Move guide
Size, test and make price moves
Your report suggested a move on at least one of these four practices:
- how price elasticity is understood;
- the analysis actually used;
- how price increases are executed;
- how fast a price decision reaches the market.
This move guide sizes every major price move before it is made, pilots it where it can, and measures it after. It also makes sure an agreed price reaches the invoice in weeks.
If you took the consumer check, go to the consumer section. If your products reach shoppers through retailers, go to the section for brands.
1. When it applies
Use this move guide if your report placed elasticity or the speed of price changes in the lowest two stages. Use it too if it placed the analysis you use or your price increases at the middle stage.
| Practice | Where it may sit today | Where this move guide aims |
|---|---|---|
| How price elasticity is understood | Lagging: "Nobody can say what volume a 5% price move would cost or win; the argument is settled by whoever fears the customer most." Basic: "Elasticity is a regression of our own price against our own volume, and it says whatever the last two years happened to say." |
Competent: "Elasticity is estimated from granular history against a stated reference price (what the customer last paid us, or the nearest comparable competitor's price), by segment." |
| The analysis actually used | Competent: "Standard analyses (price bands by customer size, discount distribution, margin bridge) run on a calendar." | Advanced: "Analysis reaches elasticity, win/loss by price level, and scenario modelling of moves before they are made." |
| How price increases are executed | Competent: "Realisation of each increase is measured, and accounts that escaped are followed up." | Advanced: "Increases are differentiated by segment and account, with talking points and escalation paths prepared for sales." |
| How fast a price decision reaches the market | Lagging: "Changing a price is a project: systems, spreadsheets and sign-offs mean months pass before the market sees it." Basic: "Some elements of the price mix can move quickly; others are effectively frozen by systems." |
Competent: "Agreed price changes reach quotes and invoices within weeks, accurately." |
If the analysis you use or your price increases sit in the lowest two stages, start with reporting price performance every month. Measuring what stuck comes before shaping the next increase.
2. What to do
Seven steps. Each major move takes six to twelve weeks, from model to rollout.
- State the reference price for each segment. It is the price the customer compares you with: what they last paid you, or the nearest competitor's price.
- Estimate elasticity for your ten biggest products. Elasticity is how much volume changes when price changes. Use two to three years of history, against the reference price. The elasticity calculator makes the estimate from your sales history. Have your sales managers review the estimates.
- Model each move. Work out the volume and margin effect, the likely customer and competitor response, and the volume you can lose before margin falls. The breakeven calculator and the 1% calculator check the sums.
- Build a set of moves. Spread them across products with different elasticities. Make segment-level moves before finer ones. Table A below does this.
- Pilot before rollout. Use friendly customers or one product line, and compare with a group that did not change.
- Differentiate the next increase. Vary it by segment and account. Give sales talking points and an escalation path. Measure what stuck, account by account. The price rise calculator shows how much of a rise you bank.
- Time how long a decision takes to reach an invoice. Remove the two slowest steps.
The four free worksheets do the recording:
- /worksheets/pricing-b2b/price-elasticity for the reference price, the estimates and one experiment;
- /worksheets/pricing-b2b/analytics for the standard analyses this builds on;
- /worksheets/pricing-b2b/price-increases for realisation and the plan by segment;
- /worksheets/pricing-b2b/price-agility for timing a change from agreement to invoice.
What the calculator gives you
Each reading uses the same months in a different way. The calculator shows the levels reading as its headline, and the range the others gave. It shows no figure when the history cannot carry one.
3. What to produce
One table, plus the four worksheets above.
Table A. The move plan
| Move | Products or segment | Elasticity estimate | Volume it can lose before margin falls (%) | Pilot: who and how long | Comparison group | How what stuck will be measured | Decision date |
|---|---|---|---|---|---|---|---|
4. A worked example
Illustrative. Written for this move guide. It describes no client.
A distributor of commercial vehicle parts sells to fleets and independent workshops.
The estimates.
- The reference price for workshops was what they last paid. For fleets it was the main competitor's contract price.
- Elasticity for the ten biggest product lines came from three years of history. Sales managers reviewed each estimate.
- Filters and brake parts responded little to price. Tyres responded strongly.
The moves.
- A rise on filters and brake parts for workshops, a hold on tyres, and no change for fleets until their contracts renewed.
- The rise ran first in one region for six weeks, with a neighbouring region as the comparison.
- Sales had talking points for the rise and an escalation path for any account that pushed back.
The speed.
- The last price decision had taken nine weeks to reach invoices.
