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
Quarterly business reviews tend to use the same deck every time: how sales split between base, promotions and media, what each media group spent and returned, and a weekly trend. The slides don't change much. The numbers do, and they have to be pulled again every time the model is refreshed. If you're a customer success manager or a marketing lead preparing for leadership, that's a few hours of copying you'd rather skip.
For Germany, Q3 2026 (2026-07-01 to 2026-09-30) versus Q3 2025: total sales split into base, promotions and media; spend, incremental sales and ROI by media group; and a weekly trend of spend and incremental sales for Q3 2026. Format each as a table I can paste into the deck.
The assistant runs one pull per slide, each with last year's quarter as the comparison window.
Illustrative demo data, USD:
| Component | Q3 2025 | Q3 2026 | Change |
|---|---|---|---|
| Base | 12 079 154 | 11 498 115 | -4.8% |
| Promotions | 3 863 781 | 3 124 155 | -19.1% |
| Media | 8 764 636 | 10 460 241 | +19.3% |
| Weather | 71 419 | -413 675 | -485 095 |
| Total sales | 24 778 991 | 24 668 835 | -0.4% |
On the top line, the quarter was flat. Underneath, it's a different story: media contributed about 1.7 million more than last year, while base demand, promotions and weather all pulled the other way.
The six largest media groups by spend (illustrative demo data, USD):
| Media group | Spend Q3 2026 | Spend change | Incremental sales Q3 2026 | ROI Q3 2025 | ROI Q3 2026 |
|---|---|---|---|---|---|
| Cross-network | 692 006 | +27.4% | 3 551 720 | 6.10 | 5.13 |
| Direct Mail | 502 857 | +29.4% | 2 106 266 | 3.58 | 4.19 |
| Paid Social | 427 029 | +13.3% | 2 913 846 | 3.04 | 6.82 |
| 325 462 | +9.2% | 475 048 | 2.01 | 1.46 | |
| TV | 291 240 | +7.1% | 316 521 | 1.20 | 1.09 |
| Paid Search | 134 042 | -33.7% | 581 108 | 3.54 | 4.34 |
| All media groups | 2 584 999 | +3.7% | 10 550 678 | 3.51 | 4.08 |
Total spend was up only 3.7%, but incremental sales from media rose 20.5% and ROI went from 3.51 to 4.08. Paid Social is the headline: a 13% spend increase, with incremental sales up by about 1.8 million and ROI more than doubling. Paid Shopping, a smaller group not shown above, was cut by 86%.
The media total here (10 550 678) differs a little from the media line on slide 1 (10 460 241). That's by design. Slide 1 counts media's effect in the weeks it lands, while slide 2 credits carryover back to the weeks the money was spent, which is how ROI is calculated. Sellforte points this out in its answer, and it's worth a footnote on the slide.
The weekly pull comes back as 14 rows, one per week, ready to chart. Weeks run Sunday to Saturday on the demo account, so the window widens to 2026-06-28 to 2026-10-03, and the assistant says so. Weekly spend climbed from 141 101 in the first week to a peak of 358 124 in the week of 2026-09-20. Weekly ROI ranged from 3.07 to 5.27 through July and August, and was 3.22 to 3.32 in the three heaviest weeks at the end of September (illustrative demo data, USD).
A flat quarter can hide a lot. Without the decomposition, this one reads as "nothing changed". With it, you can see that media did more work than last year while underlying demand and promotions softened. Those are very different conversations to have with leadership, and only a model that separates them lets you have the right one.
The weekly view adds the next question before anyone asks it: ROI was lower in the weeks with the most spend. That's a prompt to check where the curve flattens before planning the next quarter.
And because the deck is now a repeatable pull, refreshing it after a model update means asking the same question again. A tip from real use: for long weekly windows across many markets, pull one market at a time. Big weekly pulls can time out.
Book a demo to build your review deck from a model like yours, or read how to connect Sellforte MCP to your LLM client. Keep an eye on this library: new use cases arrive regularly.