← Back to the guideThe full worked example: Did last year's price increase really stick?
campo.
Field guide / Sales / UC-001 · SC-31
The result

"The 4% stuck"? Adjusted for segments, average net prices rose about 1.75%. The real price gain isn't verified yet.

A CRO at a 5,600-person packaging maker checks whether last year's price increase stuck before the board plans another, with any AI chat her company allows.

The resultNathalie's page for the CEO. Fictional company.
MIRELLE PACKAGING · REVENUE
Did last year's price increase stick?
For: Luc Bernier, CEO
From: Nathalie Gagnon · Tuesday, January 19

MY ANSWER

Not as the pricing summary says. In food & beverage, our largest segment, the average net price per tonne rose 1.5% while volume fell 13%, mostly with six customers who named price. Adjusted for segments, average net prices rose about 1.75% before $1.9M of FY26 rebates, but an average also rises if the customers who left paid less. Don't plan FY27 on another 3% on this evidence; keep it open until finance confirms rebates for both years and prices for the same customers and products. · Inputs 2, 3, 4

BY SEGMENT (AVERAGE NET INVOICE PRICE PER TONNE, BEFORE REBATES)

Food & beverage+1.5%
Volume down 13%; 6 customers left naming price
Personal care+2.0%
Volume up 25%
Industrial+2.0%
Volume up 5%
Rebates$1.9M
Not in the figures; FY25 rebates not given

WHAT'S BEHIND THE 4.1%, AND WHAT'S NEXT

MIX
More personal care, at about $3,000 a tonne, lifts the average
At last year's prices, this year's mix alone raises the average 2.3%. · Input 3
BY MON JAN 25
Finance: FY25 and FY26 rebates by segment, and prices for the same customers and products in both years
If FY25 rebates were zero, the 1.75% would fall to about 0.85%. · Input 3
FOR THE BOARD
Any FY27 increase decided by segment, with food & beverage volume shown next to it
· Inputs 3, 4
Caught on review
1
"Up 4.1%" is mostly mix: adjusted for segments, average net price rose about 1.75%
2
"Volume down only 2%" hides food & beverage down 13%
3
$1.9M of FY26 rebates aren't in the figures
UC-001 · SC-31 · Rev B
FOR
Chief Revenue Officer
SECTOR
Packaging manufacturing
COMPANY
5,600 people · 6 plants · Quebec and Ontario
TASK
Check whether last year's price increase stuck before the board plans another
WRITTEN FOR
ChatGPT · Claude · Gemini · Copilot
AI TEST
Claude, Nov 15, 2026 · 7 of 7 found
Step 0 · The mess

What landed on Nathalie's desk

A short email from the CEO, a confident summary from pricing, finance's net prices by segment and two notes from sales operations. The headline average is right. What it measures is the question.

EMAIL · CEO[1]
Check before the pre-read
“The board wants to plan another 3% for FY27 on the strength of it.”
1
The decision.
PRICING SUMMARY[2]
FY26 increase
“Average price per tonne … up 4.1%. The increase stuck.”
2
Price, or mix?
FINANCE[3]
Net price by segment
Food & beverage: 60,000 t at $2,000 … 52,000 t at $2,030.
3
Like for like.
SALES OPS[4]
Discounts and churn
“6 food & beverage customers left in FY26 and named price as the reason.”
4
Who paid for it.
You're not alone
Under half
of a planned price increase is what companies realise on average, according to Simon-Kucher's Global Pricing Study.
Simon-Kucher, Global Pricing Study 2025, more than 2,200 business leaders in 28 countries and 39 industries. Source
Steps 1–3 · How it works

Three steps, one chat

Here the AI checks a headline average that is arithmetically right. The prompt makes it split price from mix before it agrees anything went up.

1
Open a new chat

Use the AI tool your company approves: ChatGPT, Claude, Gemini or Copilot.

2
Paste the prompt and the inputs

Copy the prompt and the four fictional inputs from the appendix into one message.

3
Recalculate one segment yourself

Take food & beverage: $2,000 to $2,030 a tonne is 1.5%. Then decide what the board hears first.

The prompt, part by part
Copy-ready version in the appendix

I'm Chief Revenue Officer at a packaging manufacturer that sells to food, personal care and industrial customers. Our CEO wants my view on whether last year's 4% price increase "stuck", because the board wants to plan another 3% this year on the strength of it. The board pre-read goes out Thursday, January 28, and the board meets Thursday, February 4. Today is Tuesday, January 19. Below are the CEO's email, the pricing team's summary, finance's net price data by segment and two sales operations notes.

