TL;DR:
- Profound AI tracks brand citations across ChatGPT, Perplexity, Google AI Overviews, and other LLMs using real user conversation data, not simulated prompts, which produces a more accurate picture of actual citation patterns.
- Core value is diagnostic: the platform maps competitor share of voice and surfaces prompt-level visibility gaps across your priority query set.
- The Agents feature supports AEO content workflows, but production execution still requires a team to set strategy, review outputs, and manage content governance.
- AI-referred pipeline requires optimization across three surfaces: web search, citations, and training data. Profound focuses primarily on the citations surface.
- This review covers what Profound does well, where its attribution limits apply, and when a service-led execution partner fills what the software cannot deliver on its own.
Most B2B SaaS marketing teams evaluating Profound AI as part of a broader AI visibility platform evaluation are solving the same problem: their brand is absent from LLM responses while competitors get cited repeatedly. The platform shows you where those gaps exist. The harder question is what happens after the dashboard tells you.
This review evaluates Profound's citation tracking methodology, data refresh architecture, CRM integration limits, and pricing structure. We also cover when software-only tooling produces "dashboard fatigue" and what a full execution stack looks like alongside it.
Targeting Profound AI for B2B SaaS growth
AI visibility tools track how often and how prominently your brand appears inside LLM-generated answers. For B2B SaaS companies, this matters because buyers now use AI assistants to evaluate vendors before visiting a website. If your brand is absent from those answers, you're missing pipeline the sales team never sees and cannot attribute.
The discipline of optimizing for these citation surfaces is called Answer Engine Optimization (AEO) or Generative Engine Optimization (GEO). It covers three distinct surfaces, as we detail in our three surfaces AI visibility guide: web search, citations, and training data. Profound focuses primarily on the citations surface, which is the most commercially important for most B2B SaaS teams.
How Profound tracks AI citations
Profound builds its citation dataset from real prompts collected from active AI assistant users. The company applies statistical modeling to correct for demographic and geographic bias, so the trends reflect real-world query populations rather than a narrow sample. The Profound prompt volumes page describes the methodology in detail.
This is a meaningful distinction from tools that rely purely on API simulation. Simulated prompts miss the variation in how real users phrase questions, which affects which brands and pages get cited in practice.
You'll get the most value from Profound in three scenarios for B2B SaaS teams:
- Competitor share of voice: See which brands get cited more often than yours across a defined prompt set, broken down by LLM engine.
- Unbranded query coverage: Identify category-level and problem-level queries where competitors appear but your brand does not.
- Prompt-level diagnosis: Understand which specific question patterns trigger citations for your brand and which do not.
These use cases help a CMO build a defensible board slide showing current citation rate versus competitors. For the $2M to $50M ARR range, that diagnostic baseline is the starting point for any AI visibility investment. See our B2B SaaS case studies for how attribution paths look when the full stack is in place.
Traditional SEO platforms like Ahrefs and Semrush historically focused on keyword rankings and backlink profiles, describing performance on the web search surface. By 2026, both platforms introduced AI visibility tracking to monitor which competitors get cited in AI-generated answers like ChatGPT, Perplexity, and Gemini. However, their core strength remains web search measurement, not the passage retrieval mechanics that drive LLM citations.
Ahrefs data shows the overlap between Google top-10 rankings and AI-cited pages fell from 76% in mid-2025 to 38% by early 2026. Our own analysis of 2 million citations and 10,000 pages confirms that retrieval mechanics for LLMs have meaningfully diverged from classic ranking signals.
Profound collects data in three phases: prompt ingestion from real user panels, LLM response logging across supported engines, and statistical processing to produce brand mention rates by query theme. The platform surfaces this data through a dashboard showing share of voice, citation frequency, and competitive position.
Tracking ChatGPT, Claude, and Gemini
Profound supports tracking across multiple engines, but coverage varies by pricing tier. The Starter plan at $99/month covers ChatGPT only, 50 prompts, and one seat. The Growth plan at higher tiers adds Perplexity and Google AI Overviews with expanded prompt volumes. Gemini, Claude, and Microsoft Copilot tracking requires Enterprise pricing. For a B2B SaaS team where buyers actively use Claude and Gemini, single-engine visibility data gives an incomplete picture.
