AI Prompts for CPG & Consumer Goods Teams

Due-to bridge: explain exactly why volume moved

Quarterly business review: the volume decomposition is done, but leadership needs the driver story — not the spreadsheet.

You are a CPG insights analyst writing the driver narrative for a volume due-to (decomposition) analysis. I will paste the decomposition outputs and context. Produce:

A) A ranked DRIVER TABLE — columns: driver (distribution, velocity, base price, promotion, mix, new items, lost items), volume impact as given, direction, and a one-line plain-English explanation a non-analyst executive can read.
B) A 5-sentence NARRATIVE that leads with the single biggest driver, quantifies it from my numbers, and labels each driver structural (e.g., distribution losses) vs temporary (e.g., promo timing).
C) THREE follow-up cuts to run next (by retailer, pack size, or region) and what each would confirm or kill.

Data and context: [PASTE DUE-TO OUTPUT: driver names + volume or dollar impacts, period, geography, brand vs category trend]

Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in the data I give you, write "not provided" and flag it. Mark every claim I should verify against my syndicated data or internal reporting before using it externally. Never include retailer-confidential terms or personally identifiable shopper data.

Velocity vs distribution: is this growth real?

Your brand is up and someone wants to plan against it. Before they do, you need to know whether growth is quality (velocity-led) or bought (distribution-led).

You are a syndicated data analyst diagnosing the QUALITY of a sales trend. I will give you dollar or unit sales, distribution (TDP or %ACV), and velocity (sales per point of distribution) for my brand and, where I have it, the category. Produce:

A) A VERDICT — velocity-led, distribution-led, or mixed — stated in one sentence with the supporting arithmetic from my numbers only.
B) A decomposition TABLE: period, sales change, distribution change, velocity change, and which component carried the move.
C) RISK FLAGS in plain language: distribution-led growth that velocity cannot support (future delist risk), velocity-led growth with flat distribution (the expansion case to sell), or shrinking distribution masked by strong velocity.

My data: [PASTE: sales, TDP/%ACV, velocity by period; category comparators if available]

Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in the data I give you, write "not provided" and flag it. Mark every claim I should verify against my syndicated data or internal reporting before using it externally. Never include retailer-confidential terms or personally identifiable shopper data.

Trade promo post-mortem that tells the truth

The event is over and the lift number looks good in the deck. Did the promotion actually pay, or did it just buy volume forward?

You are a trade promotion analyst writing an honest post-event review. I will paste the event results. Produce:

A) A VERDICT — paid out, broke even, or bought volume forward — with the arithmetic shown step by step using only the numbers I provide (base vs incremental volume, lift, deal depth, feature/display support, event cost if I have it). If the payout math cannot be completed from my numbers, say exactly which input is missing rather than estimating it.
B) DIAGNOSTIC QUESTIONS for what the data cannot see: forward-buying, pantry loading, cannibalization of my own items, and halo — each phrased as a specific check I can run.
C) THREE GUARDRAILS for the next event: a depth cap, a support requirement (feature/display condition), and a timing rule — each tied to what this event's numbers showed.

Event data: [PASTE: base volume, incremental volume, lift %, deal depth, support type, retailer, dates, cost if known]

Rules: Do not invent, estimate, or extrapolate any figure — if a number is not in the data I give you, write "not provided" and flag it. Mark every claim I should verify against my syndicated data or internal reporting before using it externally. Never include retailer-confidential terms or personally identifiable shopper data.

Price-pack architecture read: what did the price change teach you?

Pricing moved — yours or a competitor's — and volume responded. Turn the response into a price-pack architecture insight before the next planning cycle.

You are a revenue growth management analyst reading a price event. I will paste price and volume by pack size or tier, before and after the change. Pr

New database sanity-check: before you trust a single number

A new data delivery, a restatement, or a changed market definition just landed. Run the QC gauntlet before anyone builds a story on it.

You are a syndicated data steward. I will describe my database setup and you will produce a NUMBERED QC CHECKLIST customized to it — each item with a

Share-loss forensics: who is taking it, where, and with what lever

Share is slipping and the meeting is Thursday. Build the suspect list and the confirmation plan before opinions fill the vacuum.

You are a competitive insights analyst running a share-loss investigation. I will paste my share trend and whatever cuts I have (retailer, region, seg

Base vs incremental: what did the promotion actually buy?

Trade spend review: the promo moved volume, but nobody can say how much you would have sold anyway.

You are a CPG trade analytics lead separating BASE volume from INCREMENTAL volume for a promoted period. I will paste weekly POS and promo detail. Pro

On-shelf availability: size the sales you never got to make

You suspect out-of-stocks are eating the number but the loss has never been quantified in dollars.

You are a CPG supply-and-insights analyst quantifying lost sales from poor on-shelf availability. I will paste store-item level data. Produce:

A) A L

Price sensitivity read from your own history — with the caveats attached

Finance wants an elasticity number before a list price move, and the only data you have is your own price history.

You are a CPG pricing analyst estimating price sensitivity from observed history, and you are deliberately conservative. I will paste price and volume

POS vs shipments: reconcile the gap before someone else finds it

Consumption is up, shipments are down (or the reverse), and the two numbers are about to meet in a leadership deck.

You are a CPG demand analyst reconciling retailer POS consumption against your shipment data. I will paste both series. Produce:

A) A GAP TABLE by pe

Category review skeleton: retailer-first, brand-last

The annual category review is due. You have the data; you need the structure and the story a merchant will actually engage with.

You are a category advisor building a retailer category review. I will paste what I have: category size and growth, segment trends, this retailer vs t

Assortment rationalization: keep / cut / watch

Reset season. The item list needs a defensible keep/cut/watch call before the planogram meeting — not a velocity sort with feelings.

You are an assortment analyst preparing a SKU rationalization. I will paste item-level data: velocity, distribution, and — where I have it — increment

Panel measures → growth levers: penetration or buy rate?

Household-panel numbers are in — penetration, buy rate, frequency, trip size. Translate them into which growth lever is actually available to you.

You are a shopper insights analyst decomposing brand buyer dynamics from household panel data. I will paste my panel measures vs year-ago and vs categ

Shopper study design: hypotheses before methodology

Someone wants to commission shopper research. Before money moves, force the hypotheses and check whether existing data already answers them.

You are an insights lead scoping a shopper study. I will describe the business question and what our existing data already shows. Produce:

A) FIVE te

Shelf-space argument builder: share of shelf vs share of sales

Reset season again — and your space ask needs merchant math (space-to-sales, days of supply), not brand enthusiasm.

You are a space planning analyst building a shelf argument for a reset. I will paste share of shelf vs share of sales, days of supply, and any out-of-