Welcome to Guru

Guru is a managed SEO and GEO content operation. AI agents run your keyword research, title generation, technical audits, and implementation - and every proposed change goes through a structured approval workflow before anything touches your live site.

How Guru works

Every month, Guru's AI agents run a set of standard procedures against your data - pulling from Google Search Console, DataForSEO, and your site's content profile. The output is a prioritised queue of changes, each one reviewed and approved through your portal before implementation begins.

Once approved, changes are applied directly to your WordPress CMS via the REST API: title tags, meta descriptions, H1s, schema markup, FAQ sections, and internal links. A verification pass runs after implementation to confirm the change landed correctly.

What the platform covers

• On-Page SEO - title proposals scored for SEO and GEO potential, meta descriptions, H1 optimisation
• Content Pipeline - keyword-driven article titles, AI-written articles, full pipeline from approved title to published URL
• Technical SEO - per-URL indexation status via Google Search Console, re-crawl submission via the Indexing API, schema and structured data implementation
• GEO & LLM Visibility - entity gap analysis, GEO scoring, citation pattern optimisation for ChatGPT, Claude, Perplexity, and Gemini
• Reddit Monitoring - brand mention tracking across Reddit with sentiment analysis
• Monthly Reporting - GSC rankings, GA4 traffic, content output, and AI mention detection in one report

Your client portal

Your portal is your single view into everything Guru is doing for your site. It's accessed via a unique link - no shared accounts, no login friction. Each section updates in real time as agents run and changes are implemented.

Portal sections

• Dashboard - live count of pending approvals, active agent jobs, and implemented changes this month
• Approvals - the main decision queue. Every proposed change with full before/after detail, confidence level, and plain-English reasoning
• Title Proposals - request new article titles, review AI-generated proposals, approve titles to enter the writing pipeline
• Keywords - view your keyword groups, generate new groups on demand, and see search volume and difficulty data per keyword
• Content Pipeline - Kanban board tracking every article from title approval through writing to publication
• Content Optimization - existing pages being refreshed, separate from new content
• Indexation - per-URL Google index status, submission controls, and coverage state from GSC
• Reports - monthly performance summaries combining GSC, GA4, content output, and AI visibility metrics
• Deliverables - what's included in your monthly plan and progress against each deliverable
• Activity - full audit log of every approval, implementation, and status change

Month 1 walkthrough

Month 1 is the foundation sprint. The scope is deliberately focused: your main navigation pages get optimised first - titles, metas, H1s, schema, and FAQs - before the broader audit expands to the rest of the site.

How approvals work

Every change Guru recommends enters your approval queue before anything happens. The queue is the single place where you stay in control of what goes live on your site.

Each item in the queue shows: the URL it applies to, the change category (On-Page, Content, Technical, or AI-GEO), the current value, the proposed value, a confidence level, and a plain-English explanation of why this change is being recommended.

Actions you can take

• Approve - queues the change for implementation. The agent applies it to your site, then verifies it.
• Skip - removes the change from the queue. It won't be re-proposed.
• Question - leave a note asking for clarification. The item stays in your queue and the team will respond.
• Backlog - deprioritises the item without skipping it. It will be revisited in a future sprint.

Approval states

Every change in Guru moves through a defined lifecycle. These are the states a change can be in.

Pending

The change has been surfaced by an agent and is waiting for your decision. Nothing about your site has changed. This is the state where you review the recommendation and decide what to do with it.

Approved

You've approved the change. It's now queued for implementation. The implementation agent will pick it up and apply it to your site via the WordPress REST API. For most changes, this happens within minutes of approval.

Implementing

The agent is actively applying the change to your CMS. After writing the change, the agent reads the live page back to verify the value was applied correctly. You can see implementation progress in your Activity feed.

Complete

The change is live on your site and has passed verification - meaning the agent confirmed the value it wrote matches what's now on the page. Completed changes are tracked in your monthly report.

Skipped / Backlog

Skipped changes are permanently removed from your queue. Backlogged changes are deprioritised but may resurface in a future sprint if conditions change (e.g. ranking drops, new competitors appear).

Change categories

Changes are grouped into four categories, each with its own sub-queue in the Approvals section of your portal.

On-Page

Changes to title tags, meta descriptions, and H1 headings. These are the most common first-month changes. Each shows the current value and the proposed replacement, with an explanation of why the new version performs better.

Content

Body copy additions, FAQ section insertions, and content refresh work. FAQ sections are injected via a WordPress get_footer hook using Code Snippets - not via Elementor widgets - which is more reliable for large sites.

Technical

Schema markup additions, redirect rules, and structural changes. Schema is injected as JSON-LD via a WordPress wp_head hook, conditional on the page ID. Technical changes may require a brief design review before implementation proceeds.

