Profound vs Otterly vs Scrunch AI: citation tools compared

Profound, Otterly, and Scrunch AI all sell the same basic promise: tell me when ChatGPT, Perplexity, or Gemini mention my brand. They do not sell the same product. Profound is built for enterprise depth, Otterly for a solo founder who wants a number today, and Scrunch for an agency-grade alert stack now bolted onto a much bigger company. Picking between them is really picking a budget and an org shape, so that's how this comparison is built.

The short answer on Profound vs Otterly vs Scrunch AI

Profound is the enterprise-grade option: the widest engine reach at the top tier, SSO and SOC 2 for procurement, and a self-serve floor that starts at $99/month for ChatGPT-only tracking. Otterly is the cheapest real entry point, $29/month for four engines with a 14-day free trial and no card required, which makes it the natural pick for a solo founder or a two-person agency testing whether AI citation is worth watching at all. Scrunch AI sits in the middle on price but leans hardest into automated alerting through a feature called Signals, and its Influence Score is a genuinely different way to score a mention. Scrunch was also just acquired by Sitecore, which changes the calculus for anyone signing a long contract. None of the three will make you more citable. That's the honest hand-off point at the end of this post, and it's worth reading before you buy any of them.

What each platform actually measures

All three run your prompts across AI engines on a schedule and score how often you show up. The differences are in which engines, at what depth, and what happens after the mention is logged.

Profound brings broad engine reach and enterprise-grade share of voice

Profound's self-serve pricing starts at Starter, $99/month for one seat, 50 tracked prompts, and ChatGPT only. Growth, its most popular self-serve tier at $399/month, expands to three seats, 100 prompts, and three engines (ChatGPT, Perplexity, and Google AI Overviews). Enterprise is custom and is where the platform actually earns its "broad reach" reputation: up to 10 answer engines, multiple workspaces, SSO/SAML and SOC 2 compliance, API access, and dedicated Slack support with a 24-hour SLA.

Profound doesn't publish an Enterprise number, but third-party reviewers who've priced it out land in the same neighborhood. Vismore's review puts real Enterprise contracts at roughly $2,000 to $5,000-plus a month, driven mostly by how many of the 10 engines you turn on, how many seats you need, and whether you want the Prompt Volumes panel that analyzes over 10 million daily prompts. Treat that as a third-party estimate, not a quote, and confirm current numbers with sales before you budget against it.

The company backs up the "enterprise" positioning with its cap table. Profound raised a $96 million Series C in February 2026 at a $1 billion valuation, led by Lightspeed with Sequoia, Kleiner Perkins, and others returning, bringing total funding to $155 million. The company says it now serves more than 700 enterprise customers, including roughly 10% of the Fortune 500. Co-founder and CTO Dylan Babbs put the underlying thesis plainly in an interview about Profound's product strategy: "every company on the planet just got a new customer: AI." That's the same reasoning that has enterprise buyers treating answer-engine visibility as a line item worth an SSO-gated procurement process rather than a side project.

Otterly is the low-cost entry point for solo teams and small agencies

Otterly answers a narrower question well: is anyone even mentioning us in AI search, and can I find out without committing a budget line to it. Its live pricing is Lite at $29/month ($25 billed annually), Standard at $189/month ($160 annually), and Premium at $489/month ($422 annually). All three tiers track ChatGPT, Perplexity, Google AI Overviews, and Microsoft Copilot by default, with Gemini and Google AI Mode sold as add-ons across every tier, not gated to the top plan. That's a meaningfully different structure than Profound's, where the number of engines is itself the thing you pay to unlock.

The catch sits in what Lite leaves out rather than what it tracks. Lite gets you 15 prompts and daily tracking, but API and MCP server access only unlock at Standard ($189/month) and above, so a founder who wants to pull citation data into their own dashboard needs to jump two tiers past the headline price. Otterly also runs a 14-day free trial with no payment details required, the lowest-friction way to try any of these three tools before paying anything.

The company's own story matches the price point. Otterly is bootstrapped, founded by Thomas Peham, Josef Trauner, and Klaus-M. Schremser, and exited stealth in December 2024 with more than 1,000 users after launching that October. Peham described the founding thesis this way: "the shift to generative AI search is one of the most seismic transformations in marketing today. We started Otterly.AI with a vision to empower brands to adapt to the fundamental changes happening in the search ecosystem." No outside war chest, no enterprise sales motion, just a product priced for a team that wants an answer this week.

Scrunch AI's Influence Score, Signals alerts, and the hallucination-detection question

Scrunch's pricing runs Starter at $300/month ($250 billed annually) with three seats, 350 custom prompts, and 1,000 industry prompts, and Growth at $500/month ($417 annually) with five seats and roughly double the prompt volume. Enterprise is custom. Coverage spans ChatGPT, Claude, Gemini, Perplexity, Google AI Mode and AI Overviews, and Meta, though Scrunch's pricing page doesn't break out exactly which engines land on which tier, so confirm the tier-by-tier list directly with sales rather than assuming Starter gets everything Enterprise does.

What differentiates Scrunch is less the engine list and more what it does with a mention once it's logged. Its Influence Score, per an independent review at Trakkr, "goes beyond mention counting to assess how prominently your brand appears, whether it is cited as a recommendation, and how much weight AI models give your content," rolling three signals, prominence, recommendation status, and model weighting, into one number meant for trend tracking and executive reporting rather than a raw citation tally. Signals, a feature Scrunch describes on its own blog, continuously assesses how AI agents represent a brand and flags threshold breaches, trend shifts, anomalies, and emerging competitors, then recommends a next action, delivered inside the app, by email, or through other channels. It's the closest thing among the three tools to an alert system that taps you on the shoulder instead of waiting for you to open a dashboard.