- The two slowest steps were a manual price-file upload and a second sign-off that added nothing. Both were removed.
- The next change reached invoices in three weeks.
After the pilot.
- What stuck was measured account by account, and the move was rolled out region by region.
On a retake, all four practices could be answered at the target stage.
What it looked like in our work
From our own work. Figures changed.
An equipment maker worked out how much volume two product lines could lose after a price rise before their margin fell.
| Price rise | Line A, about 35% contribution margin | Line B, about 45% contribution margin |
|---|---|---|
| 5% | 12.5% of volume | 10% of volume |
| 10% | 22% of volume | 18% of volume |
| 15% | 30% of volume | 25% of volume |
Its elasticity estimates said the least price-sensitive products would lose far less than that. The review proposed rises on those products, and price cuts on the most price-sensitive ones where it sat above competitors.
How an elasticity model works
Illustrative, on made-up data.
A pricing model can take a monthly history to a price decision in eight steps. Here they are, run on the calculator's made-up example.
- 1. Gather the historyOne row a month: your volume and price, a competitor's price, and the market's size where you have it.
- 2. Make two measuresA volume measure and a price measure for each month, in logs, so a 1% move reads the same at any level.
- 3. Clean themTake out each calendar month's usual level and any slow trend. Leave out months that were not normal.
- 4. Fit the lineDraw the straight line closest to the months. Its slope is the elasticity.
- 5. Check itRead the 80% range, and check that the other readings of the same months agree.
- 6. Judge what the data cannotWhether a competitor matches you, whether the market shrinks, and how long lost volume stays lost.
- 7. Set it against the marginA small rise pays while the elasticity is below 1 divided by the contribution margin on net revenue.
- 8. Keep it currentMove the price, add the new months, and estimate it again.
Two measures, month by month. Each month gives one point: the change in the price measure across, the change in the volume measure up.
| Month | Your price ÷ competitor's | Change in the price measure | Your volume | Change in the volume measure |
|---|---|---|---|---|
| 4 | 0.96 | none | 1,176 | +5.2% |
| 5 | 1.01 | +4.9% | 1,090 | -7.6% |
| 10 | 0.98 | -3.0% | 954 | -0.7% |
| 12 | 1.02 | +3.9% | 969 | +3.8% |
In month 12 the price measure rose and volume rose too, because that month always sells more. Step 3 takes that out before the line is fitted.
The estimate beside the break-even.
- 80% range: 1.40 to 1.90
- Estimate: 1.60
- Break-even at a 40% margin: 2.50
The bar is the estimate's 80% range. The line is the break-even elasticity, 1 divided by the margin.
The whole range sits below the break-even. On these made-up figures, a small rise would earn more on the units kept than it would lose on the units that went.
- 1. EstimateOn the months so far.
- 2. DecideAgainst the break-even at your margin.
- 3. Move the priceWithin about 20% of the prices you measured on.
- 4. Add the monthsAdd the months after the move, then estimate again.
The elasticity calculator runs steps 1 to 5 on your own history.
A price test against a control group
From our own work. Figures changed.
An insurer tested two ways to reprice its renewals. The control group got the usual rise, with about half of its customers at one rate. The test group got a price for each customer, set by how sensitive that customer was to price.
- +$1.5m Control group: yearly profit gain, with 1,900 customers lost
- +$0.5m Test group: yearly profit gain, with no customers lost
- About $790 Cost to win back each lost customer
- About $0 Control group's gain after paying to win those customers back
The control group looked better on profit alone. Counted at the same number of customers, the test group was expected to come out about $0.5m a year ahead.
Will the rival follow your rise?
Illustrative. The payoff logic is standard game theory.
A rise holds when the main rival follows it. Whether it follows depends on its own economics, above all its spare capacity. Draw your move and the rival's reply, then work backwards from the rival's side.
- We decide whether to raise price 5%
- Raise
- Rival follows4, 4
- Rival holds2, 3
- Hold
- Rival holds3, 3
- Raise
Our profit first, the rival's second, in made-up units. Start from the rival's replies. The rival cannot serve the customers our rise sends it, so it follows. Rising pays us.
- We decide whether to raise price 5%
- Raise
- Rival follows4, 4
- Rival holds1, 6
- Hold
- Rival holds3, 3
- Raise
Our profit first, the rival's second, in made-up units. The rival can fill its plant with the customers our rise sends it, so it holds. Holding pays us more.
The move is the same in both. Only the rival's plant changed. So when you model each move (step 3 above), put in the main rival's capacity and costs beside your customers' elasticity.
