Welcome to the Sellforte MCP Use Case Library. Each post takes one question that marketing teams really ask their AI assistant through Sellforte MCP, which is now in general availability and accessible to all marketers worldwide. It strips out anything customer-specific and walks through what the assistant does with it, step by step. We run every question on our demo account so the numbers you see are illustrative, not anyone's real results. The idea is simple: steal the prompts, try them on your own model, and see what you get back. We hope you find this series inspiring!
-Juha, CEO at Sellforte
Around the middle of every month, someone in performance marketing has to answer the same thing: are we on track to spend the budget, or are we heading for an over- or underspend? The data is all there, but stitching together spend by platform and region, a run-rate and a month-end projection usually takes a spreadsheet and half an hour. Here's how to get it in one question.
Our demo account's data runs to 2026-11-01, so November had only one day in it when we ran this. To make the example useful, we ran the check as if it were 2026-10-20, with monthly paid budgets of 650 000 for Germany and 300 000 for the US.
For 2026-10-01 to 2026-10-20, show paid spend by country and by ad platform, excluding Email, against these October budgets: Germany 650 000, US 300 000. Give me the daily run-rate over the last 7 days, the projected month-end spend at that rate, and spend as a share of sales per country.
Now show where October actually landed by country.
The assistant pulls paid spend for the month so far by country, then the same for the last seven days to get a daily run-rate, and projects the remaining 11 days at that rate. The budgets come from the prompt, since they live outside Sellforte. Twenty days in, 64.5% of the month had passed (illustrative demo data, USD):
| Country | Budget | Spend to date | Share of budget | Daily run-rate, last 7 days | Projected month-end |
|---|---|---|---|---|---|
| Germany | 650 000 | 442 459 | 68.1% | 20 122 | 663 801 |
| US | 300 000 | 192 430 | 64.1% | 9 418 | 296 026 |
| Total | 950 000 | 634 888 | 66.8% | 29 540 | 959 827 |
So the mid-month read was: Germany is running a little hot, about 2% over budget at the current pace. The US is close to plan and slightly under.
The platform view explains the pace. These are the five largest paid platforms on the demo account (illustrative demo data, USD):
| Ad platform | Spend 2026-10-01 to 2026-10-20 | Spend 2026-10-14 to 2026-10-20 |
|---|---|---|
| Google Ads | 368 202 | 127 896 |
| Meta Ads | 141 473 | 49 596 |
| Direct Mail | 57 143 | 0 |
| Adtraction | 32 796 | 15 217 |
| TikTok Ads | 21 010 | 8 141 |
Direct Mail is the interesting row. It spent 57 143 earlier in the month and nothing in the last seven days. That's why the last-7-day run-rate is lower than the month's average daily spend so far. A one-off drop like that is exactly what you want spotted before you project anything.
Finally, the assistant pulls total sales per country for the same 20 days. Paid spend was 6.8% of sales in Germany and 5.8% in the US, or 6.5% overall. It's a simple ratio, but it's the one finance usually asks about first.
Because the demo data covers all of October, we could check the projection against reality. Germany finished at 630 732 and the US at 289 023, a total of 919 755. Both ended under budget, and Germany came in about 33 000 below what the mid-month run-rate suggested. Spend slowed in the final 11 days.
A run-rate is a projection, not a forecast. It assumes the next 11 days look like the last seven, and in this example they didn't. That's not a reason to skip the check. It's a reason to run it every week, so you catch the drift while there's still time to act on it.
The other thing to watch: the most recent days may not be fully loaded yet. A run-rate built on a half-loaded day will understate your pace, so confirm the last data day is complete before trusting the projection.
And when the check says you have room to spend more, or need to cut, don't just move money to the platform with the best average ROI. The next dollar should go where the marginal return is highest, campaign by campaign. Pacing tells you how much to move. The model tells you where.
If you'd rather not ask every week, a weekly Sellforte briefing can deliver the same pacing check.
Book a demo to run the pacing check on a model like yours, or read how to connect Sellforte MCP to your LLM client. Fresh use cases are added to this library regularly.