Topical Research

Topical Research is where you turn a core topic, your brand context and your personas into a structured set of topics and keywords. The result is a hierarchy up to four levels deep that feeds SERP Clustering and, from there, your Topical Map.

Use this page as a guide for setting up your research, generating topics, editing your research, finding AI search gaps, and exporting your work.

What Topical Research Does

Topical Research helps you:

Expand a core topic into Main Topics and subtopics, up to four levels deep
Ground topic generation in real search behavior using live Google and YouTube autocomplete
Generate keywords for the topics they belong to
Check whether your research needs a fourth level before you add one
Shape and refine your research before clustering starts
Find gaps by comparing your research against what AI search engines recommend for your buyer personas

Your topical research becomes the starting point for:

Where to Find Topical Research

Open your project in Floyi.
In the left navigation, select Topical Research.

You also reach it from Site Architecture with Next: Topical Research, or with Continue to Topical Research if you skipped site architecture.

Before You Start: Brand and Personas

You will get better results if you set up Brand Foundation and at least one persona in Audience Insights first.

Floyi uses your brand name, mission, market position, site information, target audience and unique selling proposition when it generates topics. AI Search Gaps needs at least one buyer persona.

Step 1: Confirm Your Brand Context

The first section, Step 1: Confirm {Brand}’s Brand Foundation, shows the context your research is built on.

Review Site Information and Target Audience. Click the edit icon on either one to change it, then save. Your changes save back to your Brand Foundation.
Under Buyer Personas:, tick the personas this research is for. Each tick saves right away.

Floyi uses up to three selected personas, including their goals, pain points, values and fears.

Step 2: Define the Core of Your Research

In Step 2: Define the Core of Your Research:

Enter your core topic in 1. Define Your Core Topic:, for example “email deliverability for SaaS.”
Optionally, add terms under 2. Add Seed Keywords (Optional) that your research has to account for, such as product categories or phrases your customers use.

Your research is written in your brand’s output language, shown on the Output Language line.

Step 3: Generate Your Topical Research

The Generate Bar sits below your research views. It has four steps, and each one unlocks after the step before it. Steps 3 and 4 are marked Optional.

Generate MTs creates your top-level pillar topics (Main Topics). Start here, and review and clean your pillars before expanding further.
Generate ST2+ST3 builds second- and third-level subtopics under each Main Topic. Each second-level topic gets its own branch of third-level topics. This is the main expansion step.
Generate ST4s adds a fourth level of subtopics under each branch. Run Check depth first to see whether your research needs it (see Check Depth Before Steps 3 and 4).
Generate Keywords adds keywords to your deepest topics. See Step 6.

Once a step is ready to run, it shows its cost: Included while the run is part of your plan’s included research, or an estimate in credits. Generate Keywords always shows an estimate, because keyword generation always uses credits. Finished steps show a check, their topic count and Regenerate. When Floyi suggests a next step, its button is blue.

Check Depth Before Steps 3 and 4

Your ST2 and ST3 topics may already cover what people search for. ST4s under topics with no real subtopics left become near-duplicates. They also add more keywords to cluster, and clustering uses credits. The depth check tells you whether ST4s would add anything.

Generate ST2+ST3.
In the row under the four steps, click Check depth.

Floyi reviews your Main Topics, ST2s and ST3s and estimates how many ST3s still have subtopics people search for separately. While the check runs, Generate ST4s and Generate Keywords stay locked. The result shows in the same row:

Result	What it means	What to do next
Your topical research is deep enough	Most ST3s have no real subtopics left, so ST4s would mostly be near-duplicates.	Skip ST4s. Click Run AI Search Gaps to find what AI answers tell your buyers that your research doesn’t cover yet. Add keywords only where you want them.
Some ST3s can go deeper	Some ST3s have subtopics people search for separately.	Generate ST4s to add them as new topics. If you’d rather not add topics, add keywords instead.
Couldn’t check depth	The check didn’t finish.	Click Try again, or continue without it.

If you click Generate ST4s or Generate Keywords without a current result, Floyi checks depth first and shows the result in a dialog. Click Generate ST4s or Choose keywords to continue, or Continue without a check if the check didn’t finish. When your research is deep enough, Generate ST4s opens this dialog every time, with Run AI Search Gaps and Generate ST4s anyway.

The result applies to your research as it is. If you add, rename or delete Main Topics, ST2s or ST3s afterward, the row offers Check depth again. The row disappears once you generate ST4s.

