Why is "Reporting 'ROI' on total volume instead of incremental volume" a common mistake when using the Promo ROI Calculator?

Symptom: The headline read '+150% ROI' on a 20% TPR. Six months later the post-event analysis was rerun, this time separating the genuinely new volume from the volume that was coming anyway. The same event came back with a negative return and a recommendation to cancel it. The underlying sales data never changed. What changed was the top half of the calculation, the profit, and the bottom half, the cost it divides by. That alone turned a headline win into a loss. Fix: The return is **net incremental profit over everything that left the business**, and both halves have to be right. The top half counts only genuinely new volume, valued at the REGULAR gross profit per unit, the profit you make with no deal on the pack. The bottom half carries three things: the discount funded on EVERY promoted unit, any fee money you paid the retailer, and the margin the post-promotion dip takes back. Counting only the discount on the units that were not new leaves out the rest. The default 20% / 2.0x scenario here: -$15,540 over $20,980 is -74.1%, which is what the event actually returned. Leave the discount on the new units out of the bottom half, and the events most worth fixing look better than they are. Those are the ones where most of the extra volume is genuinely new. And the old shortcut of multiplying the sales spike by the margin per unit overstates the return by a factor of two to five.

Why is "Confusing volume uplift with profit success" a common mistake when using the Promo ROI Calculator?

Symptom: Sales celebrated a 2.5x uplift on the Q2 event. Trade marketing celebrated the sell-in number being hit. Three months later the bridge showed about -$40K of net profit per event. The net share of uplift had come in at 25%, not the 50% the plan assumed. And the 35% cut gave away too much margin to loyal buyers who were coming anyway. Fix: Volume uplift is **necessary and nowhere near sufficient**. The grid needs the money back AND more than half the uplift genuinely new, a net share of uplift (net incremental volume / uplift) above 50, and uplift on its own answers neither. A 2.5x uplift at a 25% net share of uplift is one to cancel. The same 2.5x at 60%, on a shallower cut, is a template worth running again elsewhere. Pair the uplift number with the net share of uplift estimate before anyone celebrates.

Why is "Ignoring the post-promo dip in the bridge" a common mistake when using the Promo ROI Calculator?

Symptom: The promo-week analysis showed +$8K net profit. The retailer called it a win. Looking 4 weeks past the event, the number was -$15K. The shoppers who had stocked up stayed away for 3 weeks afterwards, and nobody had built that loss into the bridge in the first place. Fix: **Always include Post-Promo Dip x Dip Duration** in the bridge. Heavy-promotion categories (biscuits, snacks, canned goods, cleaning) see 15-25% dips lasting 1-3 weeks. Light-promotion categories see 5-10% lasting 1 week. The default in this calculator, a 15% dip lasting 1 week, is the middle of the range. Plenty of real events come out worse. Before you sign a trade plan with deep cuts in it, go back to your scanner data and check what sales actually did in the weeks after past promotions.

Why is "Comparing depth and uplift in isolation, without incrementality" a common mistake when using the Promo ROI Calculator?

Symptom: The trade plan set a shallow cut with low uplift against a deep cut with high uplift, and chose the deep one because it shipped more total volume. The bridge, run after the event, showed the shallow template delivered 2-3x the net profit. It ran at a 60% net share of uplift because the deal only paid on a multi-buy, against 25% for the deep template, which was a straight price cut. Fix: **The share of genuinely new volume is a lever, not a constant.** How you design the promotion lifts the net share of uplift (net incremental volume / uplift) from the 35-45% an untargeted price cut typically earns to 50-65%, and the best-built events reach 70 or more. On a pack that earns $1.36 at full price, a price promotion only pays near the top of that range, which is why so few price promotions on a pack like this one pay. What does the work is a condition the shopper has to meet. A multi-buy, an offer that rewards a bigger basket, a slot aimed at shoppers who have never bought you, or a deal with a deadline on it. In this calculator, a small price cut with a high uplift and a high net share of uplift (8% / 2.5x / 70%) consistently beats a deep cut with only a moderate uplift. Depth costs you margin on every pack sold on deal. Uplift only pays back on the share of it that is genuinely new.

Why is "Repeating losing templates because the cosmetic ROI metric looks 'okay'" a common mistake when using the Promo ROI Calculator?