1
Who you are, what the board wants to decide, the pre-read and meeting dates, and today's date.

Please give me:

  1. Each claim in the pricing summary: holds, doesn't hold, or unverified, with the input that shows it.
  2. How much of the price increase we actually kept, by segment and overall, with your calculation.
  3. What the overall figures hide.
  4. What must be confirmed before the pre-read, and by whom.
  5. My recommendation to the CEO on the 3% plan, in five lines or fewer.
2
A verdict on each claim, how much of the increase you kept by segment, what the totals hide, what to confirm and by whom, and a five-line recommendation.

Separate price from mix: compare like for like before you say anything stuck. Check the arithmetic. Don't count anything twice. Don't add facts that aren't in the inputs; if you assume something, say so.

3
The heavy lifting: separate price from mix and compare like for like before saying anything stuck. It found all seven planted problems.
Where to paste
Every tool has one message box. Paste the prompt, add a blank line, paste the inputs, send. Price exports can be attached as files if your tool allows it.
Safe to try
Practise with the fictional inputs first. Real prices, rebates and customer names are confidential: use only your company's approved tool.
The AI's first answer

What Claude actually wrote

Exactly as it came back, not edited. Read it the way Nathalie did: check each point against the source it cites.

UNEDITED
Claude · fresh session, no other context · Nov 15, 2026 · one run

Short answer: the 4% did not stick. On a like-for-like basis, net invoice prices rose by about 1.7 to 1.8%, so we kept about 44% of the increase. That is before the $1.9M of FY26 rebates. Most of the reported 4.1% rise in average price per tonne comes from a mix shift towards personal care, which carries higher prices.


1. Claims in the pricing summary

Claim Verdict Evidence
List prices up 4.0%, effective Monday, February 2, 2026 Unverified No list price data in the inputs. Only the pricing summary [2] says this. Feb 2, 2026 is a Monday. Sales ops [4] counts exceptions from "February to December 2026", which suggests January FY26 sales were at old prices. That is an assumption: the fiscal year is not defined.
Average price $2,186 vs $2,100, up 4.1% Holds arithmetically From finance [3]: FY26 revenue $214.19M / 98,000 t = $2,185.6; FY25 $210.0M / 100,000 t = $2,100.0; +4.08%. found
"The increase stuck" Doesn't hold Like for like, every segment's net price rose by only 1.5 to 2.0% [3]. Most of the 4.1% is mix (see 2 and 3). found
Volume down only 2% (98,000 vs 100,000) Holds in total, misleading Food & beverage fell 13.3% (60,000 to 52,000 t). Personal care rose 25% and industrial 5% [3]. found

2. How much of the increase we kept (net invoice, before rebates)

Segment FY25 $/t FY26 $/t Change Share of 4% kept
Food & beverage 2,000 2,030 +$30, +1.5% 37.5%
Personal care 3,000 3,060 +$60, +2.0% 50%
Industrial 1,500 1,530 +$30, +2.0% 50%

Overall, like for like (FY26 tonnes at FY25 prices):

  • FY26 tonnes at FY25 prices: 52,000×2,000 + 25,000×3,000 + 21,000×1,500 = $210.5M
  • Actual FY26 revenue: 105.56 + 76.50 + 32.13 = $214.19M
  • Price effect: $3.69M, which is +1.75%. That is about 44% of 4%.
  • Cross-check at FY25 tonnes: 60,000×30 + 20,000×60 + 20,000×30 = $3.6M on $210M, +1.71%.

Splitting the $86/t average rise:

  • Mix: average price at FY26 mix and FY25 prices is $210.5M / 98,000 = $2,148, which is +2.3% on $2,100.
  • Price: $2,185.6 / $2,148 = +1.75%.
  • Check: 1.023 × 1.0175 = 1.041. So more than half of the "4.1%" is mix, not price.

Things I did not add or count twice:

  • Discount exceptions ($2.4M): already in the net prices [4], so I have not subtracted them again. found
  • January timing (assumption): if FY26 includes January at old prices, the most the year could have captured is about 4% × 11/12 = 3.67%. Against that ceiling we kept about 48%.
  • Rebates ($1.9M): not in these figures [3] and must not be ignored. That is about $19/t overall. If it were all in food & beverage it would be about $37/t on 52,000 t, more than the $30/t gain. FY25 rebates are not given, so I cannot work out an after-rebate comparison. found
  • What the rebates would mean, if FY25 rebates were zero: the price effect drops to $3.69M - $1.9M = $1.79M, about 0.85%. This depends entirely on the FY25 rebate figure.

