Key takeaways
- Brandlight is a sales-led, enterprise-only platform with no public pricing (third-party estimates range from roughly $199/mo entry to $4,000-$25,000+/mo for enterprise contracts); it pairs measurement with a strategy team but its own materials admit a dedicated attribution module is still "coming soon."
- Scrunch was acquired by Sitecore in June 2026 and now sells a $250/mo Core plan (4 engines, 125 prompts, 25-page site audit) alongside a custom Enterprise tier with 9 engines and its Agent Experience Platform (AXP).
- Promptwatch publishes self-serve pricing from $95/mo, tracks 8 models including Meta and DeepSeek, and is the only one of the three built around closing the loop: crawler logs, content gap analysis, and Content Agents that publish directly to your CMS.
- Citation slots are scarce and getting scarcer: ChatGPT typically surfaces around 5 sources per response versus roughly 10 for Google AI Overviews and Perplexity, according to Promptwatch's citation research, which raises the stakes of whatever content-gap tooling each platform offers.
- None of the three can guarantee a citation. The real decision is whether you want a platform-plus-service model (Brandlight), a technical-agent-delivery specialist now owned by a CMS vendor (Scrunch), or a self-serve platform that also ships the fix (Promptwatch).
Why this comparison matters right now
Enterprise marketing teams are past the "should we monitor AI search" question. They're now stuck evaluating three very different philosophies dressed up in similar dashboards. Brandlight sells itself as a strategy partner with a platform attached. Scrunch, now owned by Sitecore, leans into AI-agent content delivery and technical diagnostics. Promptwatch tries to do the whole thing itself, from tracking to publishing, without requiring a retainer.
I've spent time with all three vendors' public materials, pricing pages, and the comparison pages they publish about each other (which, fair warning, are never neutral). What follows tries to separate the marketing from what you'd actually get for your money.
The one-line version of each
Brandlight is an enterprise-only "AI visibility operating system" sold through sales calls, aimed at Fortune 500 CMOs managing multiple brands and markets. Scrunch is a technical AI-experience platform, recently absorbed into Sitecore, built around getting your content readable and traceable by AI agents. Promptwatch is a self-serve platform that tracks prompts across eight models and then tries to fix what it finds, using crawler logs, content gap analysis, and automated publishing.
Pricing: the biggest practical difference
This is where the three platforms diverge hardest, and it's worth being blunt about it.
Brandlight has no public pricing page. Brandlight.ai/pricing returns a 404. Third-party estimates (not confirmed by Brandlight) put entry pricing around $199/month, with enterprise contracts reportedly running anywhere from $4,000 to $25,000+ per month. Engagements typically start with a paid three-month pilot, not a free trial. If you need a number to put in a budget spreadsheet before a sales call, Brandlight isn't going to give you one.
Scrunch publishes real numbers, at least for its Core tier: $250/mo gets you 4 LLMs (ChatGPT, Perplexity, Google AI Overviews, Copilot), 125 unique prompts, 5,000 responses/month, 5 site audits/month, a 25-page sitemap cap, and 5 user licenses. Its Enterprise tier expands to 9 engines (adding Claude, Gemini, Meta AI, Google AI Mode, and Grok) plus full query fan-out, the AXP agent-experience layer, and SSO, but that tier is custom-quoted. Worth flagging: Scrunch's pricing has reportedly shifted more than once in 2026, and the company was acquired by Sitecore in June 2026, so a "standalone AI visibility startup" pitch no longer quite applies. Future roadmap decisions may now answer to Sitecore's CMS bundling priorities rather than pure GEO demand.