Symptom: The retailer scorecard showed an average return near break-even and the Customer Marketing team read it as 'roughly fine, optimise around the edges'. Commercial Finance ran the bridge across the calendar and found the average was hiding two very different groups. Most events lost money. A thin group returned very well. The average sat between them and described neither. The losing events were destroying enough to cancel out everything the strong ones earned. Fix: Track **distribution, not average**, because an average is the wrong statistic for a shape like this. Every quarter, summarize the calendar on the Performance Grid, and give every event its verdict: Scale, Hold, Fix or Cut. Treat the events to cancel as a list, rather than as a percentage. In the worked calendar on this lesson, 59% of events lose money, 18% pay back thinly, and 23% pay back properly, with a mean of +0.7% that would have told you to leave the whole thing alone. Read the verdicts and the medians. The Net Incremental Profit Bridge diagnoses one event at a time. The Performance Grid is where you decide what to do with all your events together.

Free tool · no login required

Promo ROI Calculator

Build the Net Incremental Profit Bridge for any trade promotion in fast-moving consumer goods (FMCG). The bridge starts with the gross profit your genuinely new units earned, then takes off everything that left the business to get them. Read the Performance Grid verdict, Scale, Hold, Fix or Cut, before you sign the trade plan. Same interactive model as Trade Promotion Optimization Lesson 1 in the full RGM Academy course. It is free to use and there is no sign-in.

Updated 13 October 2026Extracted from the Trade Promotion Optimization module, lesson 1: Promo ROI Fundamentals

Loading calculator...

Guided walkthrough

Explore the tool, from setup to common mistakes

Five short sections explain the scenario, what each control does, how to read the output, and the mistakes to avoid. Open whichever helps. The tool above works without them.

5.1Scenario setup

The starting SKU, market, and assumptions the model makes.

You're a Customer Marketing Manager working a national grocery account on a Mainstream Biscuits 250g SKU. It sits on the shelf at $4.99, a price the retailer sets. The retailer keeps about 25 percent of that ($1.25) and pays you $3.74, and your everyday trade terms (7.5 percent, or $0.28) leave you $3.46 of Net Sales a pack. The pack costs $2.10 to make, so a full‑price pack earns you $1.36 of gross profit, your full margin, which is 39.3 percent of your Net Sales. Both percentages are planning assumptions, so put in your own. Baseline sales are 5,000 units/week. The retailer wants a Q3 trade plan. They're pushing for a 20% TPR (temporary price reduction) running 2 weeks, expecting a 2.0x uplift. You've seen this template before, and a template here just means one set of promotion settings: how deep the cut goes, how much uplift it buys, and how long it runs. The Sales team loves this one because the volume hits the sell‑in number. Finance wants to know whether it actually creates profit.

Your job is to say whether this promotion makes money, in time for Thursday's Joint Business Plan (JBP) meeting with the account.

Start by building the Net Incremental Profit Bridge for the proposed event. It is a line‑by‑line calculation that starts with the extra volume the deal sells and ends with the profit left once everything the deal cost you has been taken out. Then read where the event lands on the Performance Grid, which sorts events into four verdicts: Scale, Hold, Fix or Cut. Come back with a green light, or with a counter‑proposal you have rebuilt.

Your objective

Use the calculator to work out the Net Incremental Profit of the promotion the retailer has proposed. Then test an alternative against the same two questions the grid asks: did the money come back, and was more than half the uplift genuinely new. Three alternatives are worth building. One cuts the price less deeply. One sells more volume for the same depth of cut. And one attaches a condition the shopper has to meet, such as buying several packs together, which raises the share of the extra volume that is genuinely new.

