Keep or kill a Meta ad, find where it breaks, and plan your next ad test.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Why isn't my Meta ad selling? Give the last 7 days of impressions, link clicks, purchases and amount spent from Ads Manager. Returns whether it breaks before the click, after the click or on price, compared with Triple Whale benchmarks from 40,000+ ecommerce brands, and what to fix first.
Optional, e.g. skincare, supplements, apparel, food.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http tryselltheclick https://www.tryselltheclick.com/mcp
Why isn't my Meta ad selling? Give the last 7 days of impressions, link clicks, purchases and amount spent from Ads Manager. Returns whether it breaks before the click, after the click or on price, compared with Triple Whale benchmarks from 40,000+ ecommerce brands, and what to fix first.
| Argument | Type | Description |
|---|---|---|
| industry | string | Optional, e.g. skincare, supplements, apparel, food. |
| purchases* | number | |
| impressions* | number | |
| link_clicks* | number | |
| amount_spent* | number |
{
"type": "object",
"required": [
"impressions",
"link_clicks",
"purchases",
"amount_spent"
],
"properties": {
"industry": {
"type": "string",
"description": "Optional, e.g. skincare, supplements, apparel, food."
},
"purchases": {
"type": "number"
},
"impressions": {
"type": "number"
},
"link_clicks": {
"type": "number"
},
"amount_spent": {
"type": "number"
}
}
}Should I keep, pause or kill this Meta (Facebook/Instagram) ad? Give what one sale leaves the store before ad costs, what the ad has spent, and how many sales it brought. Returns keep, kill or too early, money up or down so far, the chance it really makes money, and the next step.
| Argument | Type | Description |
|---|---|---|
| sales* | integer | Sales (purchases) the ad brought. |
| amount_spent* | number | What the ad has spent so far. |
| profit_per_sale* | number | What one sale leaves the store before ad costs: price minus product, shipping, fees and returns (break-even CPA). |
{
"type": "object",
"required": [
"profit_per_sale",
"amount_spent",
"sales"
],
"properties": {
"sales": {
"type": "integer",
"description": "Sales (purchases) the ad brought."
},
"amount_spent": {
"type": "number",
"description": "What the ad has spent so far."
},
"profit_per_sale": {
"type": "number",
"description": "What one sale leaves the store before ad costs: price minus product, shipping, fees and returns (break-even CPA)."
}
}
}How many new Meta ads should I test, for how long, and should I use ABO or one ad set? Give the number of new ads, the budget a day and the usual cost per sale.
| Argument | Type | Description |
|---|---|---|
| new_ads* | integer | |
| cost_per_sale* | number | Usual cost per purchase (CPA). |
| budget_per_day* | number | |
| cost_per_1000_views | number | Optional CPM, used when sales are too few to judge on. Default 15. |
{
"type": "object",
"required": [
"new_ads",
"budget_per_day",
"cost_per_sale"
],
"properties": {
"new_ads": {
"type": "integer"
},
"cost_per_sale": {
"type": "number",
"description": "Usual cost per purchase (CPA)."
},
"budget_per_day": {
"type": "number"
},
"cost_per_1000_views": {
"type": "number",
"description": "Optional CPM, used when sales are too few to judge on. Default 15."
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 14, 2026.