Before a large rise, put a small team in the rival's seat with its costs, its capacity and its share at stake. The war game worksheet runs three rounds of move and reply. Then rerun the move plan with the rival's most likely reply in it. Read the rival's position from customers, tenders and published lists. Decide the move alone, and never discuss it with a competitor.
Size the move before you make it
Illustrative. Dummy figures.
Model what each price would earn against holding today's. Do it twice: once if the competitor holds its price, and once if it follows.
| Our price, per unit | Contribution against today |
|---|---|
| $94 | −$42k |
| $95 | −$32.5k |
| $96 | −$24k |
| $97 | −$16.5k |
| $98 | −$10k |
| $99 | −$4.5k |
| $100 | $0k |
| $101 | $3.5k |
| $102 | $6k |
| $103 | $7.5k |
| $104 | $8k |
| $105 | $7.5k |
| $106 | $6k |
| $107 | $3.5k |
| $108 | $0k |
| $109 | −$4.5k |
| $110 | −$10k |
| Our price, per unit | Contribution against today |
|---|---|
| $94 | −$61.2k |
| $95 | −$48k |
| $96 | −$36k |
| $97 | −$25.2k |
| $98 | −$15.6k |
| $99 | −$7.2k |
| $100 | $0k |
| $101 | $6k |
| $102 | $10.8k |
| $103 | $14.4k |
| $104 | $16.8k |
| $105 | $18k |
| $106 | $18k |
| $107 | $16.8k |
| $108 | $14.4k |
| $109 | $10.8k |
| $110 | $6k |
- Untested: $106 to $110
- Today: $100
- Best if it holds
- Best if it follows
A business selling $1m a month at a 40% margin. Dummy figures.
If the competitor holds, the best price here is about 4% up. If it follows, a larger rise pays. Past the prices your evidence covers, the curve is a guess.
Then test how sure you need to be. A rise pays only while volume falls less than its break-even.
| Volume lost for each 1% on price | Contribution from the rise |
|---|---|
| 0.5% | $38.7k |
| 1% | $27.5k |
| 1.5% | $16.2k |
| 2% | $5k |
| 2.5% | −$6.3k |
| 3% | −$17.5k |
| 3.5% | −$28.8k |
| 4% | −$40k |
| 4.5% | −$51.3k |
| 5% | −$62.5k |
| Volume lost for each 1% on price | Contribution from the rise |
|---|---|
| 0.5% | $44.4k |
| 1% | $38.7k |
| 1.5% | $33.1k |
| 2% | $27.5k |
| 2.5% | $21.9k |
| 3% | $16.2k |
| 3.5% | $10.6k |
| 4% | $5k |
| 4.5% | −$0.6k |
| 5% | −$6.3k |
- Our estimate: 1.8%
- Break-even if it holds
- Break-even if it follows
The same business. Dummy figures.
Here the estimate sits below both break-evens, so the rise pays whether or not the competitor follows.
Read the slope from the months that moved
Illustrative. Dummy figures.
Set each month's price change against its volume change. The months with a price move set the slope. The months with none show the noise.
| Point | Price change in the month | Volume change in the month |
|---|---|---|
| −5% | 8.5% | |
| −4% | 5.9% | |
| −3% | 6.1% | |
| −2% | 2.4% | |
| −1% | 2.9% | |
| 1% | −0.8% | |
| 2% | −4.1% | |
| 3% | −4.2% | |
| 4% | −7.4% | |
| 5% | −7.1% | |
| 6% | −10.5% | |
| 7% | −10.6% | |
| 0% | 1.2% | |
| 0% | −1.5% | |
| 0% | 0.4% | |
| 0% | −0.6% |
- Slope: −1.6: from −6%, 9.6% to 8%, −12.8%
Large dots are months the price moved. Small grey dots are months it did not.
Here, volume fell about 1.6% for each 1% on price. That is the figure the response curve above is built on.
5. Effort and dependencies
- AnalystBuilds the estimates and the models.
- Pricing forumDecides each move.
- SalesCarries the increase to customers.
- ITFixes the slow steps in the systems.
What you need
- Two to three years of price and volume history from one source of pricing data.
- Competitor prices from learning what customers value and competitors charge.
- The monthly measure of what stuck from the monthly price report.
How long
- Six to twelve weeks for each major move.
What gets in the way. We have seen these in our own work:
- Fear of volume loss. An equipment maker built elasticities from three years of sales, reviewed by its sales managers. On its least price-sensitive lines, a rise held margin even after a sizeable volume loss.
