Skip to content

Topical Audit

The Topical Audit tool discovers the content structure of any public website—without needing a pre-existing Topical Map. It’s perfect for analyzing competitor sites or understanding a domain’s topical footprint.

  • How to run a topical audit on any website
  • Using CSV upload to save 50% on credits
  • Understanding your audit results
  • Reading AI-generated insights
  • Navigating the discovered hierarchy in Tree View or the Visual Map
  • Refining clusters that come out too broad or too granular
  • Managing past audits
  • Exporting your findings

Use Topical Audit when you want to:

  • Analyze competitor content strategies — See how competitors structure their content
  • Audit a new domain — Understand an acquired site’s topical coverage
  • Reverse-engineer authority — Discover what topics a site has built authority around
  • Benchmark your content — Compare your structure against industry leaders
  1. Navigate to Topical Audit from the main menu.
  2. Enter the website URL you want to analyze (e.g., https://example.com).
  3. Configure your scan settings:
SettingDescription
URLThe website to analyze
CountrySelect geo-location for page loading
Page LimitMaximum pages to crawl (1-5,000)
Force CrawlEnable if the site lacks a sitemap (uses 1.5x credits)
  1. Review the estimated credits.
  2. Click Start Audit to begin.

If you already have page content extracted from a competitor site or any public website, you can upload a CSV file instead of crawling. This method saves 50% on credits since Floyi doesn’t need to fetch and extract content from the pages.

  1. Click the Upload CSV button in the input mode toggle (next to “Crawl Website”)
  2. The form will switch to CSV upload mode

Your CSV file must include:

ColumnRequiredDescription
urlYesThe full page URL
contentYesThe main content of the page
titleNoPage title (optional but recommended)
h1NoPrimary H1 heading
meta_descriptionNoPage meta description
  • Minimum 5 pages required per upload
  • Maximum 5,000 pages per upload
  • Supported formats: .csv, .xlsx, .xls
  1. Click Download CSV Template to get a properly formatted file
  2. Fill in the page data you’ve collected
  3. Save the file and upload it
  1. Click the upload area or drag and drop your CSV file
  2. Floyi validates your file and displays:
    • Number of pages detected
    • Any validation errors (missing required columns, etc.)
  3. Review the estimated credits (50% of standard crawl rate)
  4. Click Start Audit to begin analysis

Once started, you’ll see real-time progress through four stages:

  1. Discovering — Finding pages via sitemap or crawling
  2. Crawling — Extracting content from discovered pages
  3. Analyzing — Processing content for semantic understanding
  4. Clustering — Grouping related content into topics

You can navigate away during scanning. You’ll receive an email when complete.


After completion, you’ll see key metrics at a glance:

MetricDescription
Topics DiscoveredAll topics found across every level of the hierarchy
Pages AnalyzedPages successfully processed and clustered
Pillar countShown beside the results heading, with the total topic count

Content is organized into a 4-level structure:

LevelNameDescription
1PillarBroad topic categories (e.g., “Marketing”)
2HubMajor subtopics (e.g., “Content Marketing”)
3BranchSpecific topic areas (e.g., “Blog Strategy”)
4ResourceIndividual content pieces

Each topic shows:

  • Topic name — The semantic theme of this group
  • Topic Strength — How cleanly this topic separates from the others
  • Page count — Direct URLs assigned to this topic, and the total including its subtopics
  • Match % — Average similarity of pages to this topic’s own centre
  • Primary URL — The page that best represents this topic

Topic Strength is the number to trust. It measures how cleanly a topic separates from its neighbours:

BandValueMeaning
Strong0.25+Pages clearly belong here
Moderate0.10 – 0.25Overlaps a sibling topic
Overlapping0 – 0.10Sits on the boundary between topics
MisplacedBelow 0Pages are closer to a different topic

Match % is the average similarity of pages to their own topic centre. It is high by definition and climbs as topics get smaller, so a 97% match on a two-page topic is not evidence of a better topic than an 88% match on a forty-page one. Judge quality on Topic Strength; read Match % as supporting detail.

The default view. Each pillar is a card you can drill into.

  • Click Arrows to expand/collapse sections
  • Click Expand All or Collapse All to toggle the entire tree
  • Click any URL to open it in a new tab
  • View Included URLs to see all pages grouped under a topic

Switch to Visual Map for the same results as an interactive picture of the site’s topical structure. The site sits at the centre, each pillar radiates outward, and hubs and branches sit behind them. Depth always runs outward, so a topic’s subtopics are always further from the centre than the topic itself.

The map opens showing topics only. Tick Show pages to add every analysed page as a small node fanned out from the topic it was grouped into. On a big site those page nodes are the large majority of the map, so read the structure and topic names first and turn them on when you want to see how the page volume is distributed.

