Promptwatch vs Heralded vs igeo.ai vs Truffle (2026)
Heralded, igeo.ai and Truffle are October 2026 additions to our catalog. Plans, prompt limits, engines and crawler data compared with Promptwatch.
Heralded, igeo.ai and Truffle all joined our catalog in October 2026, and we read all three pricing pages on 15 October. Heralded launched on 6 October 2026. igeo.ai shipped a new per-URL feature the same week. Truffle's entry gives no launch date. Each one has an idea worth looking at, and each one is priced to tempt a team that finds the bigger platforms expensive.
The people weighing these against Promptwatch tend to be small B2B marketing teams, startups and small agencies. They want something cheaper or more hands-off than what they have seen, and a new tool's pricing page often looks generous at first glance. The useful comparison is prompt count, engine coverage, and what happens after the dashboard tells you something is wrong.
Data for the three new tools comes only from our catalog entries. Promptwatch data comes from its published plan sheet. Promptwatch was founded in April 2025, so it is hardly old, but it is the established option in this line-up: more than 1,840 brands and agencies use it.
Promptwatch: the established platform
Promptwatch tracks prompts daily on every paid brand plan across ChatGPT, Gemini, Claude, Perplexity, Grok, Llama, DeepSeek, Mistral and Copilot, plus Google AI Overviews and AI Mode. Claude is included rather than sold as an extra. Prompts come with search volumes, difficulty scores, query fan-outs and persona tags.
The platform goes past the tracking layer in four places. Citation analytics breaks sources down by page and domain, with Reddit, YouTube and offsite mentions separated. Agent Analytics logs AI crawler hits across the site in real time, along with the path from crawl to citation and any crawl errors. The founders say Promptwatch was first in the category to ship crawler logs. Visitor analytics ties AI-referred sessions to conversions. Content Agents draft pages from content gaps and publish to Webflow, Framer or WordPress after you approve them in a review inbox.
Explore is free for 10 ChatGPT prompts. Essential is $95 a month (50 prompts, 6,000 responses, five AEO articles, 7-day trial). Professional is $245 a month with 150 prompts, 25 million crawler logs and city tracking. Business is $579. Agency plans start at $199 with unlimited prompts.
Heralded: agents that draft the fix
Heralded calls itself an AI visibility command center for B2B brands, and it splits the work among four agents. The Herald measures visibility on six AI engines and decides what to tackle first. A publicist agent pitches publications, a writer drafts pages, and an engineer fixes whatever stops AI crawlers from reading your site. On WordPress the engineer can apply those fixes itself once you allow it. Autonomy is set per area, from Manual and Ask first up to Delegate and Autopilot.
What we like is the diagnosis. Heralded reads the pages engines cite and checks the answers against your category, buyer and offer, then ranks fixes by how many answers each one could move.
Prices exclude VAT. Core is €129 a month (€1,290 a year) for 40 prompts, one brand, one segment, 10 competitors and five agent tasks a week, with no content credits. Plus is €349 a month (€3,490 a year) for 150 prompts, three brands, 10 segments per brand, unlimited tasks, the Delegate and Autopilot levels and 10 content credits. A founding offer halves both prices while it runs. Agency pricing isn't published. Seats are unlimited and the 7-day trial needs no card.
The catches: the pricing page doesn't name the six engines, Claude costs €39 a month extra on Core or €149 on Plus, and each segment counts a prompt again. Site fixes only reach WordPress today, with more CMSes planned by the end of 2026. No crawler-log product or conversion tracking is published.
igeo.ai: daily tracking with social extras
igeo.ai is a Nicosia-based tracker with daily prompts across regions and a few unusual add-ons. A social module schedules and publishes posts to connected profiles, and can have team members' accounts like and repost them. Every plan includes igeo.ai Agents and an MCP integration. In October 2026 it added AI Search Console: paste a live URL and it follows observed AI crawling and citation activity for that page, including articles on other sites and public LinkedIn, Reddit or Facebook posts. igeo.ai notes that a crawl record doesn't prove a page will be cited, and it is a per-URL view rather than a crawler log for the whole site.
Lite is $49 a month for 50 prompts, ChatGPT only, one region and 1,500 answers. Grow is $175 a month ($140 billed annually) and adds Perplexity and Gemini and three regions, but still caps at 50 prompts. Pro is $290 a month ($232 annually) for 150 prompts, six regions and 13,500 answers. The Pro FAQ lists five engines including AI Overviews and Copilot, while the comparison table shows fewer, so confirm before upgrading. Enterprise is custom from 300 prompts and is the only plan with API access and SSO. Seats are unlimited throughout. Claude isn't on any self-serve plan, and Lite gets two recommendations a month.