- One number for the whole market. An insurer found elasticity varied widely by segment once competitor prices were modelled. A single market figure would have hidden that.
- Moving everything at once. A retailer staged its moves. Segment-level moves came first and finer ones later, each round spread across products of different elasticity.
- Trusting survey elasticity. Our own process notes found survey methods give a rough figure. Volumes are usually too low for them to replace real tests.
What we have seen. "In the businesses we have worked with, a 1% gain in realised price was worth about twice a 1% gain in volume or in cost, because it falls straight to the bottom line." We have seen this in our own work. It is not a forecast for yours.
6. How will you know it worked?
The test is whether the next major move is sized, piloted and measured. Retake the check after it. Each of these should be true when you answer:
| Practice | You can answer the target cell when | A document alone does not count if |
|---|---|---|
| How price elasticity is understood | Elasticity is estimated by segment from detailed history, against a stated reference price. | The estimate is your own price plotted against your own volume. |
| The analysis actually used | Your analysis reaches elasticity, win and loss by price level, and a model of each move before it is made. | The model was built after the move. |
| How price increases are executed | The last increase varied by segment and account, and sales had talking points and an escalation path. | The increase was the same for everyone, whatever the plan said. |
| How fast a price decision reaches the market | Agreed price changes reach quotes and invoices within weeks, accurately. | The slow steps were mapped but are still there. |
For consumer businesses
This section is for the B2C Consumer Pricing Health Check. Here moves are estimated by product and channel, tested in stores, regions or groups of customers, and landed on a repricing calendar.
When it applies
Use it if your report placed any of these three practices in the lowest two stages.
| Practice | Where it may sit today | Where this move guide aims |
|---|---|---|
| How price elasticity is understood | Lagging: "Nobody can say what a 5% price move would do to units; the debate is settled by nerves." Basic: "Elasticity is our own price plotted against our own units, confounded by season, promotions and whatever competitors did." |
Competent: "Elasticity is estimated from granular history against a stated reference price (the price the customer last paid, or the nearest comparable competitor's), by product and channel." |
| Whether prices are tested | Lagging: "Price changes are all-or-nothing bets argued from opinion." Basic: "We look at what happened after changes, confounded by season, weather and marketing." |
Competent: "Meaningful changes are piloted (stores, regions, cohorts) before rollout when feasible." |
| How price changes reach the shelf/site | Lagging: "Repricing is a manual scramble; wrong prices survive for weeks." Basic: "Changes go out in batches when someone has time, with occasional misses between channels." |
Competent: "A repricing calendar and checklist exist; changes land accurately and on time." |
What to do
- State the reference price for each of your twenty highest-volume products. It is the price the customer last paid, or the nearest competitor's.
- Estimate elasticity against it, by product and channel, from your sales history. The elasticity calculator does the sums.
- Run one clean test this quarter. Change the price in some stores, regions or customer groups, and keep others as the comparison. Roll out only what the test supports. Table B below plans it.
- Put repricing on a calendar with a checklist. After each change, check that every channel shows the new price.
The three free worksheets do the recording:
- /worksheets/pricing-b2c-direct/price-elasticity for the reference prices and the estimates;
- /worksheets/pricing-b2c-direct/testing for the hypothesis, the design and the decision;
- /worksheets/pricing-b2c-direct/change-execution for the calendar and the accuracy check.
What to produce
Table B. The test plan
| Price change | Test stores, regions or customer groups | Comparison group | Weeks | What will be measured | Decision rule |
|---|---|---|---|---|---|
A worked example
Illustrative. Written for this move guide. It describes no client.
A cinema chain runs 25 sites.
- It estimated elasticity for tickets and snacks by site type, against what customers paid last time.
- It tested a lower midweek ticket price in four cinemas, with four similar cinemas as the comparison, for eight weeks.
- Midweek attendance rose in the test cinemas, but not by enough to cover the lower price. The change was not rolled out.
- Ticket and snack prices now changed on a monthly calendar, with a check that the app, the website and the box office matched.
On a retake, all three practices could be answered at the middle stage.
Effort and dependencies
- Who. An analyst models. The monthly pricing meeting decides. Store operations runs the test. E-commerce runs the accuracy check.
- What you need. Sales history by product, channel and week, with promotions flagged.
- How long. Six to twelve weeks for each test, from design to decision.
- What gets in the way. We have seen price cuts run into demand that does not respond. A consumer business's promotions needed far more extra turnover to break even than they delivered.
