You see a category is growing. The next step is a campaign brief, a prompt to write descriptions, or a job for an agency. The data you already have can become the input for the next task right away.
Analysis is rarely an end in itself. You see one category shoot up, and the next step appears immediately: let’s turn it into a campaign, write the descriptions, prepare a brief for the agency. But between “I know what’s happening” and “I’m acting” there’s a tedious gap – you have to write down again the context and numbers you were just looking at.
This is exactly where the data usually evaporates. The brief gets written from memory, the prompt for the second tool gets typed from scratch, and the agency asks anyway: “and what exactly are your numbers?”. But since the store’s data is already connected, the same material – brief, prompt, job spec – can be created with those numbers inside, in a single sentence.
Below is a real example. Starting point: last week one category grew by 29.7%. The numbers are real, the store anonymized to specialty coffee.
Starting point – something is growing
In the top 5 categories, the fastest grower was functional coffee: +29.7% (14,168 €, 15.8% of weekly revenue). It’s the third category by revenue, but the only one with double-digit growth – the rest of the leaders fell or stood still. 303 orders, average cart in the category 46.76 €.
Data → campaign brief
Functional coffee is the fastest-growing category in the top 5 (+29.7% week over week), third by revenue with a 15.8% share. In the same week ground coffees fell by 19.2% and the classics stood still – so the growth is concentrated in one clear segment. 303 orders, average cart in the category 46.76 €.
Sustain and accelerate the category’s growth: raise functional coffee’s share of revenue from 15.8% to ~20% in 8 weeks, without cannibalizing the classics.
- Existing customers buying classic coffees – the “add a functional one to your cart” offer.
- New shoppers looking for coffee “with an added benefit” (energy, focus, immunity).
- A set of creatives for Meta and Google (3 message variants).
- A homepage section + a dedicated category page.
- An email to the base of classic-coffee customers.
- No deep discounts – the category grows organically, it doesn’t need subsidizing.
- No health claims that the ingredients don’t support.
The same brief → a prompt for the next tool
Brief and prompt are the same material – in two formats
What is a brief for a person is a prompt for another AI tool. In both cases the same thing matters most: context and numbers. The agency brief has them in the header and the “context” section; the prompt has them bulleted at the top. You don’t need to retype them twice – they come from the same analysis.
The classic problem with AI “that spouts nonsense”: it didn’t know your store, so it guessed. A prompt with real numbers fixes that.
Neither of these documents is final, and that’s how it should be. You check the brief before sending it, you edit the headlines from the second chat. The point is that the input material – with the right context and current numbers – takes a dozen-odd seconds to produce, not half an hour of retyping.
The end of working with data is rarely a number in a chat. More often it’s material you carry forward to the next step, based on the store’s data.
Takeaway
The value of analysis doesn’t end when you see a number – it begins when that number moves to the next task without retyping. It’s worth treating data as input material, not just a result to read: the same set of facts can be a brief for an agency, a prompt for another tool, or a job spec for a freelancer. When the path from “I see it” to “I have something to act on” costs one sentence, you act more often and faster.
DataOrganizer · MCP
Brief, prompt, job spec – from your data, in one sentence. Set it up in 15 minutes.