ControlWhat it does
Color byTopic Strength shades every topic by band so weak and overlapping topics stand out. Level shades by depth
Show pagesAdds the individual page nodes. Off by default — pages are usually most of the map, so the structure reads more clearly without them
LegendClick a swatch to filter the map to that band. Click again to clear
HoverShows strength, match %, page count and the primary URL
ClickIsolates a topic and its connections
Double-clickOpens that page in a new tab
Drag / scrollRearrange nodes, zoom, and pan
Reset zoomFrames the whole map again after you have zoomed or panned
FullscreenExpands the map to the full window. Press Esc to exit
Export PNGDownloads the current map as an image, matching whatever is on screen

Two panels sit below the hierarchy:

  • Divergent pages — pages that clustered with nothing else, usually one-off posts outside the site’s core themes. They are listed separately rather than posing as single-page pillars.
  • Excluded from topic analysis — utility pages such as the homepage, contact, legal and cart. They carry no topical signal and would pull unrelated pages together, so they are held out of clustering and shown here for transparency.

If the topics come out too broad or too granular, click Refine Clusters and adjust the sensitivity slider. Negative values produce tighter, smaller topics; positive values produce broader groups.

This re-runs clustering on content Floyi has already stored, so no re-crawl needed. Topic names are regenerated, so they will not be word-for-word identical to the previous run.


After analysis, you’ll receive AI-generated insights about the site’s content strategy.

A 0-10 rating measuring how focused the site’s content strategy is. This is a site-level judgement of the whole domain — distinct from Topic Strength, which scores one topic at a time:

ScoreMeaning
8-10Highly focused, clear topical authority
5-7Moderate focus, some topic drift
1-4Scattered content, weak topical clustering
CategoryWhat It Tells You
StrengthsTopics where the site shows strong authority
Missing TopicsSemantically relevant topics not covered
Divergent TopicsContent that doesn’t fit the main strategy
Expansion OpportunitiesSuggested subtopics to build out authority

  1. Click the Past Audits tab

  2. Browse your audit history with:

    • Domain name
    • Scan date
    • Pages analyzed
    • Status
  3. Click any audit to load its full results

Select multiple audits using checkboxes to:

  • Delete — Remove selected audits to keep your workspace clean

Past audits are paginated. Click Load More to view additional history.


Export your discovered hierarchy for analysis or sharing.

Click the Export dropdown and choose:

FormatBest For
Excel (.xlsx)Spreadsheet analysis with multiple sheets
CSVDatabase imports, custom tools
Markdown (.md)Documentation, reports
Text (.txt)Simple flat-file output
PDFClient-ready reports
Word (.docx)Reports you want to edit before sending
HTML reportOpening the full report in a browser tab
Google SheetsCollaborative analysis
Google DocsCollaborative write-ups

Exports contain:

  • Complete topic hierarchy
  • Page counts, Topic Strength values and bands, and match percentages
  • Primary URLs for each topic
  • All grouped URLs
  • AI Insights summary

Click Copy to quickly copy the hierarchy as formatted text for pasting into documents or chat.


Yes, Topical Audit works on any publicly accessible website. Private or password-protected sites cannot be scanned.

FeatureTopical AuditOrganic Audit
PurposeDiscover structure of any siteSee how Google Search sees your topics
Data SourceWebsite crawlGoogle Search Console
Requires MapNoYes
Best ForCompetitor researchValidating your strategy with real search data
OutputDiscovered hierarchyClicks, impressions, position per topic

Why do some topics show low Topic Strength?

Section titled “Why do some topics show low Topic Strength?”

A low Topic Strength means the grouped pages don’t separate cleanly from neighbouring topics. This can mean:

  • The topic covers diverse subtopics
  • It overlaps a sibling topic, and the two may really be one
  • The topic name is a best-fit label for disparate content

A Misplaced score (below 0) is the strongest signal: those pages sit closer to a different topic than their own. Color the Visual Map by Topic Strength to see where the weak topics cluster.

Typically 5-15 minutes for 100 pages, depending on site complexity. Larger audits (500+ pages) may take 30+ minutes.

Yes! Run multiple audits over time to track how a site’s content strategy evolves.

When should I use CSV upload vs. crawling?

Section titled “When should I use CSV upload vs. crawling?”

Use CSV upload when:

  • You’ve already extracted competitor content using your own scraping tools
  • You have content exports from a CMS or database
  • You want to save 50% on credits
  • The target site has complex JavaScript rendering or bot protection

Use website crawling when:

  • You want quick, automated analysis without manual data preparation
  • The site has a standard sitemap the crawler can follow
  • You don’t have pre-extracted content available