Profound AI Review: Citation Tracking and Limits (August 2026)

If you're sizing up Profound for AI search citation tracking, here's the short answer: it does one thing really well and leaves everything else to you. This is a breakdown of how it works, where the data gets unreliable, and what the tool won't do — on any plan. My goal: you walk away knowing whether Profound fits your situation, or whether you need something that closes the loop between insight and execution.

TLDR:

• Profound tracks AI citation frequency across engines daily, but only monitors prompts you configure yourself
• Meaningful engine coverage starts at $399/month; Claude and Gemini require Enterprise pricing and a sales call
• Each configured prompt fans out into 8-12 sub-queries internally, so your citation data reflects one surface input, not the full retrieval picture
• SparkToro found no agreed-upon benchmark for AI citation consistency, so a single week of data isn't actionable signal
• Maintouch tracks citations across all five engines including Claude, pushes fixes through CMS integrations, and includes Claude coverage in its free AI visibility tier (35 prompts, one full year, no credit card) — AI visibility only; content creation and full-platform capabilities require a paid subscription

What Is Profound AI and What Does It Do

Profound is built to answer one question: when someone asks ChatGPT, Perplexity, or Google's AI Overviews about a topic in your space, does your brand show up?

It tracks citation frequency by running your configured prompts against AI engines and recording which sources appear in the response. The output is share-of-voice across AI answers: which prompts cite you, which cite competitors, and how often.

It also surfaces "Prompt Volumes" — proprietary data showing how often questions in your space are actually being asked across AI engines. No direct competitor matches that data, and it's the tool's clearest differentiator in the category.

That's where the product stops. It tells you where you stand. Acting on what you find is left to you.

How Profound Tracks AI Search Citations

Profound runs your configured prompts against AI engines daily and records what comes back. When a response includes your brand or a competitor, that citation gets logged. Over time you build frequency data: how often you appear, in which engines, against which prompts — and how that shifts week over week.

Under the hood it's simple: Profound sends the prompt, captures the response, parses it for brand mentions, and stores the result. Scheduled runs handle that automatically — no manual querying.

The catch: the prompt set is defined by you. What gets tracked is what you configured. If a prompt isn't on your list, there's no visibility into it — and that constraint matters more than it looks on a demo.

Which AI Engines Profound Covers (and What's Gated)

Engine coverage is gated by plan. Which tier you're on determines how much of the AI search landscape you actually see.

Claude and Gemini are Enterprise-only. If that's where your audience is asking questions, you won't see any of it without a sales call and a custom contract.

The Starter plan's single-engine view is a real blind spot. ChatGPT and Perplexity citations overlap just 11% of the time — a ChatGPT-only view misses the majority of what's happening across AI search.

Profound Pricing: What Each Tier Actually Includes

Starter runs $99/month billed annually: ChatGPT only, 50 prompts, 1,500 responses per month, one seat, one region, 100 Agent credits. No exports, no API access.

Growth is $399/month billed annually: ChatGPT, Perplexity, and Google AI Overviews, 100 prompts, 9,000 responses per month, three seats, 400 Agent credits, CSV/JSON exports, and a 7-day free trial. Still one language, one region.

Enterprise is custom pricing: up to 9 engines, broader prompt volumes, API access, SSO.

The real pricing cliff is engine coverage. You need $399/month to see anything beyond ChatGPT. Claude and Gemini aren't available at any self-serve price point, full stop.

Profound's Key Features Explained

Four areas make up the product.

Prompt tracking is the core. You configure prompts, Profound runs them daily across whichever engines your plan covers, and logs which sources appear in the responses. Citation frequency over time, broken down by prompt and engine.

Prompt Volumes is the standout. It shows estimated AI search demand by topic — how often questions in your space are actually being asked. No direct competitor matches this data, and it's the clearest reason to consider Profound over cheaper alternatives. It's the same intuition as keyword volume, applied to AI answers before you've written a word.

The Agent layer turns citation data into content briefs and drafts. It runs on a credit model (100 credits on Starter, 400 on Growth), so heavy use burns through your allowance fast. Output quality depends entirely on how well your prompts are configured — garbage in, garbage out.

Competitive benchmarking shows which competitors appear in the same AI responses and how your citation share compares against them. The limitation: it only reflects the prompts you've defined. If a competitor is getting cited on queries you haven't configured, you'll never see it.

