Scrunch AI review 2026: what Core acts on, and what waits for Enterprise
Scrunch AI reviewed on monitor, optimize, publish and measure: Core's real limits at $250, the Enterprise gates, user reviews, and where the fix still falls to you.
Scrunch built its product around one bet: the most important visitor to your website is now an AI agent, and that visitor deserves its own version of the site. It's a sharper thesis than most trackers start from. This site grades tools on a narrower question, though. Once Scrunch AI has found a gap, how much of the fix does it carry out, on a plan with a listed price, and can it tell you afterwards whether the fix worked? We read Scrunch's pricing page and homepage (captured November 6, 2026), its monitoring and agent traffic pages and llms.txt (read November 8, 2026), and its G2 reviews (November 15, 2026), then set every row against Promptwatch.
Quick verdict
Scrunch Core is a capable monitor with a real bot traffic view and a recommendations feed, and it stops before the page goes live. The parts of Scrunch that change your site, the Agent Experience Platform (AXP), the extra engines and every data connection, are Enterprise features without a published price. Promptwatch Professional costs $5 less than Core and adds publishing, conversions and verified crawler data.
We score each stage from 0 to 3. A 3 means the stage works in depth on a self-serve plan with a listed price.
| Stage | Scrunch Core ($250/mo) | Promptwatch Professional ($245/mo) | Promptwatch Business ($579/mo) |
|---|---|---|---|
| Monitor | 2 | 2 | 3 |
| Optimize | 2 | 3 | 3 |
| Publish | 1 | 2 | 3 |
| Measure | 2 | 3 | 3 |
| Total | 7 of 12 | 10 of 12 | 12 of 12 |
Core drops a Monitor point for four fixed engines in one country, two Publish points because no CMS step is listed, and a Measure point because its bot data has no link to citations or conversions. Promptwatch Professional drops a Monitor point because city tracking and sub-daily runs start on Business, and a Publish point because articles are written on demand rather than by the scheduled Content Agent.
What Scrunch is and who it's for
The homepage leads with the claim that humans no longer visit your website and AI agents do, and offers a free AI visibility audit with no credit card. The product menu lists Agent Pages, Agent Traffic, Site Optimization, Monitoring & Citations, Insights, Shopping and AI Search Trends. Scrunch's llms.txt now describes it as a Sitecore company.
Scrunch's homepage, captured November 6, 2026.
What it does well is easy to source. Core includes AI agent and bot traffic, split by model and by whether a bot is training, indexing or retrieving for a live prompt. Citation reports show top cited domains and URLs, citations grouped by owner type, and movement over time. Signals flag a change, explain why it matters and suggest a response. Core also has funnel stages, one persona, limited query fan-outs and basic content generation, and the company lists SOC 2 Type II.
It suits brands that see the agent-facing version of their site as the main lever, and enterprise teams, Sitecore customers especially, who will buy AXP on a contract. Teams that want the gap-to-page loop run for them on a self-serve plan will find Core hands the last steps to a person.
Pricing, plan by plan
Scrunch shows one self-serve plan for brands and one custom plan. A "For agencies" tab exists, but its content wasn't in our capture.
| Core | Enterprise | |
|---|---|---|
| Price | $250/mo (billing period not stated) | Custom |
| Prompts and responses | 125 unique prompts, 5,000 responses a month | Custom |
| Engines | ChatGPT, Perplexity, Google AI Overviews, Copilot | Those four plus Claude, Gemini, Meta AI, Google AI Mode, Grok |
| Countries, languages, personas | 1 each | Unlimited countries; custom languages and personas |
| Query fan-outs | Limited | Included |
| Site audits | 5 a month, 25-page site map | Full site audits |
| Page optimizations and content | 1 optimization a month; Basic generation | Custom; Advanced generation |
| AXP, Product Observability | Not included | Included |
| Looker Studio, query API, MCP | Not included | Included |
| Seats and sign-on | 5 licenses, Google SSO | Custom licenses, SAML and OIDC |
Scrunch's pricing page, captured November 6, 2026.
The costs that don't show on the price tag:
- The response cap binds first. Running all 125 prompts on all four engines every day needs 15,000 responses a month, three times what Core includes.
- More than half of Scrunch's engine list sits on the custom plan. If your buyers use Gemini or AI Mode, Core never sees those answers.