AI-GEO

Entity optimisations, AI mention additions, and content restructuring for LLM citation potential. These changes target how ChatGPT, Claude, Perplexity, and Gemini read and cite your pages. Each change includes an entity coverage score and an explanation of the citation pattern being targeted.

Title proposals

Title proposals are AI-generated article titles designed to rank on Google and be cited by AI systems. Each proposal is generated from your keyword groups, enriched with live SERP data from DataForSEO (top 10 competitor titles and snippets), and scored on two dimensions.

Scoring

• SEO score - keyword alignment, length, intent match, and differentiation from top-ranking competitors
• Quality score (1–5) - specificity, uniqueness, anti-pattern check, and vertical fit. Titles scoring below 3 are not surfaced. Banned patterns include "Complete Guide to…", "Ultimate Guide", "Why [Topic] Matters", and excessive use of colons or listicle formats.

Requesting a title

Open Title Proposals in your portal, enter a direction (e.g. "something about migrating from HubSpot to Salesforce"), select a keyword group, and optionally choose a search intent. The agent generates a title proposal with SERP context in seconds.

You can edit the title directly, ask the agent to regenerate with a new direction, or approve it as-is. Once approved, the title enters the content writing pipeline and is assigned to a writer.

Content types

• Standard article - core blog content, typically 1,000–2,000 words, targeting a specific keyword
• Longform article - pillar content, 3,000+ words, broader topic coverage with internal linking structure
• Content refresh - an existing page being updated and re-optimised (tracked separately in Content Optimization)

Keyword groups

All content strategy in Guru is organised around keyword groups - clusters of related keywords targeting a distinct topic pillar. Each group contains five subkeywords, each with volume, difficulty, intent, and priority data from DataForSEO.

Standard groups (set up in Month 1)

Guru generates five standard groups in Month 1 based on your site, GSC data, and competitor analysis. These cover the five intent stages:

• Branded - queries that include your company or product name
• Problem-Aware - informational searches where the user has a problem but not a solution
• Solution-Aware - commercial investigation, evaluating options like yours
• Comparison - late-stage commercial queries (alternatives, vs., best X for Y)
• Transactional - pricing, demo, trial, and signup intent

Generating a custom group

You can request additional keyword groups at any time from the Keywords section of your portal. Enter a topic direction, and the agent generates a group of five subkeywords with DataForSEO enrichment. Custom groups are saved permanently and feed into title generation alongside the standard groups.

Priority tiers

Within each group, the highest-opportunity subkeyword is ranked high, the next two are medium, and the lowest two are low. Opportunity is calculated as a function of search volume and keyword difficulty - high-volume, lower-difficulty keywords rank first.

Content pipeline

Once a title is approved, it enters the content pipeline. The pipeline Kanban tracks every article from title approval through writing, review, and publication.

Pipeline stages

• Titled - title approved, awaiting writing assignment
• Queued - writing job dispatched to the pipeline
• In Progress - article is being written (typically takes 8–12 minutes for a standard article)
• Complete - draft finished, ready for your review
• Published - article has been submitted to WordPress as a draft and is ready to publish

How writing works

The writing pipeline pulls live competitor HTML for the target keyword, builds an outline, writes the article, and generates metadata - all using Claude AI. The finished draft is submitted to your WordPress site via the REST API as a draft (never auto-published). You review and publish it in your CMS.

Content optimization

Content optimization covers refresh work on existing pages - updating outdated content, improving keyword alignment, adding new sections, and improving GEO scoring for pages that are already indexed but underperforming.

Refresh jobs appear in the Content Optimization section of your portal, separated from new content. Each job shows the target URL, the type of changes being made, and the current status.

As with new content, all proposed changes go through the approval queue before implementation. After implementation, the URL is submitted to the Google Indexing API for re-crawl.

Monthly quotas

Each plan tier includes a fixed monthly allocation of content deliverables. Quotas reset on the first of each month. Unused quota does not roll over.

Indexation monitoring

The Indexation section shows live Google index status for every URL in your site that Guru has worked on. Data comes from the Google Search Console URL Inspection API, which returns the actual coverage state Google has on record for each URL.

Coverage states

What the metrics mean

• Total pages - all URLs Guru is tracking for this site
• Indexed - pages confirmed indexed by Google (verdict: PASS)
• Issues - pages that are blocked, not indexed, or returned an error
• Not checked - pages not yet inspected via the URL Inspection API, or whose last check was more than 7 days ago

On page load, the portal auto-refreshes any URL whose GSC data is more than 7 days old. You can also manually refresh individual URLs or all URLs at once using the refresh controls in the Indexation view.