How Query Fan-Out Affects What AI Visibility Tools Actually Measure

When you configure a prompt in Profound, you're tracking one input. But the AI engine doesn't run one search — it breaks your prompt into 8 to 12 parallel sub-queries, retrieves content for each, and assembles a single answer from all of them. A December 2025 study analyzing 173,902 URLs found that 68% of pages cited in AI Overviews weren't in the top 10 organic results for the original query.

In practice, your citation data reflects one surface-level prompt — not the sub-queries actually driving the answer. A competitor could be getting cited on a fan-out branch you never configured, and you'd never know.

The Reliability Problem: How Consistent Is AI Citation Data

Citation data from AI engines isn't as stable as it looks on a dashboard.

SparkToro found no agreed-upon benchmark for AI citation consistency. Ask the same prompt 100 times and you'll get meaningfully different answers — which makes it genuinely hard to know whether a citation score reflects real visibility or statistical noise from a few lucky runs.

Treat the first few weeks as data collection, not a readout. The signal is real, but it needs volume and time before it stabilizes into something actionable.

What Profound Does Not Cover

Profound is a monitoring tool. It surfaces data and stops there.

There's no CMS integration, so citation insights don't translate into shipped fixes. You see that a page isn't getting cited, but pushing a schema update, refreshing a stale post, or correcting a metadata mismatch all happen somewhere else — in a separate tool, on a separate timeline, coordinated by you.

No technical SEO execution layer. No backlink procurement. No schema health monitoring. The tool identifies the gap; closing it is your problem.

Engine access below Enterprise is narrow. Claude and Gemini aren't available at any self-serve price point, and Starter caps you at ChatGPT. If your buyers are using multiple AI engines — and most are — you're flying partially blind on anything under $399.

The Agent workflow builder skews toward agency configurations. Smaller teams typically need an external contractor to extract real value from it, which adds roughly $2,000/month on top of the subscription. This tool rewards teams with a dedicated AEO owner — it doesn't reward solo operators running lean.

Profound vs. Other AI Visibility Tracking Tools

The monitoring-only category is crowded. Where tools actually differ is engine coverage, prompt limits, and whether any insight ever becomes executed work.

Semrush and Ahrefs cover AI Overviews and stop there — ChatGPT, Claude, Perplexity, and Gemini citations aren't in scope. Otterly is the cheapest entry point but stays firmly in the monitoring lane. Profound has the strongest data depth at the top tier (Prompt Volumes, specifically), but meaningful engine coverage starts at $399 and Claude remains Enterprise-only regardless.

Every tool in that table except Maintouch hands you a dashboard and leaves remediation to you. That's the question that actually separates them: does it report, or does it execute?

One note on the free tier: Maintouch's free AI visibility tier tracks citations across all five engines but covers AI visibility only — content strategy, content creation, and technical execution require a paid subscription.

Who Should Use Profound and Who Should Look Elsewhere

Profound earns its reputation at the Enterprise tier. If your team has a dedicated AEO owner, a budget that absorbs $399+/month, and you need Prompt Volumes data to prioritize content investments at scale, it's a strong fit. Large marketing orgs with existing SEO infrastructure and the technical resources to act on citation data will get real value here.

The fit weakens fast outside that profile. Solo operators and lean startup teams typically need an external contractor to extract value from the Agent workflow — adding roughly $2,000/month on top of an already steep subscription. Starter caps you at ChatGPT only, 50 prompts. That's a narrow view for $99/month.

Teams whose buyers are active on Claude or Gemini have no self-serve path to that data. You're looking at Enterprise pricing and a sales call before you can see anything useful.

And if you need citation tracking, content creation, schema fixes, and backlink execution in one loop — Profound leaves that loop open. You'll see the gap. Closing it is a separate project.

How Maintouch Approaches AI Visibility Tracking Differently

Maintouch tracks citation share across all five engines: ChatGPT, Google Gemini, Google AI Overviews, Perplexity, and Claude. Claude is included in the free AI visibility tier at maintouch.com/free — 35 prompts, one full year, no credit card, AI visibility tracking only. Profound gates all of that behind Enterprise pricing and a sales call.

The structural difference isn't the dashboard. It's what happens after you see a gap. Maintouch pushes fixes directly through CMS integrations — schema updates, metadata corrections, content refreshes. The insight and the fix live in the same system. Profound hands you the data and stops. You coordinate the rest yourself.

