Ask Sellforte MCP: Should we keep, scale or cut this campaign?
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
Deciding what a campaign gets after its flight ends
A campaign flight wraps up and someone has to decide what happens next: run it again, give it more money, or move the budget somewhere else. That's usually the campaign or performance marketer, squeezing the review in between planning meetings. What you need is a short, honest read of what the flight earned and a clear call for the next one.
What you ask
Review our Sales campaigns (campaign names containing "Sales") for 2026-07-01 to 2026-09-30: spend, incremental sales and ROI overall and by campaign. Compare with the same campaigns for 2026-04-01 to 2026-06-30. Which ones should get more or less budget next time?
Before we decide, what do Sellforte's performance recommendations say for these campaigns?
We ran this on our demo account, treating Q3 2026 as the flight and Q2 2026 as the earlier flight to compare against. Every figure below is illustrative demo data, in USD.
What Sellforte MCP does
Finds the campaigns behind the name
The fragment "Sales" matched six campaigns: Meta, TikTok and Pinterest, each in Germany and the US. The two Pinterest campaigns had no spend in Q3, because on the demo account they only start spending in October 2026. So the review covers four campaigns.
Worth a quick look every time: a fragment can catch campaigns you didn't mean or miss a second naming pattern. If total spend doesn't match the budget you know, fix the names first.
Reviews the flight
Next it pulls spend, incremental sales and ROI from the marketing mix model for each campaign, with Q2 alongside.
| Campaign | Spend Q3 | Incremental sales Q3 | ROI Q3 | ROI Q2 | ROI change |
|---|---|---|---|---|---|
| Meta Ads_Meta Sales_DE | 108 949 | 1 500 584 | 13.77 | 8.70 | +58% |
| Meta Ads_Meta Sales_USA | 47 937 | 608 025 | 12.68 | 7.59 | +67% |
| TikTok Ads_TikTok Sales_DE | 17 032 | 27 174 | 1.60 | 1.59 | 0% |
| TikTok Ads_TikTok Sales_USA | 9 118 | 13 920 | 1.53 | 1.48 | +3% |
| Total | 183 036 | 2 149 704 | 11.74 | 7.31 | +61% |
The flight spent 5.2% more than Q2, and incremental sales rose 69%. Nearly all of that came from Meta. Both Meta Sales campaigns spent 6.67% more, but impressions went up 30.34% and CPM fell 18.16%, so the same money bought a lot more reach. TikTok spent slightly less, and its ROI barely moved.
Asks the engine what to change
This is where the review stops looking backwards. Past ROI tells you what a campaign earned on average. It doesn't tell you what the next dollar will earn. So the follow-up asks for Sellforte's performance recommendations for the same campaigns, which work from marginal return (miROAS), campaign by campaign. The recommendation snapshot covers 2026-10-05 to 2026-11-01.
| Campaign | Daily spend | Recommended daily spend | Change | miROAS |
|---|---|---|---|---|
| Meta Ads_Meta Sales_DE | 1 406 | 1 406 | Hold | 6.24 |
| Meta Ads_Meta Sales_USA | 618.55 | 618.55 | Hold | 6.20 |
| TikTok Ads_TikTok Sales_DE | 158.97 | 104.01 | -34.6% | 1.42 |
| TikTok Ads_TikTok Sales_USA | 83.67 | 67.65 | -19.1% | 1.46 |
| Pinterest Ads_Pinterest Sales_DE | 22.66 | 11.82 | -47.8% | 0.861 |
| Pinterest Ads_Pinterest Sales_USA | 9.97 | 5.20 | -47.8% | 0.879 |
No increases, four cuts and two holds. Recommended daily spend across the six goes from 2 300 to 2 213, with an expected drop of about 156 in daily incremental sales. The cuts land on TikTok and Pinterest, where miROAS sits between 0.861 and 1.46. For both Meta campaigns the spend stays put. The only suggested change there is to bring the daily budget setting in line with what the campaign actually spends: 1 343 to 1 405.8 in Germany and 700 to 618.55 in the US.
Why it matters
Put the two answers next to each other. On past ROI alone, Meta Sales looks like the obvious thing to scale: an ROI of 12.68 to 13.77, up sharply on the previous flight. The recommendation engine doesn't suggest scaling it, though. It says hold. A strong average and room for more budget aren't the same thing, and a review built only on historical ROI can't tell them apart.
So ask in that order: what did the flight really earn, measured as incremental sales from the model, and then what should change, based on marginal return. That beats repeating last time's budget plus a bit.
Two practical tips. If you want the campaign on its own, say "no comparison" in your first message, because assistants sometimes add a period comparison you didn't ask for. And if you're planning the budget across the whole mix rather than one campaign, use an optimizer scenario, since the optimizer plans channels, not single campaigns.
Try these follow-ups
- Show the Meta Sales campaigns week by week for 2026-07-01 to 2026-09-30, so I can see whether the ROI held through the whole flight.
- Review the same campaigns for 2026-07-01 to 2026-09-30 again, but with no comparison.
- Which ad sets inside Meta Ads_Meta Sales_DE spent the most during the flight?
- Is there a saved optimizer scenario for next quarter, and what does it do with Meta and TikTok?
Related
- Sellforte MCP
- Marketing mix modeling
- Sellforte Performance
- Media Buyer Agent
- Ask Sellforte MCP: Which Meta campaigns should we change first?
- Ask Sellforte MCP: Did our budget changes pay off?
Try it yourself
Book a demo, or if you're already a customer, here's how to connect Sellforte MCP to your AI assistant.
New use cases land in this library regularly, so check back soon.