3. What the overall figures hide

  • Mix: most of the "4.1%" comes from more personal-care tonnes at about $3,000/t and fewer food & beverage tonnes at about $2,000/t. found
  • Food & beverage volume: the segment lost 8,000 t (-13.3%). Six customers who left and named price account for 6,500 t of that, about 81% [4]. found
  • What that churn cost: at FY25 prices, the lost volume was about $13.0M of revenue. The food & beverage price gain on remaining volume is about $1.56M (52,000 × $30).
  • Concessions are concentrated in food & beverage: 31 of the 38 discount exceptions [4], most of the rebates [3], and the weakest realisation (1.5%).
  • Possible survivor effect (unknown): if the customers who left paid below-average prices, the remaining food & beverage average rises without any real price gain. Customer-level prices are not in the inputs. B: put it on the page
  • Source of personal-care growth: the inputs don't say whether the extra 5,000 t came from new or existing customers, or why.
  • Rebates: the figures exclude $1.9M of FY26 rebates. The headline therefore overstates what we kept, by an amount we can't pin down yet.

4. Confirm before the January 28 pre-read

What to confirm Who
FY25 rebates in total and by segment, and the final FY26 rebate by segment (the FY26 rebates are being paid now) Finance C: by Monday
Fiscal year dates, and whether January 2026 was billed at old prices Finance
That the 4.0% list increase applied to all segments and customers, and any exempted contracts or price-locked agreements Pricing
Churn list: tonnes, segments and FY25 prices of the six customers, and whether anyone left in other segments Customer success / Sales ops
What drove personal-care growth (+5,000 t) and the remaining 1,500 t food & beverage decline Sales ops / regional sales
The discount exceptions by segment in $, to confirm the $2.4M split Sales ops

5. Recommendation to the CEO (five lines)

  1. Don't present the 4% as having stuck. Like for like, net price rose about 1.75% before rebates, which is about 44% of the increase. A: lead with food & beverage
  2. The reported 4.1% is mostly a mix shift towards personal care, not price.
  3. Food & beverage kept only 1.5%, took most of the exceptions and rebates, and lost 13% of its volume, mostly to customers who named price.
  4. On the current evidence, don't plan FY27 revenue on 3% realised. Keep it open until finance confirms the after-rebate figures for both years, before January 28. found
  5. If the board still wants an increase, model it on realised net price by segment rather than the list price, and treat food & beverage volume risk as a separate decision.
foundcaught a planted problemA–CNathalie changes it; see the next section
The human part

What Nathalie changed, and why

Claude found every planted problem, split price from mix and noticed that FY25 rebates were missing too. It also flagged that the 1.75% might not be a real price gain. Nathalie's edits decide what the CEO reads first, put that limit on the page, and give finance a date.

What Claude found: 7 of 7 planted problems
One run
✓4.1% is mostly mix
✓Rebates left out
✓Food & beverage volume down 13%
✓Customers who left over price
✓One segment lifts the average
✓Discounts not counted twice
✓Don't plan 3% on this
Three edits, all judgment calls
A
CLAUDE WROTE
“Don't present the 4% as having stuck. Like for like, net price rose about 1.75% before rebates, which is about 44% of the increase.”
NATHALIE'S VERSION
Nathalie leads with food & beverage: in our largest segment the average net price per tonne rose 1.5% while volume fell 13%, mostly with customers who named price. The overall 1.75% comes second.
WHY · PRIORITY
The board is deciding on another increase. Where the last one hurt is what they need first, and an average hides it.
B
CLAUDE WROTE
“Possible survivor effect (unknown): if the customers who left paid below-average prices, the remaining food & beverage average rises without any real price gain.” One caveat among several.
NATHALIE'S VERSION
Nathalie puts it on the CEO's page next to the 1.75%, and asks finance for prices on the same customers and products in both years. The January detail stays on her own list.
WHY · EVIDENCE
The 1.75% adjusts for the shift between segments, not for which customers and products are inside each one. If the six who left paid less, the average rises with no real price gain. That limits what the headline proves, so the CEO needs to see it.
C
CLAUDE WROTE
“Keep it open until finance confirms the after-rebate figures for both years, before January 28.”
NATHALIE'S VERSION
Finance confirms FY25 and FY26 rebates by segment, and prices for the same customers and products, by Monday, January 25, three days before the pre-read goes out.
WHY · DATE
An answer that arrives the day the pre-read goes out is too late to rewrite the page.
What only you can do
Decide what the board reads first, keep any limit that changes what the headline proves on the page, and give finance a date that leaves time to act. The AI split price from mix. You decide what the board hears.
Your turn

Make it yours

Use it whenever someone reports that an average went up: price per unit, deal size, revenue per customer, margin per order.