Key assumptions

The calculator models a single SKU x single retailer event in isolation, and it leaves out three things. It does not count the volume this promotion takes from your own other packs, which is cannibalization. It does not count a competitor running a promotion in the same weeks. And it gives no credit for a halo, the pull‑through a promoted pack gives the packs of yours that are not on deal. Those cross‑SKU effects sit in the TPO Lesson 2 Source‑of‑Volume lesson.
Net share of uplift is net incremental volume / uplift: of the extra units above baseline, the share that is genuinely new demand. It is not a share of everything sold on deal. That share is incrementality (net incremental volume / promoted volume), which the calculator shows beside it and which is smaller whenever the uplift is above 1.0x. The rest of promoted volume is units not new: the subsidized base (loyal buyers who would have bought at full price but now enjoy the discount) plus the part of the uplift that only moved from your other packs or another week. The split is the single most consequential assumption in the entire model.
Who pays for the cut. The calculator assumes the retailer keeps its usual cash margin on every pack sold on deal, so you fund the whole cut the shopper sees. At 20 percent off the $4.99 shelf price that is about $1.00 of your money a pack, most of the $1.36 a full‑price pack earns you. Some retailers share the cost. If this one kept its percentage of the lower deal price instead, you would fund about three quarters of the cut, and the default event (2.0x uplift, 40 on the slider) would break even at a cut of about 5.9 percent instead of 4.4.
Margin dilution = units not new x (full margin - promo margin). This is the cost of the discount on the volume you'd have sold anyway. It's a real P&L cost; the retailer just doesn't see it on their side.
Post‑promo dip is the volume reduction below baseline AFTER the event ends, because pantry‑loaded shoppers stay away. Where a category is promoted hard and often, sales run 15‑25% below normal for 1‑3 weeks after the event ends. Where it is promoted lightly, the dip is 5‑10% and lasts 1 week. Multiply that by Dip Duration to get the total volume the event pushed into the future. The calculator's dip settings describe a 2‑week event, and it grows the dip in line with the weeks on deal, because a longer deal lets more shoppers stock up.
ROI denominator is everything that left the business: the discount funded on the units not new, the discount funded on the genuinely new units (they were sold at the deal price too), and the margin the post‑promo dip takes back. Those are the three middle bars on the waterfall, so you can add up the denominator straight from the chart. A small ROI % on a large investment can still be a large absolute loss. Read Net Profit alongside ROI, never one without the other.

5.2Controls & toggles

Every input the calculator exposes, its range, and what it changes.

Control	Range	Default	What it changes
Discount Depth	5% to 50% in 1% steps	20% (the retailer's proposed template)	How deep the price cut goes during the promo. Drives both uplift AND margin dilution. Doubling depth doubles the discount on every promoted unit, and a deeper cut also sells more units at that bigger discount. So the margin given away grows faster than the uplift, and returns fall in the deep‑discount zone.
Volume Uplift	1.0x to 5.0x in 0.1x steps	2.0x (the retailer's headline expectation)	The multiplier on your baseline weekly volume while the promotion runs. Baseline is what the pack sells in an ordinary week with no deal on it. Realistic uplifts in fast‑moving consumer goods (FMCG) sit between 1.5x and 3.0x. Above 4.0x you are usually looking at pantry‑loading rather than genuinely new demand. Pantry‑loading is shoppers stocking up on packs they would have bought later anyway. Uplift on its own answers neither question the Performance Grid asks, which is the point. Pair it with two other things. The net share of uplift (net incremental volume / uplift) has to clear 50, a rule of thumb meaning more than half the extra volume is genuinely new demand. And the money has to come back.
Net Share of Uplift	10% to 90% in 5% steps	40% (a common planning assumption for mature FMCG categories)	The single most important LEVER, and the one the grid cuts on at 50, a rule of thumb. The slider is net share of uplift: net incremental volume / uplift, meaning of the extra units, the share that is genuinely new. Do not take the dip after the promotion off this number, because the Post‑Promo Dip controls already count it as a cost. The Net Incremental Units tile also shows that same volume as incrementality, net incremental volume / promoted volume. That share is the smaller of the two whenever the uplift is above 1.0x, because the baseline volume sitting underneath the promotion was never incremental. Healthy events typically clear 40 to 50 on the slider. Designing the promotion type well lifts that to 60 to 80. What does the work is a condition the shopper has to meet before the deal applies. A multi‑buy that only pays when several packs go through together, an offer that rewards a bigger basket, or a slot aimed at shoppers who have never bought your brand.
Duration	1 to 6 weeks in 1-week steps	2 weeks (standard TPR slot)	Weeks the promo runs. Longer durations boost total volume but train shoppers to only buy on deal. The share of genuinely new volume erodes with each repeat exposure. Run this same deal for 4 weeks instead of 2 and it loses exactly twice as many dollars, at the same return, because the dip after it grows with the weeks too.
Post-Promo Dip	0% to 40% in 5% steps	15% (heavy-promotion category mean)	Percentage volume falls below baseline after the promo ends, as pantry‑loaded shoppers stay away. Multiply that percentage by Dip Duration to get the total volume the event pushed into the future. Value that volume at your full margin, which is the margin you earn on a pack with no deal on it. The result is a straight cost to the P&L, with nothing coming back the other way. A small discount on something shoppers do not stockpile gives a dip of 5‑10%. A deep discount gives 25‑30% where the pack stores well at home, like canned goods or cleaning products.
Dip Duration	0 to 4 weeks in 1-week steps	1 week (post-2-week-TPR default in scanner data)	Weeks the post‑promo volume dip lasts. A 15% dip lasting 1 week costs much less than a 15% dip lasting 4 weeks. Heavy‑promotion calendars where consecutive events run back‑to‑back never let the dip fully resolve. The modeled dip cost is then chronically understated.