- Every data connection is Enterprise: Looker Studio, the query API, advanced API access and MCP.
- The FAQ asks what happens when you exceed plan limits and whether annual pricing exists, but our capture shows neither answer. Ask before you plan around overages.
- A sixth colleague means a sales conversation, since Core stops at five licenses.
Promptwatch publishes every self-serve price: Essential $95 a month, Professional $245, Business $579, Enterprise custom, and agency plans at $199, $399 and $799. All six self-serve plans include a 7-day free trial on monthly billing, the same length as Scrunch's, and unlimited seats. Professional, the closest match to Core, includes 150 prompts and 18,000 responses on two projects. Essential is cheaper but has no crawler logs and no AEO articles, so Core's bot view beats it at the entry price.
Features, one area at a time
Tracking and engines
Core tracks four engines in one country and one language, and Scrunch's pages don't state how often prompts run (llms.txt promises "a real-time read" without a frequency). Promptwatch lets every paid plan track any supported engine: the real interfaces of ChatGPT, AI Overviews, AI Mode, Gemini, Perplexity, Copilot, Alexa and Meta AI, plus Claude, DeepSeek, Mistral and Grok through their APIs. Runs are daily, or every 12, 6 or 1 hours on Business. It also tracks the coding agents Codex, Claude Code and OpenCode, which Scrunch doesn't list.
Prompt research
AI Search Trends gives approximate prompt volumes and trending topics, three topics on Core, with no difficulty score listed. Promptwatch's Prompt Volumes add difficulty and citation rate across AI search, Google and Bing on every plan, the fan-out view shows each model's exact sub-queries, and personas can be stacked and compared in Persona Insights.
Citations and sources
This is Scrunch's strongest monitoring area, and it's all on Core: top cited domains and URLs, a "Social" owner segment, citation movement, and a per-URL view of your own pages with the prompts citing them. The monitoring page also shows a third-party mentions feed without explaining it.
Scrunch's Monitoring & Citations page, captured November 8, 2026.
Promptwatch's Citation Trends sorts citations into 22 content types and 20 source types with a decay half-life per page, Page Tracker follows any URL including a competitor's, and Reddit, YouTube, LinkedIn and Instagram each get a separate report.
Crawler and traffic data
Agent Traffic on Core shows visits by model, bot type and page, recent requests, and a human versus bot split. What we couldn't find on Scrunch's pages is the step after the visit: whether the bot was real, whether the page it read got cited, why a crawl failed, and whether AI-referred visitors converted. Promptwatch answers the first three with Agent Analytics (from Professional) and the fourth with Visitor Analytics (every plan), and Search Insights adds Search Console and Bing data on every brand plan, which Scrunch doesn't list either.
Content and publishing
Core's "Basic" generation finds prompts no page answers and drafts on-brand content, and Page Optimizations allow one page a month. Scrunch's pricing and product pages (checked November 2026) don't list publishing to WordPress, Webflow, Framer or another CMS; llms.txt news titles mention SitecoreAI and Zapier workflows without describing a publishing step.
Promptwatch's Content Agents page, captured November 8, 2026. The product image is Promptwatch's own illustration.
Promptwatch writes AEO articles from Professional (5 a month, 10 on Business) and publishes them to WordPress through its plugin, to Webflow or to Framer, as a draft or live. On Business the Content Agent works from a calendar with a monthly budget and publish windows, through a review inbox or fully automated. Content Refresh proposes diff-style edits to existing pages, and publish status tracking follows each page from live URL to crawls, citations and AI traffic.
Reporting and integrations
Both Scrunch plans have a reporting dashboard and the Explorer BI layer; exec reporting, Looker Studio, the API and MCP are Enterprise only, and PDF or scheduled reports aren't listed. Promptwatch includes dashboard PDFs and 10 scheduled reports on every brand plan, Looker Studio from Professional, and the REST API and a hosted MCP server with 110+ tools on every paid plan.
Where it falls short
The fix ends at a draft
Insights, Signals and basic generation tell you what's missing and hand you text. Then someone formats it, pastes it into the CMS, adds internal links and remembers to check it a month later. For a team publishing weekly, that handoff is where work slips. Promptwatch removes it with CMS publishing from Professional and the scheduled Content Agent on Business.
