MCP 101
This is how a user signs in to Meltwater MCP
Meltwater sign in OR SSO
This is the name of the consumption unit that powers MCP pricing
Intelligence Credits
The Pro tier: credits, price, and ideal customer?
3M credits, $32,500/yr — small to medium in-house PR team.
Checking, creating, and managing searches, lists, tags, dashboards, and alerts costs credits.
Zero — everyday actions in the Meltwater apps stay included under the core license. Credits meter AI work only.
What is trial length and credit volume for a Meltwater MCP net-new trial?
15 days, 10M credits
These are the two AI assistants supported at general availability.
Claude and ChatGPT
(T/F) Customers can buy MCP as a standalone offering
False
The Enterprise tier: ideal customer, credits and price?
6M credits, $62,000/yr — for a global organization and scope.
An "Intelligence Action" — a single question of your media dataset — runs about this many credits.
25 credits
Objection handling: Won't this replace our Meltwater seats?" Your response?
No — MCP expands who can ask.
Deep work stays in the platform. More of the org touches Meltwater data through MCP than would ever log in directly, so it grows the footprint rather than shrinking it.
Every Meltwater MCP answer includes these, so the team can trust the numbers and trace them back to the source.
Citations
The Lite tier includes this many credits at this annual price and is best for what size teams
1M credits at $11,500/yr and is best for lean teams
The Advanced tier: credits and price
15M credits, $144,000/yr — large enterprise, multi-brand org.
"Advanced Synthesis" — deep AI analysis in one pass, like competitive research or a narrative map — runs about this many credits.
200 credits.
Objection Handling: "How do I know it won't just make numbers up?"
any one of three for credit:
(1) When there's no Meltwater data, it returns nothing rather than inventing an answer — it shows the exact query it ran.
(2) Requests outside a user's entitlements are refused, not filled in with other data.
(3) Anything the assistant adds from general knowledge is labeled as general knowledge, never attributed to Meltwater.
Explain Meltwater MCP to a customer in one sentence or less.
USB-C for AI
or
Meltwater MCP brings Meltwater intelligence into the AI assistants your team already uses — sign in, ask a question in plain language, and get an answer grounded in your Meltwater data with cited sources.
This is the maximum discount allowed on MCP
20%
A customer wants 45M credits. How do you structure it?
Stack PLIs! (Sell 3x Advanced Package)
A "Full AI-Built Deliverable" — a complete end-to-end output like a branded report or media briefing pack — runs about this many credits.
800 credits.
Name an MCP use case that stands out from basic platform/mira use cases:
Executive briefs on a schedule — "Give me a 5-bullet morning briefing on competitor activity from the last 24 hours." This one lands hard with VPs who will never log into Meltwater themselves.
Cross-source analysis — the sleeper differentiator. Inside the assistant, Meltwater data can be combined with the customer's other connected tools — internal docs, HubSpot, Salesforce, Notion. Meltwater becomes one trusted input alongside their own systems, which no dashboard can do.
Open-ended exploration — "Drill into that. What are the top three outlets driving negative sentiment?" → "Now compare that to last month." Each question builds on the last instead of starting a new search every time. This is the "aha" moment in most demos.
This is what the letters M-C-P stand for
Model Context Protocol
Describe the credit rollover policy in full
Annual Hard Cap; Credits must be used within the contract term, non-transferrable, and no rollover year over year.
Name the four annual credit tiers
Lite, Pro, Enterprise, and Advanced.
Do the math: a Media Research Analysis deliverable plus four follow-up questions of the dataset totals this many credits.
900 credits (800 for the deliverable + 4 × 25 for the questions)
Meltwater MCP vs. the Meltwater Insights API
whats the core difference?
Insights API returns structured raw mention data (articles, posts, metadata) for BI pipelines like Power BI and Looker Studio.
MCP returns AI-generated conversational answers with citations — no raw mention data or article text.