Here's the part worth being careful about. Several third-party review sites describe Scrunch as offering dedicated hallucination detection, catching when a model states the wrong price, invents a partnership, or otherwise fabricates a claim about your brand. Scrunch's own site does not market a feature under that name. Its product pages list "Insights" as one of three components, alongside Monitoring and the Agent Experience Platform (AXP), and describe Insights as offering optimization guidance, citation analysis, and "error detection to spot when AI bots can't crawl your site." That's crawlability error detection, not a check on whether a model's answer about your brand is factually wrong. Scrunch's pricing page separately promises "Page Audits" and "Insights to spot opportunities and content gaps" on every tier, but neither is described as hallucination detection in Scrunch's own copy. The honest answer is that the capability may exist in some form, but it isn't a named, independently confirmable feature on Scrunch's own marketing pages as of this writing. If catching brand-hallucinations specifically is why you're buying, ask Scrunch to demo it before you sign, rather than taking a third party's word for what the product does.

One more fact worth knowing before you evaluate Scrunch as a standalone vendor: it isn't fully standalone anymore. Sitecore announced on June 3, 2026 that it had acquired Scrunch, a deal Bloomberg reported at roughly $225 million. Scrunch reports more than 500 brands and agencies on the platform, and CEO Chris Andrew addressed the deal directly in a post on Scrunch's own blog: "that mission hasn't changed. But our timeline to execute it just got a lot faster." Scrunch's pricing page and self-serve signup were both still live when we checked. But an agency locking in a long-term contract right after an acquisition should confirm current terms directly, since roadmaps and packaging both tend to shift once a platform gets absorbed into a larger company's suite.

Pricing and team fit for startups, agencies, and enterprise

The honest way to choose isn't "which tool scores highest," it's "which org shape am I." A five-person startup, a ten-person agency running this for clients, and a marketing team inside a public company are solving genuinely different problems with the same category of tool.

A solo founder or two-person startup testing whether AI search visibility is worth tracking at all should start with Otterly's $29 Lite plan or its 14-day trial. You get four engines and daily tracking for less than most teams spend on coffee, with no commitment if the data says AI referral isn't material yet.

An agency running this across client accounts needs seats, prompt volume, and something that doesn't require a human to check a dashboard every morning. Scrunch's Signals alerting and multi-seat Starter and Growth plans fit that workflow, provided the Sitecore acquisition doesn't complicate the vendor relationship for a shop juggling several clients on one contract.

An enterprise marketing org buying with a procurement checklist, SSO, SOC 2, dedicated support, a named account team, is Profound's actual market. The $399 Growth tier is really a trial run for teams that already expect to land on a custom Enterprise deal once they need more than three engines or want the Prompt Volumes data.

Profound vs Otterly vs Scrunch AI scorecard

Prices and plan details above are current as of this writing and change often, especially for Scrunch post-acquisition, so confirm them on each vendor's pricing page before you buy.

Where citation tracking stops and content strategy has to start

Every one of these three tools answers the same question: where do I stand right now. None of them answers the harder one: how do I stand somewhere better. A tracker can tell you that a competitor got cited for "best project management tool for remote teams" and you didn't, with a trend line proving it's gotten worse over the last quarter. It cannot rewrite your page so the model prefers you next time. That gap between measurement and content is the same split we mapped out across the broader category in our roundup of AEO platforms: tools split cleanly into "tells you the score" and "changes the score," and a citation tracker, however good its alerting, is permanently in the first bucket.

That distinction matters more than the pricing table above, honestly. A team that buys Profound's Enterprise tier and points it at the same thin, unverified, poorly structured content it had before will get a very precise, very expensive report confirming that it still isn't cited. The fix was never going to come from the tracker. If you want the fuller measurement picture, including the free GA4 setup that catches some AI referral traffic before you pay for any tracker at all, we cover that ground in AI citation tracking. This post is the deep comparison of the three paid tools; that one is the free-tier-first methodology, and they're built to be read together rather than as competing answers to the same question.

Where Lyra fits

Lyra is a content tool, not a tracker, and that split is deliberate. Pair her with whichever of these three matches your team's shape and budget, and you have the full loop: a tracker that tells you where you stand, and a writer that actually moves it. She researches the topics your buyers ask about, writes the post in your blog's existing voice, verifies every fact and link against a live source the way this comparison itself was built, and structures each page the way answer engine optimization rewards: the answer stated early, headings shaped like real questions, and citations that hold up when a model or a human checks them. Then she opens the post as a pull request you review and merge. Nothing publishes on its own.

We built her after running this loop by hand for years, a story we tell in why we built Lyra. If you're already paying for Profound, Otterly, or Scrunch and the score isn't moving, the content is very likely the reason, and that's the half of the problem a citation tracker was never going to solve for you. You can try Lyra directly, or see the plans on the pricing page if you want the numbers before you talk to anyone.

A tracker can tell you a competitor got cited instead of you. It can't write the page that fixes it. Lyra writes the posts built to earn the citation, verified and opened as a PR you merge.

Try Lyra → · Talk to the founder