Ask Sellforte MCP: What does this saved Q4 plan really change?

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

Explaining a saved Q4 plan before you present it

Someone on your team built a Q4 scenario in the Media Optimizer weeks ago, and now you're the one presenting it. You need to say what it actually changes, what it costs in sales, and how it stacks up against what you really spent last year. Media planners and agency analysts get this one all the time, often with a skeptical stakeholder asking why the numbers look the way they do.

What you ask

Summarize scenario "Q4 2026 Optimized -10%" against "Q4 2026 BAU (Q4 2025 spend)". What changed in budget and by channel, and what happens to incremental sales and ROI? Show me how each scenario was set up first.

Now compare the plan with the same weeks last year: spend, incremental sales and ROI, and explain the biggest differences.

What Sellforte MCP does

Checks the setup before the results

The assistant lists the saved scenarios and tries to read each one's configuration. On the demo account, neither has a recorded setup: both were created before the optimizer started recording settings. So it says that plainly and doesn't guess the settings from the scenario names. It also flags both scenarios as stale. They were built on 2026-08-20, and the marketing model was refreshed on 2026-11-02, so the figures may reflect the previous model.

That's two useful warnings before you've seen a single number.

Compares the plan with its baseline

Both scenarios cover the same 15 weeks, 2026-09-27 to 2027-01-09. Here are the totals (illustrative demo data, USD):

So the plan takes out 10% of the budget and loses about 8% of incremental sales, and ROI goes up a little because the dollars it removes were earning less than average.

Where the money comes from is the interesting part. It isn't an even trim. Two channels carry most of the cut: Direct Mail drops by 145 960 (−29.5%) and TV by 62 358 (−29.8%). Most other channels come down by around 6%, and Google Performance Max gives up 77 213 in absolute terms simply because it's the biggest line. One channel goes the other way: Google YouTube gets 6 971 more (+7.7%) and adds about 37 000 incremental sales.

Puts it next to last year

For the second prompt, it pulls the MMM results for the same 15 weeks last year, 2025-09-28 to 2026-01-10 (illustrative demo data, USD):

And here's the catch. The baseline is called "Q4 2025 spend", but its 4 276 232 is well below the 6 919 069 actually spent in those weeks last year. On the demo data, the name doesn't describe the number. Without a recorded setup, the assistant can't say how that baseline was built, and it says so. That's exactly the kind of thing you want to find before a stakeholder does.

The channel lines show where the gap sits. Last year Direct Mail spent 820 000 and TV 739 339 in those weeks; the plan has 348 415 and 147 189. Performance Max goes from 2 028 020 to 1 313 453. A couple of Meta channels move the other way: Advantage+ barely ran last year (1 388) and has 176 542 in the plan.

One more caveat it adds: last year's ROI is the model's historical read, while the plan's ROI is a forecast. They're worth comparing, but they're not the same kind of number.

Why it matters

A saved scenario is easy to present and hard to defend. Splitting it into "what the smaller budget costs" and "where the mix moves" turns one total into two decisions you can discuss separately. Here, a 10% cut costing about 8% of incremental sales is the budget story, and the concentrated cuts in Direct Mail and TV are the mix story. If you want the full picture, run the same budget with last year's mix next to it.

The last-year check is what makes the plan credible. If the baseline doesn't match what you really spent, that's the first thing to fix, and a stale scenario should be re-created on the current model before anyone signs off on it.

Try these follow-ups

• Show the -10% plan week by week against BAU. Where does the cut land in the calendar?
• Compare "Q4 2026 Optimized (flat budget)" with both of these. How much of the change comes from the mix and how much from the budget?
• Split the plan by country. Does one market take more of the cut?
• Which channels in the plan sit furthest from their last-year spend, in percent?

Related

• Sellforte MCP
• Sellforte Optimizer
• Media Planner Agent
• Multi-market MMM
• Use case: Where should we move budget mid-month?

Try it yourself

Curious what your own saved plans look like under this kind of scrutiny? Book a demo, or see how to connect Sellforte MCP to your LLM client. The launch post has more on what Sellforte MCP can do.

We add new use cases to this library regularly.