Tracking brand mentions in AI responses
Profound's Agent Analytics feature connects to your CDN (Cloudflare, Akamai, Fastly, AWS CloudFront, Google Cloud CDN, and Netlify, with WordPress support coming soon) and reads server logs to show which AI bots visit which pages. It identifies GPTBot, PerplexityBot, ClaudeBot, GoogleOther, and others, then cross-references IP ranges to filter spoofed crawlers.
Profound's citation monitoring captures every URL and domain that LLMs reference when answering questions. The platform surfaces these at both the domain and individual page level, showing you exactly which content assets drive citation wins. This granularity helps teams map specific content pieces to AI visibility outcomes.
How Profound handles data staleness
Profound's data refresh frequency varies by plan tier. For teams running active content optimization programs, understanding when citation data updates helps set realistic expectations for measuring content changes. Many visibility tools overstate precision because they test prompts in ways that don't replicate real retrieval conditions.
How Profound validates AI citations
Profound partners with Notified to connect press release distribution with AI citation tracking. When a press release is distributed through Notified's network, teams can measure how that content gets discovered and cited across AI answer engines. This is useful for comms teams running product launch or funding announcement content. For organic citation building at scale, treat press release distribution as a supporting tactic rather than your primary driver.
How Profound identifies and captures AI queries
Profound's query identification methodology starts with its real-user prompt corpus and applies topic clustering to group similar questions into themes. The output is a prompt universe showing which question patterns are most common for your category and which competitors dominate each theme.
Optimizing queries for buyer intent
Most queries in Profound's prompt universe don't drive pipeline. For B2B SaaS teams, commercial value concentrates in queries that map to vendor evaluation, category selection, and problem-to-solution matching. Informational queries about general industry topics may generate AI citations but rarely produce AI-sourced MQLs.
The filter to apply is: does this query appear when a buyer is choosing between vendors or evaluating whether to solve the problem at all? The AEO payback period post walks through how to model which query clusters justify optimization investment at different ARR levels.
Measuring competitive citation gaps
Profound's Topic-Level Visibility feature organizes citation data by theme rather than individual keywords. This is genuinely useful for a CMO: instead of seeing individual keyword performance, you see topic-level share of voice showing where competitors dominate category conversations and where your brand has opportunity.
That gap is the board slide. It quantifies the problem in terms a CFO can interpret without needing to understand LLM retrieval mechanics. Our Reddit and ChatGPT influence research, analyzing 144,000 AI citations, found that Reddit appeared in roughly 27% of ChatGPT's internal search slots during query processing despite appearing in only 0.35% of visible citations. A links-only view of off-page work misses a significant share of what shapes AI answers.
Profound's architecture centers on prompt-to-response logging rather than API simulation, which gives it a more accurate picture of real-world citations than tools that generate synthetic prompts. The platform also includes an Agents feature with a drag-and-drop workflow builder for AEO content research, writing, and publishing. The interface allows any marketer or content team member to build and run workflows using pre-built templates without coding, though teams still need someone to set strategy, review outputs, and manage content governance.
Mapping content to buying stages
Profound allows prompt categorization by buying stage, distinguishing awareness-level queries from decision-stage queries. This is useful for identifying whether your brand appears at the right point in the research process or only in early-stage informational responses that rarely convert. For a B2B SaaS team, the highest-value citation wins come from comparison queries and vendor selection queries, where the buyer is actively evaluating options.
Syncing CRM data to track AI-sourced MQLs
Connecting AI citation data to Salesforce pipeline is what CMOs request most often and what vendors overstate most frequently. Attribution between citation visibility and pipeline fields typically requires custom integration work, whether through API development, middleware platforms, or manual data workflows.
Streamlining Salesforce data workflows
The middleware stack for AI-to-Salesforce attribution requires custom API development, a connector platform like Zapier or Make, and custom Salesforce fields for "AI-referred" lead source. Most marketing operations teams can build this, though setup time varies and the solution does not solve the underlying referrer-stripping problem at the browser level.