The depth check is free. It isn’t available on the free plan, where Steps 3 and 4 need a paid plan.

Regenerating

Regenerating Main Topics replaces everything under them, and regenerating ST2+ST3 replaces the second-, third- and fourth-level topics and keywords. A Confirm Generation dialog warns you before either one runs. Topics you added from AI Search Gaps are kept where their parent still exists.

Background Generation

Generation runs in the background. While it runs:

A progress banner shows what’s being generated, with Cancel to stop it
All four Generate buttons are disabled, with the message A generation is running. Generate buttons unlock when it finishes.
Manual edits and persona changes wait until the run finishes
When the run ends, the banner shows that generation is complete and dismisses itself

Live Autocomplete: Google and YouTube

During subtopic and keyword generation, Floyi automatically queries Google Web autocomplete (and YouTube autocomplete for the ST2+ST3 step). This grounds your topics in real search behavior rather than only what the AI model learned during training.

Google Web autocomplete surfaces commercial, tool and product search patterns that reveal how users explore your topic in Google.
YouTube autocomplete adds tutorial demand, audience segments (e.g., “for musicians”, “for ecommerce”) and comparison queries that often don’t appear in web search.

This is fully automatic. There are no settings to configure, and you don’t see the autocomplete results directly. They show up in the quality of the generated topics.

Step 4: Choose Your View

Switch between three views with the Silo, Outline and Spreadsheet tabs above your research.

Silo View

The default view. Each Main Topic gets its own column.

Columns scroll independently and resize by dragging the right edge.
Topics appear as a nested tree. Use Expand All and Collapse All to open or close everything.
Each column shows how many topics it holds, and deepest topics show how many keywords they have.
Topics added from AI Search Gaps show a teal G badge (“Gap topic from AI Search”).
Best for seeing the shape and balance of your research across pillars.

Outline View

A read-only list with collapsible groups and search.

Search topics… filters across every level and keyword as you type.
Keywords display next to their topics.
Best for scanning a large research set or finding a specific topic.

Spreadsheet View

A table with columns for Main Topic, Subtopic 2, Subtopic 3, Subtopic 4 and Keywords. Use it for bulk editing (see Step 5).

On screens narrower than 1024 pixels, Silo view isn’t available, and Topical Research falls back to Outline view.

Step 5: Edit Your Research

Quick Edits in Silo View

Double-click any topic name to rename it. Press Enter to confirm or Escape to cancel.
Right-click a topic (or use its … menu) for Rename and Delete.
Deleting a topic removes it and all its children.
For topics added from AI Search Gaps, the menu shows Accept topic and Dismiss instead.

Outline view is read-only. Rename and delete in Silo or Spreadsheet.

Spreadsheet Edit Mode

For bulk editing, click Edit. Floyi switches to Spreadsheet view.

Select rows with the checkboxes, or all rows with the header checkbox.
Add a row with the + button.
Click any cell to edit it. When you rename a topic that appears in other rows, Floyi asks whether to update only this row or every matching cell.
Delete selected rows to remove off-scope topics or duplicates.

Click Save to apply your changes, or Cancel to discard them. Manual edits are free.

Step 6: Generate Keywords

Keywords are optional. Your topic names already count as keywords in clustering, so add keywords selectively, by intent.

Clean your topics first, at least down to the third level.
Click Generate Keywords in the Generate Bar. If you haven’t run the depth check on your current research, Floyi checks depth first. Click Choose keywords to continue.
In Select Keyword Generation Scope, check the topics you want keywords for. Use Select All or Deselect All to start from all or none.
Choose a Keyword Type:
- Informational Keywords Only (the default)
- Commercial Keywords Only
- Generative Keywords Only, for requests where people want AI to produce something for them, like a plan or a summary
- Info and Comm Keywords
- All 3 Types
Review the estimated cost and click Generate Keywords.

Keywords attach to the deepest topic in each branch. They add search terms to your topics and never add new topics. Keywords display at every level where they exist.

You can rerun keyword generation after adjusting topics, but each run uses credits.

Step 7: Find AI Search Gaps

After you’ve built your research, AI Search Gaps finds topics that AI search engines recommend for your audience but that your research doesn’t cover yet. Instead of starting from keywords, it starts from the questions your buyers actually ask AI assistants like ChatGPT, Google Gemini and Google AI Mode.

