Skip to content

Discontinuation Impact Analyzer

Assess the risk of dropping a SKU before you decide to cut it.

What is the Discontinuation Impact Analyzer?

The Discontinuation Impact Analyzer is a free AI skill that assesses the risk of dropping specific SKUs before a delist decision gets made. You give it the candidate SKUs, their current performance, the surviving range, and what you know about their shopper base; it returns a cannibalization read on which surviving SKUs would likely absorb the volume, a halo-loss assessment for any shelf, traffic, or cross-category effect the SKU quietly provides, a shopper-defection risk score, and a go, hold, or investigate-further recommendation per SKU. It is built for category managers and portfolio planners weighing a cut who need more than a velocity ranking to decide. Because it forces every candidate through the same four lenses before a decision is final, it catches SKUs that look weak on sales but are quietly protecting shelf presence, a loyal segment, or a halo effect. Pairing the analysis with live food and beverage demand data helps confirm whether a SKU's low velocity reflects fading demand or a fixable distribution problem.

Who it's for

  • Category managers screening SKUs before a formal delist decision
  • Portfolio planners weighing a rationalization ahead of a reset
  • Brand owners deciding whether a legacy SKU still earns its space
  • Finance and category teams jointly reviewing a range-simplification list

What you get

  • A cannibalization read: which surviving SKUs would absorb the volume
  • A halo-loss assessment for shelf, traffic, or cross-category effects
  • A shopper-defection risk score per candidate SKU
  • A go, hold, or investigate-further call per SKU
  • The specific evidence needed before a final decision
  • A one-page risk summary for a leadership or finance audience

How to use it

  1. 1Paste the prompt into ChatGPT, Claude or Gemini.
  2. 2Fill the bracketed inputs (candidate SKUs, performance data, surviving range).
  3. 3Answer its clarifying questions about what you know versus don't know yet.
  4. 4Ask it to build the follow-up data-gathering list for any 'investigate further' SKUs.

See it in action

Example input

We're evaluating whether to drop 2 of 8 SKUs in our fictional 'Fernwood Bakery' artisan bread line at Coastal Basket — a Seeded Rye and a Rosemary Focaccia, both in the bottom quartile of velocity but each with a small, vocal following on social media. We need to know the real risk before we tell the sales team to prepare a delist.

Example output

Discontinuation impact analysis — Fernwood Bakery, Seeded Rye and Rosemary Focaccia, Coastal Basket:

Seeded Rye

Cannibalization read: Rye's closest surviving SKU is your Multigrain loaf, which shares the "hearty/dense" texture profile. Expect a moderate share of Rye's volume to transfer there. The remainder likely does not transfer within your range — rye is a distinct enough flavor profile that some buyers may simply stop buying Fernwood in this occasion rather than switch loaves. Assumption to validate against any SKU-switching data Coastal Basket can share.

Halo-loss assessment: low. Rye does not appear to anchor an end-cap or drive traffic beyond its own sales; nothing in the data provided suggests it pulls shoppers into the rest of your bread set.

Shopper-defection risk: moderate-to-low. Bottom-quartile velocity across 8 SKUs suggests a small buyer base, but "small and vocal on social" does not necessarily mean commercially significant — vocal does not equal volume. Recommend checking whether the social following correlates with actual repeat purchase data before weighting this risk higher.

Call: GO for delist, with a standard transition plan pointing displaced demand toward Multigrain.

Rosemary Focaccia

Cannibalization read: weaker internal transfer than Rye — focaccia's flat, herb-forward profile doesn't map cleanly onto any other SKU in your 8-item range. This is a higher risk of losing the volume outright rather than retaining it within your portfolio.

Halo-loss assessment: investigate further. Focaccia formats often anchor an "artisan/craft" positioning within a bread set even at modest volume — before cutting it, check whether Rosemary Focaccia is disproportionately represented in any bakery-department secondary placement, seasonal display, or entertaining-occasion merchandising Coastal Basket runs. If it is, cutting it may cost more than its own sales line suggests. This is the single most important thing to verify before deciding.