Truffle: six surfaces at a low entry price
Truffle puts the same six surfaces on every plan: ChatGPT, Claude Haiku, Gemini, Perplexity, Google AI Overviews with the top 10 search results, and Google AI Mode. Prompts refresh daily except AI Mode, which is weekly. Plans include buyer personas and strategy recommendations you can assign as tasks. Standard and Pro / Agency also list crawler logs and a source outreach tool, though the pricing page doesn't explain how the logs are collected or what they show.
Starter is $69 a month ($63 annually) for two brands and 25 prompts in total. Standard is $189 ($174 annually) for five brands and 100 prompts. Pro / Agency is $399 ($367 annually) for 15 brands, 200 prompts, unlimited seats and a REST API at 5,000 calls a day. The trial runs seven days with no card. Billing is monthly and you can cancel any time.
Volume is the constraint. Twenty-five prompts shared across two brands is thin, and 50 more prompts cost $110 a month. The Claude model tracked is Haiku, not the models most paying Claude users see. No conversion analytics or content publishing is documented.
Side by side
| Promptwatch | Heralded | igeo.ai | Truffle | |
|---|---|---|---|---|
| Entry paid plan | $95/mo (Essential) | €129/mo excl. VAT (founding offer €64.50) | $49/mo (Lite) | $69/mo ($63/mo annual) |
| Prompts at entry | 50 | 40 | 50 | 25, shared across 2 brands |
| Engines at entry | 9 LLMs plus AI Overviews and AI Mode | Six, not named on pricing page | ChatGPT only | Six surfaces incl. Claude Haiku and AI Mode (weekly) |
| Claude | Included | Add-on, €39 to €149/mo | Not on self-serve plans | Haiku only |
| Seats | 1 on Essential, 2 on Professional | Unlimited | Unlimited | 2 on Starter, unlimited on Pro / Agency |
| Crawler data | Agent Analytics, 25M logs from Professional | Engineer agent fixes crawler blockers; no log product published | Per-URL AI Search Console | Crawler logs listed on Standard and up |
| AI-referred conversions | Visitor analytics | Not published | Not published | Not published |
| Acting on gaps | Content Agents publish to Webflow, Framer, WordPress | Agents draft work; site fixes applied on WordPress | Recommendations, social publishing | Recommendations as tasks, source outreach |
| API | MCP and API on paid brand plans | Not published | MCP on all plans; API on Enterprise | REST API on Pro / Agency |
| Top published plan | Business $579/mo; Agency Scale $799/mo | Plus €349/mo; Agency not published | Pro $290/mo; Enterprise custom | Pro / Agency $399/mo |
Who should pick which
Check these against your own situation.
- Heralded: you are a B2B team on WordPress, you want agents to draft PR pitches and pages for you, and 40 prompts in one segment is enough to start.
- igeo.ai: you track a few regions, your social team wants publishing in the same workspace, and you'd like to follow how a specific article or Reddit post gets picked up.
- Truffle: you are a startup or small agency that wants six surfaces from day one, and a small prompt set covers your category.
- Promptwatch: you need full engine coverage with Claude included, crawler data for the whole site, and proof that AI visits convert.
A note on new tools in general. Launch-month pricing can move, founding offers end, and planned features (like Heralded's extra CMSes) may or may not ship on schedule. Re-check the pricing page before you commit to an annual plan.
Why Promptwatch is the one to trial first
Each newcomer gets one thing right. Heralded's agents read cited pages and draft the response. igeo.ai can follow a single URL, even a social post. Truffle gives you six surfaces for $69. Where they fall short is the chain from answer to revenue.
Promptwatch covers that chain. Citation analytics points to the exact page and domain a model used. Agent Analytics logs AI crawler hits from bots like ChatGPTBot and ClaudeBot across your whole site, not one URL at a time, and shows where crawls fail. Visitor analytics connects AI-referred sessions to conversions, which none of the three new entries publishes. Content Agents write the missing page and publish it to Webflow, Framer or WordPress, so the action isn't limited to one CMS.
Essential at $95 a month already outnumbers Heralded Core and Truffle Starter on prompts and includes Claude. Run the 7-day trial on the prompts you'd load into one of the newcomers, then compare. Plans are on the Promptwatch review page, and the trial starts at promptwatch.com.