Bot visits you can't turn into a task
A page bots never fetch needs an access fix. A page they fetch and skip needs a rewrite. Scrunch shows the visits but doesn't list the link to citations, or IP checks that rule out spoofed user agents, so you can't tell those two cases apart. Promptwatch's Agent Analytics (Professional, $245) gives each page a citation rate and verifies every bot against its operator's published ranges.
No line from answer to revenue
AI-referred conversions aren't described on Scrunch's pages, and a G2 reviewer quoted below reports trouble keeping the GA4 connection working. Promptwatch's Visitor Analytics measures AI-referred visits and conversions on every plan, and a UTM toggle keeps counting ChatGPT visits after ChatGPT stopped sending a referrer in October 2026.
Most engines behind a sales call
Gemini, AI Mode, Claude, Meta AI and Grok appear only in the Enterprise column, and a G2 reviewer says the tiering "wasn't made clear upfront". Promptwatch has no per-engine surcharge. Our caveat: it tracks Claude, Grok, DeepSeek and Mistral through their APIs, so you see what those models answer from training data rather than live search.
One country and no stated cadence
Core covers one country, and no frequency is stated. Promptwatch tracks any country on every paid plan, and Business adds state and city tracking, coordinate targeting from 20 m to 5 km, and runs every 12, 6 or 1 hours.
Nothing tests what AI claims about you
Scrunch scores sentiment by persona and topic, but claim-level narrative and fact checking aren't listed. Promptwatch's Brand Narrative (every plan) groups the claims AI makes by theme. Fact Check (Business, Agency Scale, Enterprise) tests a claim against your site and the 50 offsite pages AI cites most for it, and "Fix with content agent" writes one refresh article per flagged page.
What users say
Scrunch holds 4.6 out of 5 from 74 reviews on G2, where the listing now reads "By Sitecore" (checked November 15, 2026). Capterra showed no reviews.
The praise centres on the bot data. A senior website marketing manager at a mid-market company wrote on G2 (April 2026): "I've been really enjoying Scrunch's Agent Traffic monitoring tool as well as Site Maps. This really gives me insights into which pages are crawled by bots the most and for what reason (retrieval, indexing, training)." A growth marketing SVP wrote on G2 the same month: "We tested pretty much every AI Visibility measurement tool on the market and concluded that Scrunch has both the best price and presents the data in the most intuitive and actionable way"
The criticism is about integrations and plan terms. The same marketing manager added: "There has been some bugginess with the GA4 integration where I can't easily switch between our two properties. The Site Maps section also disconnects the GA4 integration and I have re-enable it often." A reviewer who rated it 3.5 out of 5 (April 2026), while calling the product solid, wrote that "certain models are locked behind higher tiers, which wasn't made clear upfront."
Two worked team scenarios
Example: a developer-tools company whose buyers ask Gemini and Claude
Assumptions (illustrative): 60 prompts in one country, daily on ChatGPT, Perplexity, Gemini and Claude. The docs site is heavy, so the team wants to see which bots read it. Articles are written by hand.
On Scrunch, Core fits the prompt count but only two of the four engines; ChatGPT and Perplexity daily use 3,600 of its 5,000 responses. Gemini and Claude need Enterprise at a custom price, and so does AXP, which a heavy docs site might want. On Promptwatch, 60 prompts is over Essential's 50, so the plan is Professional at $245: four engines daily use 7,200 of its 18,000 responses, with crawler logs and per-page citation rates included. Uncovered on Promptwatch: nothing serves agent-optimized pages, and Claude runs through its API rather than its live interface.
Example: a four-person content team at a travel booking site
Assumptions (illustrative): 120 prompts on ChatGPT, AI Overviews, AI Mode and Perplexity, one market, daily. The team publishes on Webflow about eight times a month and wants pages drafted from the gaps.
On Scrunch, 120 prompts fit Core's 125, but AI Mode is Enterprise only. On the other three engines a daily schedule needs 10,800 responses against Core's 5,000, so each prompt gets checked roughly every other day, and every draft reaches Webflow by hand. On Promptwatch, 14,400 responses fit Professional's 18,000 at $245, with five AEO articles a month published to Webflow. Eight pieces is more than that, so the team buys extra articles (an add-on we have no price for) or moves to Business at $579 for 10 articles and the scheduled Content Agent.