What we have seen. "For one consumer travel business we sized a 2% rise in realised price at about $45m of EBIT, more than twice the value of 2% more customers. Price falls straight to the bottom line." We have seen this in our own work. It is not a forecast for yours.
How will you know it worked?
| Practice | You can answer Competent when | A document alone does not count if |
|---|---|---|
| How price elasticity is understood | Elasticity is estimated by product and channel from detailed history, against a stated reference price. | The estimate is your own price against your own units, with season and promotions mixed in. |
| Whether prices are tested | Your last meaningful price change was piloted in some stores, regions or customer groups before it was rolled out. | The test was designed but the change went out everywhere first. |
| How price changes reach the shelf/site | Repricing runs on a calendar with a checklist, and changes land accurately and on time. | The calendar exists, but wrong prices still survive for weeks. |
For brands selling through retailers
This section is for the B2C via trade Brand Pricing Health Check. For a brand, a price move means estimating what shoppers will do, and taking a written case to the retailer before input costs force it.
When it applies
Use it if your report placed either practice in the lowest two stages.
| Practice | Where it may sit today | Where this move guide aims |
|---|---|---|
| How price elasticity is understood | Lagging: "Nobody can say what a price-point move would do to volume; the buyer's reaction is the only forecast." Basic: "Elasticity is our shelf price plotted against scan volume, confounded by promotions and the retailer's own moves." |
Competent: "Elasticity is estimated from scan data, or invoice history where there is none, against a stated reference price (the last price paid, or the nearest comparable competitor or own-brand price), by pack and channel." |
| Whether cost-driven price increases are scheduled, argued and realised | Lagging: "Input costs rise and we absorb them until the year is lost, and when we do ask the retailer, we ask without a case." Basic: "Increases go in when the retailer allows, often with a promotion given back to get them through." |
Competent: "A quarterly input-cost review decides whether and when to move, and the case to the retailer is written before the meeting." |
What to do
- Name the reference price for your top pack. It is what the shopper last paid, or the nearest competitor or own-brand price.
- Estimate elasticity for your ten biggest items against it, by pack and channel. Elasticity is how much volume changes when price changes. Use scan data, or invoice history if you have none. Do it before the next price-point review.
- Decide which store clusters deserve a controlled test.
- Review input costs every quarter. Keep a one-page commodity scoreboard, and decide whether and when to move. Table C below does this.
- Write the case for the retailer before you need it. Take it to the meeting with the increase.
The two free worksheets do the recording:
- /worksheets/pricing-consumer-trade/price-elasticity for the reference price and the estimates;
- /worksheets/pricing-consumer-trade/price-increase-execution for the scoreboard, the quarterly decision and the case.
The price rise calculator shows how much of a rise you bank.
What to produce
Table C. The commodity scoreboard
| Input | Share of product cost (%) | Change this quarter (%) | Effect on cost per case ($) | Level that triggers a move |
|---|---|---|---|---|
A worked example
Illustrative. Written for this move guide. It describes no client.
A pasta sauce maker sells through the two big grocers and independents.
- Tomato and glass jar costs had risen for three quarters. The business absorbed them until it was too late to recover the year.
- A quarterly scoreboard now tracked its five main inputs against a trigger level for each.
- When the trigger was reached, the commercial manager wrote the case for the retailer before the meeting. It set out the cost movement, the proposed list change and its effect on the shelf price.
- Elasticity for the ten biggest items came from two years of scan data, against the main competitor's price. The two least price-sensitive items carried most of the increase.
On a retake, both practices could be answered at the middle stage.
Effort and dependencies
- Who. An analyst builds the estimates and the scoreboard. The monthly pricing meeting decides. The account managers take the case to the retailers.
- What you need. Two to three years of scan data or invoice history, and input costs by quarter.
- How long. Six to eight weeks for the first estimates. Then a quarterly cost review.
- What gets in the way. An increase asked for without a case. Write it before the meeting, with the cost evidence and the effect on the shelf price.
How will you know it worked?
| Practice | You can answer Competent when | A document alone does not count if |
|---|---|---|
| How price elasticity is understood | Elasticity is estimated from scan data, or invoice history, against a stated reference price, by pack and channel. | The estimate is your shelf price plotted against scan volume, with promotions mixed in. |
| Whether cost-driven price increases are scheduled, argued and realised | A quarterly input-cost review decides whether and when to move, and the case to the retailer is written before the meeting. | The scoreboard exists, but the last increase was asked for without a case. |
The worked examples are illustrative.
Think we've used your work without proper credit? Tell us and we'll fix it.