Key takeaways
- Sitecore acquired Scrunch on June 3, 2026 for a reported $225 million. Scrunch's founder said the roadmap would accelerate, but six months of observable signals suggest the opposite for self-serve customers.
- The clearest evidence isn't missing features. It's packaging churn: Scrunch cycled through Starter, Growth, and Core plan labels in 2026, and as late as July 16, 2026 was serving two different pricing structures to different visitors simultaneously.
- This category is unusually vulnerable to vendor slowdowns. AI search platforms change their citation behavior overnight, so a visibility tool that refreshes slowly or deprioritizes data infrastructure quietly becomes wrong, not just stale.
- If your 2026-2027 roadmap assumes Scrunch ships specific features on schedule, treat that assumption like a pinned library version: named, reviewed quarterly, and replaceable.
The short version of what happened
Scrunch had a good run. Founded in 2023 in Salt Lake City by Chris Andrew (previously the first employee and CPO at Hearsay Systems), it raised a $4M seed in March 2025, a $15M Series A led by Decibel in July 2025, and grew past 500 brand customers by mid-2026, including Lenovo, Skims, Crunchbase, Penn State, ADP, and agencies like Publicis and Seer Interactive.
Then, on June 3, 2026, Sitecore acquired the company for a reported $225 million. Bloomberg reported the figure; Sitecore itself never confirmed terms.
In the acquisition announcement, Andrew wrote that Scrunch isn't going away, that it would continue operating as a standalone platform, that the product roadmap would accelerate, and that "the mission holds, the pace accelerates." I don't doubt he meant it. Acquirers say this every time, and sometimes it's even true.
But the observable evidence since June tells a more complicated story, and if your content roadmap or client deliverables depend on Scrunch shipping things on schedule, you should read the rest of this carefully.
What a slowing cadence actually looks like
Here's the thing about post-acquisition slowdowns: they almost never look like silence. They look like churn. Scrunch is a textbook case.
The packaging keeps moving
Through 2026, Scrunch cycled through Starter, Growth, and Core plan labels. As late as July 16, 2026, two different pricing structures were live at the same time for different visitors: a Starter/Growth ladder covering seven engines, and a single "Core" plan at $250/month covering four engines with the Agent Experience Platform gated to Enterprise.
When a company is shipping quickly, pricing tends to stabilize. When a company's packaging mutates every quarter, it usually means the product organization is being re-planned around someone else's priorities, in this case Sitecore's enterprise deal motion. That's a strong signal about where the roadmap is going.
The feature cadence pivots from shipping to repositioning
Scrunch did ship real things in 2026. The Q2 digest announced Scrunch MCP, Knowledge Studio, and Sentiment Trends, and later posts referenced Site Diagnostics, a Keyword Volume Estimator, and Prompt Templates. Credit where due.
But the most telling move came on July 8, 2026, five weeks after the acquisition: Scrunch repositioned itself from an "AI visibility" tool to an "Agent Experience Platform (AXP)," complete with new sub-products (Signals, Explorer) and a new category claim from the CEO. Sitecore then folded Scrunch's citation data into "SitecoreAI" and its Marketing IQ launch, wiring it into the DAM, CMS, Page Builder, and Design Studio.
A category pivot mid-integration isn't a feature release. It's a strategy document. And if you're a self-serve customer at $250/month, the strategic center of gravity just moved toward Sitecore's DXP stack, where Scrunch's data becomes an input to enterprise workflows rather than a standalone product competing for your renewals.
Small cracks that matter
A few more data points from reviews and third-party analysis:
- Core content generation was listed as "coming soon" on Scrunch's pricing page as of mid-2026. "Coming soon" features on a pricing page are roadmap promises, and roadmap promises from a company mid-acquisition deserve extra scrutiny.