Promptwatch publishes three tiers: Essential at $95/mo (1 project, 50 prompts, 6,000 responses/month, all 8 models, MCP and API access), Professional at $245/mo (150 prompts, 18,000 responses, crawler logs, 5 AEO articles/month), and Business at $579/mo (350 prompts, 42,000 responses, ChatGPT Shopping insights, Ads Radar, city-level tracking). No engine is locked behind a higher tier, which isn't true of Scrunch, where Claude and Gemini only show up once you're paying enterprise rates.
| Platform | Entry price | Pricing model | Engines at entry | Free trial |
|---|---|---|---|---|
| Brandlight | Unpublished (~$199/mo est.) | Sales-led, custom quote | 7 named (per-market adapted) | No, paid pilot instead |
| Scrunch (Sitecore) | $250/mo (Core) | Self-serve + custom Enterprise | 4 (ChatGPT, Perplexity, AI Overviews, Copilot) | 7 days, no card |
| Promptwatch | $95/mo (Essential) | Fully self-serve | 8 (incl. Meta, DeepSeek, Grok) | Free Explore tier |
What each platform actually does
Brandlight: measurement plus a strategy team
Brandlight's pitch is that it runs a full loop it calls See, Shape, Act, Transform, spanning owned, third-party, social, retail, and emerging agentic-commerce surfaces. Its content module routes drafts through a strategist-writer-editor-validator chain with deterministic brand and legal rules, then hands them to your CMS for human review. It does not auto-publish, by design.
The proof point Brandlight leans on hardest is Kimberly-Clark: 12 of 12 brands moved into the top three across major LLMs, with a reported 100% paid-pilot-to-annual conversion rate. That's a genuinely strong case study for a CPG portfolio. But the fine print matters too. Brandlight's own documentation says its dedicated attribution module is still "coming soon," and that its AI Share of Voice, Sentiment, and Narrative Consistency scores are proxy metrics, not proof of revenue impact. Brandlight is also upfront that no tool can force a citation; its influence model works indirectly, by improving the third-party signals (review profiles, schema, content freshness) that AI models pull from.
Brandlight named partnerships with Demand Spring and Data Axle reinforce that this is platform-plus-service. You're generally not running this alone; you're running it with an agency or strategist layered on top, which is reflected in the pricing opacity and the multi-week implementation timelines reviewers describe (2-4 weeks standard, 6-8 weeks enterprise, up to 30-90 days to "full productivity" with governance features switched on).
Scrunch: technical depth, now under a CMS parent
Scrunch organizes its platform around three pillars: Agent Experience (Agent Pages, Agent Traffic, Site Optimization), Monitoring & Insights (Citations, Insights, Shopping), and AI Search Trends. The emphasis on AI crawler behavior and machine-readable content delivery is genuinely its strongest differentiator, not just marketing copy. If your biggest unknown is "can AI agents actually parse our product pages correctly," Scrunch's diagnostics are built for exactly that question.
The Sitecore acquisition is the detail most comparison guides miss because it happened mid-year. Sitecore now positions Scrunch as "an AI customer experience platform that helps brands understand and improve how they appear in AI search." That's a fine description, but it also means Scrunch's roadmap now sits inside a CMS vendor's broader commercial strategy, which is worth asking about directly if you're signing a multi-year contract.
On content execution, Scrunch is comparatively thin. Independent comparisons (admittedly competitor-authored, so read with a grain of salt) note it lacks content generation beyond basic recommendations, has no ChatGPT Shopping support, and no query fan-out tracking outside its Enterprise tier. Its site audit is capped at 25 pages on the entry plan, which is a real constraint for anything beyond a small marketing site.
Promptwatch: self-serve monitoring that tries to finish the job
Promptwatch's pitch, stated plainly on its own comparison pages, is "insights don't rank, published content does." The platform tracks prompts on real AI interfaces (not just API simulations), matches that to AI crawler logs through its Agent Analytics feature, and runs a content gap index built from a full crawl of your own site, not a capped 25-page sample. When a visibility drop shows up, the workflow is designed to point at a specific page, not just a chart.
The execution piece is Content Agents: they draft and can publish directly to Webflow, Framer, or WordPress on a schedule, either through a review inbox or fully automated. That's a meaningfully different posture than Brandlight's human-review-only content module or Scrunch's recommendation-only approach. Promptwatch also covers ground the other two skip almost entirely: dedicated Reddit and YouTube citation reporting, ChatGPT Shopping and Ads Radar tracking, and prompt-level difficulty and volume scoring.