Your prompt: fill in the brackets

I'm [your role] at a [industry] company that sells to [your segments]. [Who] wants my view on whether [the change] worked, because [the decision that depends on it]. [The deadline]. Today is [date]. Below are [the summary, the data by segment, and notes on discounts, rebates and customers lost].

Please give me: 1. Each claim in the summary: holds, doesn't hold, or unverified, with the input that shows it. 2. How much of [the change] we actually kept, by segment and overall, with your calculation. 3. What the overall figures hide. 4. What must be confirmed before [the deadline], and by whom. 5. My recommendation to [who], in five lines or fewer.

Separate price from mix: compare like for like before you say anything stuck. Check the arithmetic. Don't count anything twice. Don't add facts that aren't in the inputs; if you assume something, say so.

Five checks before you tell the board it worked
1
Each segment's change is shown on its own.
2
Mix is separated from price.
3
Rebates and discounts are in, and counted once.
4
Customers who left are named next to the volume.
5
Missing figures are named, with who supplies them and when.
Make it repeatable (optional)
Save the prompt once and reuse it after every price change: a custom GPT in ChatGPT, a project in Claude, a Gem in Gemini, or an agent in Copilot. Some companies switch these off; copy and paste always works.
With real data
Prices, rebates and customer names are confidential. Use only the AI tool your company approves, and follow its rules.
What this example shows
One run of this prompt in Claude found all seven problems planted in fictional inputs and noticed the missing FY25 rebates on its own.
What it doesn't show
That every tool, setting or run will do the same, how much time it saves you, or what price increase a real company should plan. Mirelle Packaging and its people are fictional.
Appendix

Copy-ready: the prompt and the four inputs

Copy the prompt, then all four inputs, into one message in a new chat. Everything here is fictional and safe to paste.

The prompt

I'm Chief Revenue Officer at a packaging manufacturer that sells to food, personal care and industrial customers. Our CEO wants my view on whether last year's 4% price increase "stuck", because the board wants to plan another 3% this year on the strength of it. The board pre-read goes out Thursday, January 28, and the board meets Thursday, February 4. Today is Tuesday, January 19. Below are the CEO's email, the pricing team's summary, finance's net price data by segment and two sales operations notes.

Please give me:

  1. Each claim in the pricing summary: holds, doesn't hold, or unverified, with the input that shows it.
  2. How much of the price increase we actually kept, by segment and overall, with your calculation.
  3. What the overall figures hide.
  4. What must be confirmed before the pre-read, and by whom.
  5. My recommendation to the CEO on the 3% plan, in five lines or fewer.

Separate price from mix: compare like for like before you say anything stuck. Check the arithmetic. Don't count anything twice. Don't add facts that aren't in the inputs; if you assume something, say so.

[1] Email from Luc Bernier, CEO

Nathalie, pricing says last year's 4% increase stuck. The board wants to plan another 3% for FY27 on the strength of it. Can you check it before the pre-read goes out on the 28th?

[2] Pricing team summary (FY26 price increase)
  • List prices up 4.0%, effective Monday, February 2, 2026.
  • Average price per tonne: $2,186 in FY26 against $2,100 in FY25, up 4.1%. The increase stuck.
  • Volume down only 2%: 98,000 tonnes against 100,000.
[3] Finance: net invoice price by segment (after discounts, before year-end rebates)
Segment FY25 tonnes FY25 $/tonne FY26 tonnes FY26 $/tonne
Food & beverage 60,000 2,000 52,000 2,030
Personal care 20,000 3,000 25,000 3,060
Industrial 20,000 1,500 21,000 1,530

FY26 year-end volume rebates of $1.9M, mostly food & beverage, are paid in January 2027 and are not in these figures.

[4] Sales operations notes
  • 38 discount exceptions were approved between February and December 2026, worth $2.4M in total; 31 were food & beverage. They are already in the net invoice prices above.
  • Customer success churn list: 6 food & beverage customers left in FY26 and named price as the reason. Their FY25 volume was 6,500 tonnes.