- G2 reviews (73 reviews, 4.6 rating) praise the dashboard's ease of use but flag a "complete lack of visualizations for AI visibility trends," with at least one user manually rebuilding reports in Excel. Exporting is awkward on lower tiers; Looker Studio access only kicks in at Agency/Enterprise levels.
- Scrunch tracks a modeled set of prompts on a refresh schedule rather than capturing real user sessions, which means its trend data can lag what buyers actually see in the engines.
- GEO Toolbox, a competitor, raised the question everyone in this market is asking: whether Scrunch keeps selling self-serve at its current price, or gets folded into enterprise Sitecore deals over time.
Individually, none of these is damning. Together, they sketch a company whose product energy is being redirected.
Scrunch isn't alone, and that's the point
If this were a one-off, you could shrug it off as bad luck. It isn't. The whole GEO and AI visibility market went through a consolidation wave in 2026:
| Event | Date | What happened to the acquired product |
|---|---|---|
| Adobe completes Semrush acquisition | April 28, 2026 | Semrush's AI Visibility Toolkit folded into Adobe Experience Manager, the Adobe LLM Optimizer, and "Adobe Brand Visibility" |
| HubSpot launches HubSpot AEO | April 14, 2026 | Built on XFunnel, which HubSpot acquired in October 2025 |
| Sitecore acquires Scrunch | June 3, 2026 | Folded into SitecoreAI and Marketing IQ; self-serve tier's future an open question |
Scrunch's own Q2 2026 industry digest called this out almost a little too honestly: "when companies like Adobe, Sitecore, and HubSpot are making major moves, you know a market has truly arrived... Expect more martech consolidation ahead. But it's important not to confuse mergers and acquisitions with product capabilities. Buying a monitoring dashboard is easy. Everything that comes after — site auditing, content optimization, content delivery — is what ultimately generates results."
That was Scrunch describing its own fate before the ink was dry on the acquisition. I couldn't have put it better myself.
Why this category can't afford a slow vendor
Here's the part that makes vendor cadence in GEO tools a bigger deal than in, say, project management software. The platforms these tools monitor change their behavior overnight, without warning, and in ways that invalidate your data collection.
Three examples from Promptwatch Data, all from 2026:
- When GPT-5.3 rolled out on March 4, 2026, average citations per ChatGPT web-search response dropped roughly 27% (from about 6.4 to 4.7-4.9), across all model variants, in a single day. Any visibility tool that didn't immediately flag "this is a platform change, not an on-page problem" left its users chasing phantom causes.
- ChatGPT's query fan-out behavior collapsed from 2.15 searches per response in early December to exactly 1.0 by April, with average query length falling from ~117 characters to ~53. Your prompt-tracking methodology from Q4 2025 was simply wrong by Q2 2026.
- Microsoft Copilot swung from under 2 to nearly 17 sources per response within weeks, evidence that Microsoft is still actively re-architecting its retrieval. And on August 8, 2026, ChatGPT Search started using the
site:operator at scale, jumping from ~0.4% to ~17% of fanout queries overnight.
The lesson: in this category, a vendor that slows down doesn't just become less innovative. It becomes incorrect. A monitoring platform with last quarter's crawling logic, last quarter's citation classification, and a modeled prompt set on a refresh schedule will quietly misreport your visibility while the real engines move on. If your vendor's engineering attention is being spent on enterprise integration plumbing rather than data infrastructure, you absorb that error directly into your reporting.
This is exactly why I'd argue for platforms whose core business is the data layer itself. Promptwatch, for example, monitors the actual user interfaces of ChatGPT, Gemini, AI Overviews, AI Mode, Perplexity, and Claude (not just their APIs, which can diverge from what users see), and its crawler logs track 400+ AI bots in real time, which is how the platform-change events above get caught within a day instead of a quarter. Tools like this can also help you sanity-check whether a sudden visibility dip is your problem or the platform's.
The two clocks problem
There's a structural issue underneath all of this that's worth naming. In a widely shared essay from May 2026, the