Mapping AI traffic to Salesforce UTMs
When users click a link from inside a ChatGPT or Claude response, the app often strips referrer data, causing the session to register as direct traffic in GA4 (Google Analytics 4). The practical workaround is a UTM convention applied to your highest-priority pages. These UTMs survive the referrer strip if the user clicks a link that carries them. However, they don't capture zero-click behavior, where the AI answers the question without the user visiting your site.
Tracking AI-referred MQLs in Salesforce
Self-reported "how did you hear about us?" fields on your demo and contact forms provide a direct attribution signal for AI-sourced MQLs (marketing qualified leads). When you combine this with UTM data from sessions that do carry referrer information, you get a triangulated picture: UTM-tracked AI sessions plus self-reported AI discovery. Monthly reporting on direct traffic volume trends correlates with citation rate improvements and provides a secondary signal for the board narrative.
Deployment timeline for CRM integration
A realistic timeline for a clean tracking stack spans multiple weeks for implementation and calibration. You'll need time for UTM convention documentation and form field additions, Salesforce custom field configuration and connector setup, initial data collection, and calibration before the numbers stabilize for executive reporting.
Using Profound effectively requires a structured deployment that sequences diagnosis, optimization, and measurement in the right order. The teams that get value from the tool define a tightly scoped prompt universe of high-value buyer queries first, not a sprawling report across every competitor.
Week 1: defining your AI visibility baseline
In week one, configure your initial prompt set around your most commercially important buyer queries. Start with your top conversion-driving keywords from Google Search Console, rephrase them as natural language questions, and add direct competitor comparison queries. This gives you a starting citation rate across your priority query universe to carry into board review.
Weeks 2-4: measuring baseline AI citations
Run weekly pulls across the prompt universe and log citation rates by engine and by query theme. You're looking for the pattern, not the perfect number. Which queries produce citations consistently? Which produce citations for competitors but never for you? Most teams find that the majority of priority queries have zero citations for their brand at baseline, which defines the content gap map.
Months 2-4: measuring AI-attributed pipeline
Connecting citation improvements to pipeline requires the Salesforce UTM stack described above and at least 90 days of data. The incident.io engagement shows the trajectory: within four months of working with Discovered Labs, organic meetings booked grew 22%, with AI visibility climbing from 38% to 64%, as detailed in the incident.io case study.
"I have recommended you to multiple peer CMOs. There are large organizations like Hubspot and Ramp who have dedicated teams to work on large projects like AEO. For everyone else (except my competitors) there's Discovered Labs!" - Tom Wentworth, CMO at incident.io (incident.io case study)
The diagnosis came first. The citation lift required structured content execution against specific buyer queries, off-page consistency work, and schema implementation, none of which the software delivered on its own.
Transparent monthly pricing and flexible terms
Profound publishes three pricing tiers publicly, but the features most B2B SaaS teams need (Gemini and Claude tracking, higher prompt volumes, daily refresh) sit behind enterprise pricing that requires a demo.
Tiered retainer options explained
Profound's Starter plan at $99/month covers ChatGPT only, 50 prompts, and one seat. The Growth plan at $399/month adds Perplexity and Google AI Overviews with expanded prompt volumes. Enterprise plans covering 10-plus models require custom pricing quotes. At $99/month for ChatGPT-only coverage, the cost per prompt is approximately $2, which positions Profound competitively in the single-engine monitoring category.
Month-to-month vs annual commitment
Annual software contracts in the AI search category carry real risk. Retrieval mechanics change with each model update, and a platform's citation data for one version of GPT-4o does not automatically transfer to later releases. Committing to 12 months of a tracking tool in this environment means paying for a measurement framework that may need replacing mid-contract.
Discovered Labs' retainers are month-to-month at every tier. No annual lock-in on any service.
