaigns. A brand new structure not only helped reduce overall account size, but also gave us the opportunity to resolve keyword overlap. The new structure does not have the same keyword in multiple campaigns, which helps us evaluate the true metrics of each keyword and set better bids. Having the same keyword in multiple campaigns breaks down your data into less useful segments. Multiple keyword locations also hurt the account because your bid for a particular keyword might be higher in a campaign with products that have a low conversion rate, while your good products receive a smaller portion of that keyword’s traffic.

We always recommend breaking campaigns apart by product type, and segmenting by similar keyword groupings or product differentiations. For example, this could mean a campaign for all saw blades, with ad groups that have similar keywords, but make a distinction in product type, like wood cutting or metal blades.

As we built our new campaigns, we slowly began transitioning spend from the old campaigns to their new counterparts. It takes about 14 days for a new campaign to start earning impressions for the majority of searches for which it is eligible, and 30 days to be running at full speed. To account for this, we started by halving budgets in the old campaigns and moving the rest of the spend over to the new versions. Over time, we adjusted the budgets more until the transition was complete for each campaign, about 1 month.

Lastly, we reviewed all the search term data from the old campaigns and added in new negative keywords. We also created a “feeder” campaign with a high budget. The feeder campaign has automatic targeting, with ad groups for each product type. This allows us to have low bids but still generate impressions and search terms. Over time, the search terms can be reviewed both for new negative keywords, but also for new high value keywords for the manually targeted campaigns.

The campaign restructure focused on Sponsored Products ads, as they do make up the majority of ad sales, but our team also had to design an overall company strategy for all of the client’s brands. This exercise included running Headline, Sponsored Products, and Product Display campaigns, as well as building some of the brand-new replacement to Brand Pages, Storefronts.

The preference was to keep as much advertising as possible within Seller Central. Even though Vendor Central/Amazon Marketing Services has the ability to run all 3 campaign types, the interface and reporting capabilities of Seller Central are vastly superior to AMS, thus making Seller the obvious place to run the majority of the ads. This particular client had 2 brands, and owned them both within the new Brand Registry 2.0. We absolutely recommend you own the rights to all your brand in the new registry as it will benefit your company and allow you to create Storefronts and campaigns with Storefront landing pages.

A complete Amazon strategy helps vendors increase their brand visibility on helps them target ads to people at different stages of buying intent and each campaign type plays an important role. Headline ads act as branding and category awareness, sponsored products target people specifically interested in a certain product, and product display ads help you keep potential customers within your own Amazon shopping environment. Storefronts also provide a good landing page for some ads, and serve as your brand hub on Amazon. Our client didn’t have a storefront, so we created one for each brand, being sure to use multiple widgets and group like products onto several different tabs within each storefront.

Before WTM Digital, ACoS had been creeping up from 10% until it hit 16% the month we began.

• Immediate focus on improving keyword quality reduced the ACoS to target within 1 month.
• ACoS is within target in 1 month, down 30% from previous month
• New keyword research and negative keywords helped increase CTR week over week
• Better bids decreased average cost per click while revenue and sales grew.
• Revenue up 56% and spend is down 3%
• 165% increase in Return on Ad Spend Dollars.
• 9:1 Return on Ad Spend.