Shopper-defection risk: moderate. Focaccia buyers skew toward an entertaining/sharing occasion that is somewhat distinct from your everyday bread buyers — losing them may not show up as "bread category" defection but could reduce basket size on the occasions when they do shop Fernwood.

Call: INVESTIGATE FURTHER before deciding. Do not delist in the same cycle as Rye without first checking the secondary-placement and occasion questions above.

Evidence still needed
- SKU-switching or panel data showing where Rye and Focaccia buyers' spend goes when the item is unavailable, if any test markets or out-of-stock periods exist.
- Confirmation of whether Focaccia appears in any seasonal, entertaining, or secondary display program at Coastal Basket.
- A basic cross-check of the "vocal social following" against actual repeat-purchase records, so social visibility isn't mistaken for commercial importance in either direction.

One-page risk summary

Fernwood Bakery is evaluating two bottom-quartile SKUs for delist: Seeded Rye and Rosemary Focaccia. Rye shows a clear internal transfer path to Multigrain and low halo risk — it is a reasonably safe cut once a transition plan is in place. Focaccia carries more uncertainty: its volume is less likely to transfer within the range, and it may be quietly supporting seasonal or entertaining-occasion merchandising that isn't visible in velocity data alone. Recommend proceeding with Rye on the standard delist track while holding Focaccia for one more review cycle, pending the secondary-placement check above.

Want me to turn the Rye decision into a full delist transition plan once you're ready to execute it?

The prompt

Here's the start of the prompt. Download the free bundle for the full, ready-to-paste version — plus the installable Claude Skill and Custom-GPT instructions.

# Role
You are a senior category portfolio analyst who screens SKU discontinuation risk before a delist decision is final. You look past velocity to cannibalization, halo effects, and shopper defection, and refuse to wave through a cut just because a SKU sits in the bottom quartile.

# Context I'll provide
- Candidate SKUs for discontinuation: [CANDIDATE SKUS]
- Current performance data: [PERFORMANCE DATA — velocity, sales, whatever you have]
- Surviving range these SKUs would exit from: [SURVIVING RANGE]
- Shopper base notes per candidate (optional): [SHOPPER NOTES]
- Known secondary roles, e.g. display or seasonal (optional): [SECONDARY ROLE NOTES]
- The decision this feeds and its timing: [DECISION CONTEXT]

# Your task

Frequently asked questions

What is a discontinuation impact analysis?
A discontinuation impact analysis assesses what happens if a specific SKU is dropped, before the decision is finalized: whether its volume would transfer to other SKUs in the range, whether it quietly supports traffic or halo effects beyond its own sales, and how likely its shoppers are to defect from the brand entirely. This skill runs that analysis per candidate SKU and returns a go, hold, or investigate-further call for each.
How is this different from the Delist Transition Planner skill?
The Delist Transition Planner assumes a SKU is already being cut and plans the execution: run-down timeline, inventory clearance, shopper transition, and retailer communication. This skill comes before that decision — it assesses whether and which SKUs should be cut at all, weighing cannibalization, halo loss, and defection risk. Use this skill first to decide, then the Delist Transition Planner once a cut is confirmed, to manage the exit.
Which AI models can run this prompt?
Any capable chat model — ChatGPT, Claude, or Google Gemini. The prompt is model-agnostic, so paste it into a chat, save it as a Custom GPT, or store it as a reusable skill so every discontinuation candidate gets screened through the same four-lens analysis before it reaches a delist conversation.
What if I don't have panel or switching data?
Provide whatever performance data you have — even basic velocity and facings — and describe what you know about each SKU's shopper base and any secondary merchandising roles. The skill will build the analysis around that and explicitly flag which candidate SKUs need real switching or panel data before a confident call can be made, rather than guessing at transfer rates on your behalf.

Related skills

Want the live data behind sharper outputs?

These skills get better with real-time F&B intelligence. See what Tastewise can do for your team.