How to Identify Negative Keywords for Google Shopping Query Sculpting in a Scalable Way


While specific queries drive higher conversion rates, obtaining negative keywords for Google-looking question sculpting will be challenging. These approaches will assist you to save time and budget whereas scaling.

It is widespread in advanced Google looking campaign structures to separate the traffic into totally different campaign buckets of:

•Generic: HIGH Campaign priority

•Branded: MEDIUM Campaign priority

•Product Specific: LOW Campaign priority

to attain this, we’ve to outline these negative lists:

• Brand Identifiers + Product Identifiers ↔ Generic

• Product Identifiers ↔ Branded

Usually, you’ll observe higher conversion rates on a lot of specific queries. If the questions can be split up successfully, then you’ll be able to begin moving the budget from generic to particular by adjusting your bidding strategies.

The main challenges and solutions of Google Shopping queries

In looking ads, search queries match your things by Google; therefore, you can’t trigger the impression with keywords. However, victimization negative keywords can assist you in confirming that your ads are relevant. In Google looking question sculpting tasks, the biggest challenge is to induce negatives for your product brands and products. Let’s use the merchandise feed for starters.

In theory, the setup sounds nice, but it is time-intense to get the correct set of negatives, although it can be worn out a climbable means with the subsequent approaches.

•Use N-Grams rather than entire queries for your negatives: By doing this, you may block the utmost variety of specific questions sharing that pattern. This additionally avoids achieving the Google Ads limits for the number of negative keywords

•Build a look engine supported your product feed information which will offer you this result:

• If there is much merchandise from totally different product wholes matching → GENERIC

• If there’s one brand matching with numerous products → BRANDED

• If there is a little set of products matching at intervals an equivalent brand → PRODUCT SPECIFIC

•Use the general queries in your Google Shopping account, produce 1-Grams and 2-Grams out of it and classify them with our program approach. This approach already split your traffic quality a great deal. Once you investigate the GENERIC campaign, you may still notice a lot of queries that are at the incorrect place for these reasons:

• Brand names and products identifiers typically seem with misspellings and typos. Along it’s an enormous quantity of traffic.

• There are synonyms or abbreviations for a lot of brands and product identifiers.

By adding each case to our search index, to quality of the classification will any improve.

Performance-based approach for setting negative keywords list

Let’s mix the previous cases with a second performance-based mostly approach for setting negatives. The categories of queries that Google sends to the looking accounts might not match at all. It’s troublesome to spot them (or n-grams that belong to them) with a bit of quantity of samples. The excellent issue is that they seem within the GENERIC bucket wherever we tend to pay less already.

Our second approach makes it even a lot of controllable.

• If you have enough samples on an n-gram with a HIGH or HIGH price per click, we can block it in general.

• If we’ve got a HIGH VPC (value per click), we tend to block it in BRANDED.

Important: During this case, we’re solely victimization performance-based, mostly negatives for N-Grams that may not be classified with the primary approach. Otherwise, you’d block an entire complete in your branded bucket if the VPC performance is excellent.

For the VERY HIGH and HIGH definition, we’re employing a distribution-based approach supported by the entire account.

More control means better performance.

You will notice right away that the resulting buckets are performing differently. After successfully splitting the traffic quality, you can now use different bidding strategies or target values for each campaign.