Beyond the execution layer, every paying Maintouch account gets a dedicated account strategist — a forward-deployed marketer embedded in your growth motion, with weekly standing syncs and a dedicated Slack channel. Profound's standard support cadence is bi-weekly check-in calls. Teams migrating from Profound have reported needing to hire an external contractor (~$2,000/month on top of their subscription) to extract real value from the system. That contractor layer is what the dedicated strategist replaces.

If you want to see what that execution loop looks like in practice, shoot me a message at [email protected].

Final Thoughts on What Profound AI Does and Where It Stops

Profound earns its spot at the Enterprise tier for teams with dedicated AEO owners and the infrastructure to act on citation data. For everyone else, the engine coverage gaps and monitoring-only architecture are real constraints worth knowing before you sign.

I've watched enough teams buy a monitoring tool, stare at the citation numbers for three months, and then ask who's going to do the actual work. That moment — when the dashboard is full and the execution queue is still empty — is exactly what Maintouch is built to avoid.

If you want to see what it looks like when the insight and the fix live in the same system, shoot me a message at [email protected].

FAQ

What does Profound AI do, and what are its core limitations as an AI visibility tool?

Profound tracks how often your brand gets cited across AI engines like ChatGPT, Perplexity, and Google AI Overviews by running configured prompts daily and logging which sources appear in the responses. The core limitation is architectural: it's a monitoring tool that hands you citation data and stops there — no CMS integration, no schema fixes, no content execution. Meaningful engine coverage starts at $399/month, and Claude remains gated behind Enterprise pricing regardless of plan.

Profound AI visibility review: is it worth it for lean startup teams?

For lean teams, the fit is weak. The Agent workflow builder skews toward agency configurations, and most solo operators end up needing an external contractor (roughly $2,000/month on top of the subscription) to extract real value from it. The Starter plan caps you at ChatGPT only and 50 prompts, which is a narrow view for $99/month, and the Growth tier still excludes Claude and Gemini at the self-serve level.

How do fan-out queries affect what Profound AI search actually measures?

When you configure a prompt in Profound, you're tracking one input — but the AI engine breaks that prompt into 8 to 12 parallel sub-queries, retrieves content for each, and assembles a single answer from all of them. A December 2025 study of 173,902 URLs found that 68% of pages cited in AI Overviews weren't in the top 10 organic results for the original query. Your citation data reflects one surface-level prompt, not the sub-queries driving the actual answer — so a competitor could be getting cited on a fan-out branch you never configured, and you'd never see it.

What is the relationship between SEO and AEO, and does Profound AI search cover both?

SEO and AEO share the same foundation: AI chatbots run a web search under the hood before generating answers, so strong SEO directly feeds AEO performance. Profound is a citation monitoring tool, not an SEO execution system, so it covers the AEO measurement layer but has no mechanism to fix the underlying SEO signals driving whether your content gets retrieved in the first place. Maintouch treats SEO and AEO as one unified workflow, tracking citations across all five engines while also pushing schema updates, content refreshes, and metadata corrections through direct CMS integration.

How does Maintouch AI visibility tracking compare to Profound for teams that need Claude and Gemini coverage?

Maintouch tracks citation share across ChatGPT, Google Gemini, Google AI Overviews, Perplexity, and Claude, with Claude coverage included in the free tier at maintouch.com/free, which tracks 35 prompts across all five engines for one full year with no credit card required. Profound gates Claude behind Enterprise pricing and a sales conversation, and Gemini is similarly restricted to the top tier. Beyond coverage, the structural difference is execution: Maintouch pushes fixes directly through CMS integrations when a citation gap is identified, while Profound surfaces the data and leaves remediation to you.

How long does it take to get reliable citation data out of Profound?

More than a week. SparkToro found no agreed-upon benchmark for AI citation consistency, meaning a single week of data reflects statistical noise as much as real visibility. You need enough prompt runs and enough time for the signal to stabilize before you can confidently act on it. Treat the first few weeks as data collection, not a readout.

Does Profound let you track what your competitors are being cited for?

Yes, competitive benchmarking is built in. Profound shows which competitors appear in the same AI-generated responses as your brand and how your citation share compares against them. The limitation is scope: it only reflects the prompts you've configured yourself, so if a competitor is getting cited on queries you haven't added to your list, that activity is invisible to you.