How AI Search Gaps Works

The analysis runs in the background in five stages:

Generating persona prompts: Floyi writes conversational questions your selected persona would realistically ask an AI assistant, reflecting their pain points, goals, journey stage and vocabulary.
Selecting best prompts: Floyi narrows them to the most diverse and representative questions.
Querying AI search engines: Floyi sends those questions to the engines you selected and collects the full answers.
Extracting topics: Floyi pulls every distinct topic out of the answers.
Matching against your research: Floyi classifies each topic as a gap, already covered, or not relevant to your brand.

Running an Analysis

The AI Search Gaps button appears in the toolbar once you have at least one buyer persona and at least one Main Topic, while your research is still editable.

Click AI Search Gaps.
Under Select a buyer persona, choose the persona whose questions to ask.
Under AI search engines to query, choose ChatGPT, Google Gemini and/or Google AI Mode. Cost grows with the number of engines.
Review the estimated cost and click Find AI Search Gaps.

A progress indicator shows each stage as it completes. You can close the modal and come back, and the results will be waiting.

Reviewing Results

Results are grouped into Gaps found, Already covered and Not relevant. Expand Prompts used to see the exact questions Floyi asked.

Each gap shows its suggested level and parent topic, how confident the match is, how many AI answers mentioned it, and any subtopics that come with it.

Adding Gap Topics to Your Research

Gaps arrive selected.

Deselect any topics you don’t want, or use Deselect All to start fresh.
Click Add N to Research to insert the selected gaps at their suggested places.
A banner confirms how many topics were added. If a topic’s suggested parent no longer exists, Floyi lists it instead of adding it.

Added topics show the teal G badge in your research. Accept or dismiss each one later from its menu in Silo view, or click Remove next to it in the results.

Click New Discovery Run to run again with a different persona or different engines. When you reopen AI Search Gaps, Floyi loads your most recent results.

Step 8: Export or Import Your Research

Open the export menu (download icon) in the toolbar:

Export to XLSX downloads an Excel file with columns for each level and keywords.
Export to Google Sheets creates a Google Sheet with the same structure. Floyi asks you to connect your Google account on first use.
Import from XLSX loads an XLSX or CSV file with the same columns.

[!WARNING] Importing replaces all of your current topics and keywords. Floyi asks you to confirm before it does. Import before you start editing.

Exports are snapshots. If you update your research later, export a fresh copy.

Summary Stats

The toolbar shows a running count of Main Topics and subtopics, including how many came from AI Search Gaps (for example, ”· 6 gap from AI Search”). The stats bar at the bottom shows your topical research allocation for the current cycle.

Included Research and Credit Usage

Included Topical Research

Each plan includes a monthly allocation of topical research runs:

Plan	Included Research
Free	1 (lifetime, limited to 3 levels, so no fourth level)
Pro	5 per month
Scale	15 per month

Research is tracked per brand. The first time you research a new brand in a billing cycle, it uses one allocation slot. Regenerating topics on the same brand within the same cycle is included. After your included research is used, additional runs use credits.

The allocation counter appears in the stats bar and in Settings > Credits Overview.

Free Plan Limitations

Free users can generate Main Topics, ST2 and ST3. Generate ST4s and Generate Keywords show Requires paid plan on the free plan, and the depth check isn’t available.

What Uses Credits

Generating topics at any level, once your included research is used
Generating keywords (always shows a credit estimate, even within your allocation)
Running AI Search Gaps (cost grows with the number of engines)

Editing your research by hand, the depth check, importing and exporting are free.

Choosing an AI Model

You choose the AI model with the global model picker. While a run is included in your plan’s allocation, Floyi uses its default model for that run, whatever the picker shows.

How Topical Research Connects to Other Tools

PAA Explorer: expand any topic’s People Also Ask questions into a question tree for FAQ and subtopic ideas.
SERP Clustering: when your research is ready, click Start Clustering above your research and confirm with Let’s Get Started. Floyi clusters every topic and keyword by the search results they share. Starting clustering locks your topical research, so finish your edits and gap runs first.
Topical Map: your clusters become the hierarchy of your Topical Map.

A simple pattern if you’re new to Floyi:

Set your Brand Foundation and one main persona.
Run Topical Research for one core topic.
Generate MTs, then ST2+ST3. Review and clean.
Click Check depth. Generate ST4s only if some ST3s can go deeper.
Run AI Search Gaps to find topics AI engines recommend that your research misses.
Add the relevant gaps.
Generate keywords by intent where you want more search terms.
Start clustering.

That gives you a focused, strategy-aligned set of topics instead of a random keyword list.