Promptwatch vs Scrunch AI
| Row | Promptwatch | Scrunch AI |
|---|---|---|
| Entry price and allowance | Essential $95: 50 prompts, 6,000 responses | Core $250: 125 prompts, 5,000 responses |
| Engines | Any supported engine on every paid plan | 4 on Core; 5 more on Enterprise only |
| Query fan-outs | Exact sub-queries per model | Limited on Core |
| Location | Country every plan; city and coordinates on Business | 1 country on Core |
| Fastest frequency | Hourly on Business | Not listed |
| Page-level citations | Citation Trends and Page Tracker | Yes, on Core |
| AI crawler logs | Professional and up, IP-verified | Yes, on Core |
| Crawl-to-citation and errors | Yes | Not listed |
| AI-referred conversions | Every plan | Not listed |
| Search Console and Bing | Every brand plan | Not listed |
| Fact check | Business and up | Not listed |
| Articles and CMS publishing | AEO articles from Professional; Content Agent on Business | Basic generation on Core; CMS publishing not listed |
| MCP and API | Every paid plan | Enterprise only |
| Seats and trial | Unlimited seats; 7-day trial | 5 licenses; 7-day trial |
Sources: Scrunch's pricing page and homepage captured November 6, 2026, its monitoring and agent traffic pages and llms.txt read November 8, 2026; Promptwatch's pricing page and documentation re-checked in November 2026. "Not listed" means we didn't find it on Scrunch's pages, not that Scrunch has ruled it out.
Promptwatch content editor, from the Promptwatch changelog. Example project data.
Which rows matter depends on what your team does after the report. Content teams should read the publishing row first: Scrunch drafts and stops, while Promptwatch drafts in an editor where each AI edit can be reverted, then publishes. Technical SEOs should read the crawler rows, where both tools show bots at a similar price and only Promptwatch ties visits to citations. Anyone reporting to a CFO needs the conversion row. Promptwatch's own weak spots: Essential has no crawler logs or AEO articles, Ads Radar and Shopping start on Business, agency plans have no Content Agent, and nothing in Promptwatch replaces AXP.
Who should stay, who should switch
- Stay on Scrunch AI if AXP is why you bought it and you're on Enterprise, or if ChatGPT, Perplexity, AI Overviews and Copilot cover your buyers and the bot traffic view is the report you use most.
- Stay if you run Sitecore and want your AI tooling from the same company.
- Switch to Promptwatch if your gaps keep ending as drafts in a shared doc.
- Switch if you need Gemini, Claude or AI Mode without an Enterprise contract, or more than five people in the tool.
- Switch if someone asks what AI answers bring in, since conversions aren't listed on Scrunch's pages.
Our verdict and next step
Scrunch has the most interesting idea on the agent side of the website, and most of it lives on Enterprise. Core watches well, drafts a little and leaves the rest to you. For most teams the deciding feature is Promptwatch's content loop: a brief built from a tracked prompt and the pages AI already cites, a draft pushed to your CMS, and Agent Analytics showing whether verified bots fetched it and engines cited it. Start the 7-day free trial on Professional with your Scrunch prompt list, publish one article from a gap you've been sitting on, and compare what each tool reports about it two weeks later. Our Promptwatch vs Scrunch AI comparison has the full table, and the Scrunch AI alternatives roundup scores five other options.
FAQ
How much does Scrunch AI cost?
Core is $250 a month for 125 prompts, 5,000 responses, four engines and five licenses; our capture doesn't state the billing period. Enterprise is custom.
Does Scrunch AI publish content to a CMS?
Not on the pages we read. Core includes basic content generation and one page optimization a month. Promptwatch publishes to WordPress, Webflow and Framer from Professional.
Does Scrunch track Gemini and Google AI Mode on Core?
No. Core lists ChatGPT, Perplexity, Google AI Overviews and Copilot; Gemini and AI Mode are in the Enterprise column.
Is Scrunch's bot traffic the same as Promptwatch's crawler logs?
They overlap. Both show which AI bots visit which pages, and Scrunch splits them by purpose on Core. Promptwatch, from Professional, also verifies each bot's IP, explains failed crawls and gives each page a citation rate.