Introducing Two Important Updates to Bluedot Intelligence: Integration Management and Prompt Recommendations

We’re excited to introduce an important feature update to Bluedot Intelligence, our GEO analytics and execution automation platform: Integration Management and Prompt Recommendations.

One of the questions we hear most frequently when meeting with customers is: “How should we create prompts?”

We also often hear questions like, “Is there any way to know what our customers are asking ChatGPT or Gemini?”

Unless OpenAI, Google, or Anthropic decides to disclose this data, there is currently no way to access it directly. Moreover, as people become increasingly comfortable asking questions in natural language, it is virtually impossible to determine how many times the exact same question has been asked.

At best, we can make educated guesses by examining the sets and scale of queries generated through query fan-out.

This is why designing effective prompts is far from easy.

There is, however, one valuable source of data that can provide clues about what users are actually asking: query data from Google Search Console (GSC).

Google Search Console provides the queries users entered before arriving at a page, along with impression and click data. Recently, we have seen a rapid increase in queries consisting of four or more words, rather than the traditional one- or two-word searches.

We believe this is a result of the growing number of users who have become accustomed to entering natural-language questions in AI search.

These queries can also be understood as outputs of query fan-out.

When users ask long, natural-language questions, AI search systems break those questions down into multiple queries through a process known as query fan-out. These resulting queries are often relatively long themselves.

Compared with the past, we are now seeing a significant number of four- and five-word queries appearing in Google Search Console.

Whatever the interpretation, one thing is clear: this is real data generated directly by users.

We’ve included an example below. As you can see, queries containing four or more words now account for a significant share of the data.

Reverse-Engineering Customer Questions from GSC Query Data

At Bluedot AI, we have been eagerly waiting for the day when these types of queries would become mainstream.

Until just three or four months ago, one- and two-word search terms consistently dominated the top impression rankings. Over the past one or two months, however, we have frequently seen long-tail queries of four or more words rise to the top.

We interpret this as a clear signal that AI-driven search behavior is increasing.

Naturally, we began looking for ways to make use of these queries.

The result is our new Prompt Recommendations feature.

This feature extracts high-impression queries containing four or more words from Google Search Console and reverse-engineers them into customer-intent-based prompts, or questions.

Based on intent analysis, each prompt is then mapped to our five-stage customer journey model.

Is the user asking about the problem itself? Are they exploring potential solutions? Are they comparing or validating options? Or are they preparing to purchase or take action?

Bluedot Intelligence classifies each query accordingly.

Through this analysis, Bluedot Intelligence recommends up to 20 prompts, as shown below.

You no longer have to struggle with the question, “What prompts should we enter to collect meaningful data?”

With Bluedot Intelligence’s Prompt Recommendations alone, you can identify up to 20 real-data-based customer questions.

Connect Google Search Console with Bluedot Intelligence through Integration Management

To use Prompt Recommendations, you first need to connect the Google Search Console property you manage to Bluedot Intelligence.

This connection allows Bluedot Intelligence to retrieve and analyze your query data.

To support this process, we have added a new Integration Management section to the left-hand menu.

Currently, integrations are available for Google Search Console and two AI channel services. We plan to rapidly expand the number of supported services and platforms going forward.

The integration process is extremely simple.

All you need is a Google Search Console account. With just a few clicks, the connection between the two services can be completed.

Once the integration is finished, you can immediately start using Prompt Recommendations.

With the introduction of this feature, we have removed the previous menu that used AI to generate prompts based solely on its training data.

The reason is simple: designing prompts based on real data from Google Search Console is a far more scientific and data-driven approach.

GA4 Integration Is Coming Soon to Bluedot Intelligence

We are also preparing to integrate GA4 with Bluedot Intelligence.

The goal is to further strengthen the reliability of Bluedot Intelligence’s GEO data.

As in the example above, connecting Bluedot Intelligence’s GEO metrics with key GA4 metrics will allow us to provide even deeper insights.

We have already tracked and compared the relationship between these two datasets multiple times with our customers. Through this work, we have confirmed that meaningful correlations exist.

We are currently designing a correlation chart between citation rates and AI-driven sessions.

Once completed, we believe this will significantly help brands establish reliable GEO KPIs.

In other words, Bluedot Intelligence will provide the data needed to answer one of the most common questions about GEO:

“What actually changes when we invest in GEO?”

The GSC-data-based Prompt Recommendations feature is currently available to both Pro and Starter subscribers.

For now, we have opened the feature to both plans so that users can experience it firsthand. However, differences between the two plans—including the number of recommended prompts and prompt quality—will be introduced soon.

Until then, feel free to explore and use the feature as much as you like.

Design Prompts with GSC Data and Discover Your GEO Weaknesses

Use Google Search Console data to understand the questions your users are asking.

Then, use those questions to evaluate how vulnerable your brand is in GEO.

Once you identify those weaknesses, ask Bluedot Intelligence how you can address and overcome them.

That is how your brand can move to the forefront of GEO.

Bluedot Intelligence, the GEO analytics and execution automation platform, is here to help you succeed in GEO.