I already described some basic concepts of entity recognition here – today I want to show you how it looks like when you start using those concepts for your google ads accounts. The goal is to classify/tag all available search query data as good as possible.

First we have to build up an entity database. This is very time consuming and an ongoing process. Of course you already have some product data you can use easily, such as brand names – but you will realize that with this approach there are still many unclassified words in your search queries.

Speeding up the entity management with AI on user searches

Doing all the entity management manually is simply to much effort. For that reason I set up a model that is giving me semantic similar words. First we still have to do some manual work: My first entity is color. It is very easy to start with adding some color instances: red, blue, yellow, brown, …
You can make the test: how big is your list when you think about colors out there? You can also google for color lists and enrich your list with those values. Still not sufficient for me – AI to the rescue!

By using the similarity model I was able to add around 300 colors to my entity database in around 10 minutes. All color entity values are based on real world search queries of your users – when just copy and pasting RAL Code color names they will not always be relevant.

So for which colors searched our users? why is the list that big? There are a lot of crazy combinations the users are searching for. Only for “blond” I found around 70 combinations in total.

Let’s add some more entities

The same process I did with color I applied to some more entities that make sense to me:

  • Product Brand Names
  • Competitors
  • Transactions
  • Questions
  • Product Categories
  • Gender/Usergroups
  • Sizes
  • Discount

When adding the above lists a lot of new ideas came to my mind – one of them was “user pain/needs/problems”.

I started with “falten” and “pickel” and suggestions of similar words found in the user queries are this:

I ended up with 170 entity values about problems our users are searching for.

My real world example should show you this:

  • There is still a lot of work to do when it comes to defining entities that make sense for your business
  • AI can speed up the process a lot
  • Assigning new values and adding new entities has to be seen as repeating process: everyday new search queries appear in your adwords account

Great! We have a big entity lookup list!? Which of my problems are solved with that?

You are totally right if you might ask this question. The answers will be part of an additional blog post – this use cases might be interesting:

  • How we can tag search queries with our entity lookup list to get new performance insights of search patterns (with bigger sample sizes). You can use that for an advanced bidding approach with entities as features or you can block bad patterns with negative keywords.
  • We can also use the tagged queries for deriving a user driven google ads account structure – of course you can also use sources such as the google autocomplete as source.
  • Yes, we focus on PPC, but we also have some custom projects for other topics. Think of a SEO Tool that is a) Scrape the search result page for your full keyword set. b) Scrape the content of the ranking pages as well. c) Now extract keywords of our entity database and run analysis of how the content of your competitors look like