Submitting URLs to Google

After a change is implemented on a page, Guru submits that URL to the Google Indexing API to request a re-crawl. This tells Google to revisit the page and reflect the updated content in search results.

How submission works

Submissions happen automatically when a change's status moves to Complete. The portal records the submission date and then tracks the coverage state via GSC inspection on a rolling basis.

You can also manually submit any URL from the Indexation view - enter the URL and click Submit. This is useful for pages you've updated outside of Guru.

After submission

Most pages are crawled within 24–72 hours of submission, though this varies by site authority and crawl budget. You can check the Last crawled date in the Indexation view - this is the timestamp Google most recently visited the URL.

GEO & entity scoring

GEO (Generative Engine Optimization) is the practice of structuring content so it's cited by AI systems - ChatGPT, Claude, Perplexity, and Gemini - when they answer questions related to your domain.

Entity coverage score

Each page tracked by Guru is assigned an entity coverage score (0–100). This reflects how well the page covers the key named entities - your brand name and variations, your product names, your location, your industry terms, and the entities LLMs associate with your topic area.

The AI-GEO approval category surfaces specific entity gaps: places in your content where adding or repositioning a mention of a key entity would improve citation probability. Each suggestion includes which LLMs it targets and why the entity placement matters.

What improves GEO score

• Clear entity mentions in the first 200 words of a page
• Structured, extractable content (definition-style sentences, numbered lists, comparison tables)
• Topic authority signals - internal links from related pages, consistent use of topic terminology
• Matching the query patterns that LLMs use to retrieve answers in your category

AI mention tracking

Your monthly report includes an AI mentions section showing where your brand was detected in AI-generated answers across major LLMs during the reporting period. This data feeds the GEO optimisation recommendations for the following month.

Reddit monitoring

Reddit is a significant signal source for both Google rankings and LLM training data. Guru tracks your brand's mentions across Reddit, a dedicated Reddit brand monitoring tool.

What you see

• Mention feed - individual posts and comments that mention your brand, with direct links to the Reddit thread
• Sentiment breakdown - positive, neutral, and negative mention counts for the period
• Top subreddits - the communities where your brand is discussed most
• Upvote and comment counts - to gauge the reach of each mention

Reddit engagement

Depending on your plan, Guru's team posts a set number of engagement comments per month in relevant Reddit threads - contributing to discussions in a way that's authentic and topically aligned. Starter: 2 comments/month. Growth: 5 comments/month. Authority: 10 comments/month.

Monthly reports

Your monthly report is generated automatically at the end of each month and appears in the Reports section of your portal. It combines data from GSC, GA4, and the changes Guru implemented during the month.

Report sections

• Changes summary - total changes implemented, broken down by category (On-Page, Technical, Content, AI-GEO), with notable optimisations highlighted
• Search Console metrics - clicks, impressions, CTR, and average position vs. the prior month, plus a 3-month trend chart
• Ranking movements - top 10 keywords that gained positions, top 10 that lost positions, and new keywords entering the top 20
• GA4 traffic - sessions and users vs. prior month, with AI-sourced traffic attribution broken out separately
• Content output - articles published this month, plus which published articles are now ranking in the top 20
• AI mentions - platforms where your brand was detected in AI-generated answers during the reporting period

Integrations

Guru connects to several external services. All integrations are configured during onboarding by the Guru team - you don't need to set up API credentials manually.

Google Search Console

Powers keyword research, title proposal scoring, indexation monitoring, and ranking reports. Guru uses both the Search Analytics API (clicks, impressions, positions) and the URL Inspection API (coverage state, verdict, last crawled date). Connected via OAuth.

OAuth - set up by Guru team

Google Analytics 4

Adds session and user data to your monthly report, including AI-sourced traffic attribution (organic search from AI overviews and LLM referrals). Connected via OAuth.

OAuth - set up by Guru team

Google Indexing API

Used to request re-crawls after each page implementation. Guru submits URLs automatically when changes are marked complete. Quota: 200 submissions/day. The portal shows a warning when you're approaching the limit. Uses a Google Service Account.

Service account - set up by Guru team

Reddit

Guru monitors your brand mentions across Reddit, tracking sentiment, top subreddits, post reach, and individual threads. Growth and Authority plans include monthly engagement comments on relevant threads.

FAQ

Get support

Three ways to get help:

• Book a walkthrough - 30-minute live session. We'll walk you through the approval workflow with real data from your site.
• Email support - Growth and Authority plans get same-business-day responses. Starter gets two business days.
• In-portal questions - use the Question action on any approval item if you're unsure about a specific recommendation. The team reviews every question and responds in the portal.