What is Profound's Prompt Volumes feature and how do you actually use it?

Prompt Volumes is Profound's proprietary estimate of how often specific questions are being asked across AI engines. It's the closest thing the category has to keyword volume data for AI search. In practice, you use it to prioritize which topics to target before creating content, the same way you'd use search volume to size organic opportunity. No direct competitor has matched this data, which is its main differentiator in the monitoring-only category.

Does Profound have a free trial or a free plan?

The Growth plan ($399/month billed annually) includes a 7-day free trial. There's no permanent free tier. Maintouch offers a free AI visibility tracking tier at maintouch.com/free that tracks 35 prompts across all five engines for one full year with no credit card required, including Claude coverage that Profound gates behind Enterprise pricing.

What are Profound Agent credits and how quickly do they run out?

Agent credits power Profound's content brief and draft generation layer. The Starter plan includes 100 credits and Growth includes 400. Heavy use burns through the allowance fast, and once you're out, that workflow stops until the next billing cycle. The output quality also depends on how well your prompts are configured and how much context you've fed the system, so credits spent on poorly configured prompts aren't recoverable.

Does Profound track Google AI Overviews specifically, or just ChatGPT and Perplexity?

Google AI Overviews is included starting at the Growth tier ($399/month). The Starter plan ($99/month) covers ChatGPT only, so you'd need to upgrade to see AI Overviews data. Enterprise adds up to nine engines total, though the specific lineup beyond the core three requires a sales conversation to confirm.

What does it actually mean when Profound says your brand was "cited" in an AI response?

A citation in Profound's data means your brand name or domain appeared in the AI-generated response when a configured prompt was run. It's a mention count, not a ranking position. The AI engine referenced your content (or your name) somewhere in its answer, which could mean a direct source link, an inline mention, or a recommendation. Profound logs the frequency of those appearances over time, broken down by prompt and engine.

Does Profound integrate with any CMS, or does it stay separate from your publishing workflow?

Profound has no CMS integration. It's a standalone monitoring and analytics product — citation data lives inside the Profound dashboard and doesn't connect to wherever you publish or manage content. Any fixes you decide to make — schema updates, metadata changes, content refreshes — happen in a separate tool on a separate timeline. That gap is the core architectural limitation for teams that need insight and execution in one place.

What's the difference between AI citation tracking and traditional SEO rank tracking?

Traditional rank tracking tells you where your page shows up in Google's blue-link results for a given keyword. AI citation tracking tells you whether your brand or content gets mentioned inside an AI-generated answer — a fundamentally different signal. A page can rank on page one in Google and never appear in a ChatGPT or Perplexity response, and vice versa. Profound tracks the AI layer; it doesn't replace your rank tracker or tell you anything about your organic search positions.

If my citation rate drops in Profound, what should I actually do about it?

Profound will surface that your citation rate dropped, but it won't tell you which content or schema issues caused it — and it won't push any fixes. The typical remediation sequence: check whether the content being cited has gone stale (AI engines heavily favor pages updated in the last 90 days), audit your structured data for schema drift, and review whether a competitor recently updated content on the same topic. All of that happens outside Profound, in whatever content and technical SEO tools you use separately.

How does Profound compare to tools like AthenaHQ or Otterly for AI visibility tracking?

AthenaHQ is priced around $400/month and sits in the same monitoring-only category as Profound — it tracks AI visibility but doesn't execute fixes or push changes to your CMS. Otterly is the cheapest entry point in the space at $29/month and covers multiple engines, but it also stays firmly in the reporting lane. None of these tools close the loop between insight and shipped work. Maintouch tracks citations across all five engines including Claude, starting with a free tier at maintouch.com/free, and pushes fixes directly through CMS integrations rather than handing you a dashboard and stopping there.

Do you need a technical SEO background to get value out of Profound?

Not to read the dashboard, but yes to act on what it shows you. Profound surfaces citation data clearly enough that a non-technical marketer can understand the numbers. The technical requirement kicks in when you try to remediate: updating schema markup, correcting structured data mismatches, and pushing metadata changes all require either technical SEO knowledge or a developer. If your team doesn't have that capability in-house, the dashboard gives you a detailed readout of a problem you don't yet have the tools to solve.