Core features by pricing tier
Table 1: Platform capabilities by tier
Feature | Profound Starter ($99/mo) | Profound Growth ($399/mo) | Discovered Labs Starter (€6,995/mo) |
|---|
LLMs covered | ChatGPT only | ChatGPT, Perplexity, Google AIO | ChatGPT, Claude, Perplexity, Gemini, Google AIO |
Prompt volume | 50 | 100+ | Custom query map |
Citation tracking | Page-level + brand | Page-level + brand | Citation rate + passage-level |
Salesforce integration | Custom integration | Custom integration | Custom UTM + CRM mapping |
Table 2: Execution capabilities and terms
Feature | Profound Enterprise | Discovered Labs Starter (€6,995/mo) |
|---|
LLMs covered | 10+ models | ChatGPT, Claude, Perplexity, Gemini, Google AIO |
Content execution | Agents workflow builder | Up to 20 CITABLE articles/mo (managed) |
Off-page consistency | Contact for details | Reddit + brand consistency |
Schema implementation | Contact for details | Included |
Contract terms | Contact for details | Month-to-month |
Understanding our service fee breakdown
Discovered Labs' Starter retainer at €6,995/month covers up to 20 SEO and AEO articles built using the CITABLE framework, AI visibility tracking and competitor monitoring, structured data implementation, backlinks and brand consistency work, and strategic Reddit engagement. The team includes an SEO manager, SEO specialist, off-page specialist, and content editor. That's the execution stack Profound's software assumes you have assembled internally.
When to deploy Profound for AI visibility
You should choose Profound when you have a clear brief on which queries matter, an execution team to act on the gaps it identifies, and a CRM integration plan ready to build. Without those three elements, the platform produces a monthly report that circulates internally but does not change what gets published.
Best fit: $2M to $50M ARR SaaS
This ARR range can justify dedicated AI visibility tracking because the pipeline math supports it. A 20% lift in AI visibility for a $10M ARR company with an average deal size of $25,000 and a 5% AI-sourced lead volume could represent several hundred thousand dollars in incremental annual pipeline. See our AEO cost model for B2B for the payback period math by ARR range.
Staffing models for AI search growth
An AI visibility platform requires at minimum one person who reviews citation data regularly, maps gaps to content priorities, briefs writers, and tracks citation rate changes post-publication. Teams without a dedicated SEO or content function accumulate dashboard data without acting on it. Our AEO agency vs in-house cost breakdown covers the full staffing comparison.
Integrating with your current CRM
Before deploying Profound, the minimum prerequisites are a custom lead source field in Salesforce or HubSpot that captures "AI-assisted discovery," a UTM convention applied to all high-priority pages, and a "how did you hear about us?" field active on your demo and contact forms. Without these in place, Profound's citation data and your pipeline data remain in separate systems.
The AI visibility tracking market has expanded significantly with multiple software platforms now available. The most relevant comparison for B2B SaaS teams is not software versus software but software versus service-led execution.
Profound is a software-first platform for AI visibility tracking, while Discovered Labs combines proprietary tracking with managed execution. Peec AI is another software-only citation tracking tool that offers competitive monitoring capabilities.
Capability | Profound AI | Peec AI | Discovered Labs |
|---|
Citation tracking across LLMs | Yes (tier-gated) | Yes | Yes (proprietary tracker) |
Managed content creation | Agents workflow builder | No | Yes (CITABLE framework) |
Off-page consistency building | Contact for details | No | Yes (Reddit + brand work) |
Schema implementation | Contact for details | No | Yes |
CRM integration | Custom integration | Custom integration | Custom UTM + mapping |
Contract flexibility | Contact for details | Contact for details | Month-to-month |
Evaluating in-house vs outsourced AI
For Series A to C B2B SaaS companies, a specialist agency with proprietary tooling is typically more capital-efficient than an in-house hire. The first-year cost difference is significant, and the ramp time to productivity delays pipeline impact. At Series D and above, a hybrid model often works best.
Why agencies struggle with AI citations
Most SEO agencies struggle with AEO because they treat citation optimization as link building. LLM passage retrieval selects content by semantic similarity between query and passage, not by backlink profile. An agency without in-house AI/ML engineering cannot build or validate the retrieval-aware content structures that drive citation rate lift. The AEO expertise evaluation guide covers the questions to ask any agency before committing a retainer.
Implementation timelines and pipeline measurement
Setting accurate timeline expectations is the most important thing any vendor in this space can do for a CMO preparing a board narrative.