The tradeoff is scale of brand. Promptwatch's customer list (Yelp, Duolingo, Rabobank, Shutterstock, Fireflies, Monks) skews toward mid-market and digitally native enterprises rather than the Fortune 500 CPG and healthcare accounts Brandlight names (Kimberly-Clark, Estée Lauder, Aetna, Humana). If your procurement process requires a dedicated named account team and a governance layer built for a multinational, Brandlight's model is arguably closer to what you're used to buying.
Feature comparison at a glance
| Capability | Brandlight | Scrunch (Sitecore) | Promptwatch |
|---|---|---|---|
| Public self-serve pricing | No | Yes, Core tier | Yes, all tiers |
| Engines at entry tier | 7 named, market-adapted | 4 (Claude/Gemini locked to Enterprise) | 8, all included |
| Crawler / server logs | Yes ("Server logs") | Yes, core to Agent Experience | Yes, full Agent Analytics |
| Content gap analysis | Structural, narrative-based | Site audit capped at 25 pages | Full-site crawled index |
| Automated content publishing | No (human/legal review required) | Basic generation, no auto-publish | Yes, direct CMS publishing |
| Reddit / YouTube citation tracking | Not highlighted | Source-level | Dedicated reports |
| ChatGPT Shopping / Ads tracking | Commerce module (SKU-level) | Shopping insights on higher tiers | Included from Business tier |
| Ownership | Independent, $36M raised | Owned by Sitecore (acquired June 2026) | Independent, seed-funded |
| Best fit | Multi-brand global CMO with budget | Teams prioritizing AI-agent technical delivery | Teams wanting monitoring plus execution, self-serve |
The scarcity problem none of these platforms can fix
Whatever platform you pick, you're optimizing for a shrinking pool of citation slots. Promptwatch's own citation research shows ChatGPT typically surfaces around 5 sources per web-search response, while Google AI Overviews and Perplexity cite closer to 10 each; Microsoft Copilot is the wildcard, swinging from fewer than 2 to nearly 17 sources per response within a matter of weeks. You can read the full breakdown in Promptwatch's average sources per response data.
That scarcity is exactly why the content-gap features matter more than the dashboards. A platform that tells you you're absent from a prompt but leaves you to figure out the fix on your own isn't finishing the job. This is also why crawler behavior matters: Promptwatch's own crawler data shows OpenAI's share of verified AI crawler requests dropped from 94.8% in early June 2026 to 79.8% by early September, with Anthropic, Google, Perplexity, and Mistral picking up the difference. If you're only watching ChatGPT crawl logs, you're increasingly missing a growing chunk of the picture.
Which one should you actually pick
If you're a global, multi-brand enterprise with a dedicated AI marketing function and the budget for a strategy partner, not just software, Brandlight's model fits that buying pattern, even with the pricing opacity. You're paying for a team as much as a tool.
If your core concern is whether AI agents can technically parse and crawl your site, and you're comfortable that your vendor now reports into a CMS company's roadmap, Scrunch's Core tier at $250/mo is a reasonable entry point, though the 25-page audit cap and locked-tier engines (Claude and Gemini require Enterprise) will bite larger catalogs fast.
If you want to see the problem and fix it without hiring an agency or sitting through a three-month pilot, Promptwatch's self-serve tiers starting at $95/mo cover more engines out of the box and pair that tracking with actual publishing. It's the only one of the three built to close the loop from a single subscription.
For teams still mapping out the wider field, the GEO software directory at bestgeosoftware.com is a useful place to keep comparing before signing anything, and Promptwatch's own 2026 platform comparison is worth reading in full if you want the scoring methodology behind claims like these.
If what you actually need is someone to execute the fixes rather than more software to manage, that's a different conversation. 1001 SEO Media runs technical SEO, content, and GEO work for brands trying to close exactly this kind of visibility gap, using Promptwatch as its own AI visibility platform of choice.