Using AI to Find SEO Keywords

Search Engine Results Page, otherwise known as SERP, refers to the page displayed by search engines in response to a user’s query. Scout apps can utilize SERP blocks to help look up various pieces of information.

An AI powered SERP app can serve various purposes including:

• Competitor Analysis: To analyze the SERP rankings of competitors, track their keyword performance, and understand their strategies.
• Keyword Research: To identify high-performing keywords, track changes in keyword rankings, and discover new keyword opportunities.
• Content Optimization: To understand which content types perform best on SERPs and guide content creation efforts to improve rankings.
• Personalized Search Results: To develop systems that provide users with more personalized and relevant search results based on their preferences and behavior.

Overall, a SERP AI bot can provide valuable insights and automation capabilities that enhance various aspects of digital marketing, SEO, market research, and competitive analysis, and more. So today, we’ll work on setting up a SERP powered Scout App.

Once you have signed into Scout and created an organization you will land on the dashboard. From here we can start building out our keyword research app. Click on the “+” icon located in the side menu to create a new app. After naming your App we can proceed to the App configuration.

To build out your keyword research tool we’ll be utilizing SERP blocks to look up various pieces of information. To add blocks, click on the “+” icon in the center of the screen, from the dropdown of available blocks select “web.”

Once you have added a web block we can now configure it. Our first block will be used lookup trending topics and keywords. For this we don’t want to search a particular site and instead want just a web search. Your search term for this block will look as follows:

{{inputs.input}} Trending topics

Your first block should look something like this. Make sure to change the slug, and title of the block for easy identification. The last step of configuration will be to adjust the “Web Page Limit” and “Chunk Limit.”

Next, we’ll add a few more blocks to grab some more data on our trending topics. Go ahead and add another web block. This time we will search trends.google.com to find out the volume and days trending for a given keyword. This blocks config. should look something like this:

And for the last web block we’ll do a general search to gather missing volume and days trending information. This block should look like:

With all of our web blocks in place its time to work on our prompt and tell the app what we’re looking for in a response. To configure the LLM block lets first select our language model, we recommend using GPT-4o. Next is “Temperature,” on this app we want a more conservative response so let’s set the value here to “0” and lastly lets set the “Max Tokens” to 500. (This dictates your response length, for longer answers or if you notice an answer cut-off try increasing this value.)

The last step is to write your prompt. For this app I have written a simple prompt that pulls in the results from each SERP block to form our response. This is a rather basic prompt and can be expanded upon depending on your use case. For those looking for a plug-and-play prompt, feel free to copy the prompt below:

{% set system_message %}
You are an expert at SEO and Digital Marketing and tasked with helping the user with keyword research. 

Using data from {{kw.output}} and {{trends.output}} list the keywords in order of which ones provide the best ranking potential and are trending with significant volume. Also, attach a volume value for each keyword if available, lastly, note how many days a particular keyword has been trending using information from {{days_trending.output}}.
{% endset %}

{% if not chat_history %}
$SYSTEM: {{ system_message }}
{% endif %}
{% for message in chat_history %}
{% if message.role != "system" %}
{% if loop.last %}
$SYSTEM: {{ system_message }}

{% endif %}
${{message.role | upper}}: {{message.content}}
{% endif %}
{% endfor %}

$USER: {{ inputs.input }}

Our keyword research app is now ready to test! You can open a chat window to interact with your new app by clicking on the text bubble icon at the top right of the page.

Start building with Scout and transform the way you handle research. If you have any questions or need further assistance drop into our community Slack where our users and staff can lend a hand.