Timeline for initial AI citations
Initial citation signals from new content typically appear within one to two weeks of publication, as LLM crawlers index the page and retrieval systems update their document stores. Meaningful citation rate lift across a priority query set takes three to four months of consistent content production, off-page consistency building, and schema implementation. This is a three to four month compounding effect, not a 30-day result.
Can Profound track attribution to closed revenue?
Profound cannot track closed-won revenue directly. The platform measures citation visibility and brand mentions, not conversion from those citations to pipeline stages in your CRM. As we documented in our AI tracking platform measurement flaw post, the invisible-read problem means many AI citation moments never generate a trackable session at all. Solving this requires custom UTM frameworks, Salesforce custom fields, and self-reported attribution from your demo forms, built alongside the software rather than delivered by it. That is the attribution stack Discovered Labs implements as part of onboarding.
We're an organic search agency for B2B SaaS, working across web search, citations, and training data, with a full-time AI/ML engineering team building the tooling that powers our audits, content operations, and knowledge graph. Pricing is public. Retainers are month-to-month. If you want a custom AI visibility audit that maps your current citation rate against competitors and identifies the 10 highest-value query gaps to close first, book a call and we'll tell you honestly whether we're a fit.
You can also run your existing content through our free AEO content evaluator to see how it scores against the CITABLE framework before committing to any engagement.
Conclusion
Profound gives B2B SaaS teams a diagnostic baseline: citation rates by engine, competitor share of voice by topic, and prompt-level visibility gaps mapped against a real-user query corpus. That diagnostic is valuable. What it doesn't deliver is the execution stack needed to close those gaps. Content production, off-page consistency building, schema implementation, and CRM attribution all sit outside the software. If your team has those functions covered internally, Profound is a strong tracking layer. If it doesn't, pairing the platform with a service-led execution partner gives you the measurement and the output working together.
FAQs
How much does Profound AI cost?
The Starter plan is $99/month (ChatGPT only, 50 prompts, 1 seat), the Growth plan at $399/month adds Perplexity and Google AI Overviews with expanded prompt volumes, and Enterprise plans covering 10-plus LLMs require custom pricing quotes.
Can Profound track which specific pages get cited?
Yes. Profound tracks both domain-level and individual page-level citations, showing you exactly which URLs LLMs reference when answering questions.
Does Profound integrate directly with Salesforce?
Profound's Salesforce integration requires custom setup. Connecting citation data to Salesforce pipeline fields typically requires API development or middleware platforms, plus custom Salesforce fields for AI-referred lead source tracking.
How often does Profound refresh its citation data?
Citation data refresh frequency varies by plan tier. For active content optimization programs where you want fast feedback on structural changes, understanding the update cadence helps set expectations for the optimization cycle.
How long before I see citation rate improvements?
Initial citation signals from new content typically appear within one to two weeks. Meaningful citation rate lift across a priority query set takes three to four months of consistent content production and off-page work.
Key terms glossary
AI visibility: The frequency and prominence with which a brand is cited or mentioned in LLM-generated responses across engines like ChatGPT, Claude, Perplexity, and Gemini.
Citation rate: The percentage of analyzed queries where an AI engine explicitly cites a specific brand or content source. A citation rate of 40% means your brand appears in 40 out of every 100 relevant queries tested.
Passage retrieval: The process where an LLM extracts semantically relevant blocks of text from the web to synthesize an answer. Dense passage retrieval, as demonstrated in Karpukhin et al.'s paper, selects by semantic similarity, not by keyword match or backlink count.
Information consistency: The alignment of product claims and brand facts across multiple independent web sources. LLMs use cross-source agreement to validate whether a claim is accurate enough to include in a generated answer.
Share of voice: The percentage of AI-cited responses in a defined query set where your brand appears, measured relative to competitors in the same category.
AI-referred pipeline: Revenue opportunities where the buyer used an AI assistant (ChatGPT, Claude, Perplexity, or Gemini) to research vendors before contacting sales, and your brand appeared in at least one AI-generated answer during that research. Also measured as ARR (annual recurring revenue) from AI-influenced deals.