5.3Step-by-step exploration

7-step guided exploration of the scenario.

Read the proposed template at the default
The calculator opens on the retailer's proposed template: 20% discount, 2.0x uplift, 40% net share of uplift (net incremental volume / uplift), 2 weeks, 15% post‑promo dip, 1 week dip duration. Twenty percent off the $4.99 shelf price is about $1.00 a pack, and you fund all of it. Look at the main card: its two figures and its Performance Grid verdict.
Net Incremental Profit reads -$15,540 on a $20,980 investment, a return of -74.1%. The Net Incremental Units tile shows 4,000 units, an incrementality (net incremental volume / promoted volume) of 20% of everything sold on deal, against a slider set at 40% of the uplift. The Performance Grid verdict reads Cut, because the event failed both questions.
The waterfall shows why. The first bar, the gross profit on the new units, is $5,440: 4,000 new packs at the $1.36 each one earns you. The 16,000 units that were not new are 10,000 of baseline shoppers who were coming anyway, and 6,000 of uplift that only moved from elsewhere. The three middle bars are the costs set against it: $15,968 of discount funded on those 16,000 units, $3,992 of discount funded on the new ones, and $1,020 of margin lost in the dip after the promotion. Add those three together and you have the number the percentage divides by.
Expected outcome: Net Incremental Profit = -$15,540 on a $20,980 investment. Return = -74.1%. Share of everything sold that was genuinely new = 20.0%. Grid = Cut. The retailer's proposal (20% TPR, 2x uplift, 2 weeks) destroys roughly $15.5K per event on one SKU. Only one pack in five sold on deal was demand the brand did not already have. Industry pattern: McKinsey, working from Nielsen data, puts the share of trade promotions that lose money at around 59 percent, and this template is a textbook example of how.
Slide Net Share of Uplift to feel the dominant lever
Hold every other control at default. Drag Net Share of Uplift (net incremental volume / uplift) down to 25. Net profit falls to -$17,580, a -83.8% return, and only 12.5% of what shipped was new. Now drag UP to 60. Net profit recovers to -$12,820, a return of -61.1%, and the Performance Grid verdict changes from Cut to Fix. Nothing about the money moved: the investment is $20,980 in all three cases, because the depth never changed. What moved is the answer to the second question, and with it what you should do about the event.
Expected outcome: Net Profit at 25% / 40% / 60% slider = -$17,580 / -$15,540 / -$12,820. Incrementality (net incremental volume / promoted volume) at the same three points = 12.5% / 20.0% / 30.0%. Each 10pp on the slider is worth roughly $1.4K of profit on this single SKU per 2‑week event. The slider, net share of uplift, is the input you control and the number the Performance Grid scores. Incrementality is the diagnostic that explains the money. Now multiply that by the dozen or more SKUs and 50+ promotional slots in a typical national trade calendar. Customer Marketing, the team that plans and funds these events, gets more money back from measuring properly how much of the volume is genuinely new than from any other capability it can build.
Compare two strategic templates head‑to‑head
Reset, then build the shallow and high‑uplift template: discount 8%, uplift 2.5x, net share of uplift 70%. Shallow means a small cut in price. A multi‑buy at about 8 percent off, with the condition aimed at shoppers who do not buy you now, can reach 70% on that slider. That sits above the 50 to 60 percent this course's mechanic table gives a multi‑buy, so it takes a condition aimed at new shoppers, and your own past events decide whether you can claim it. The shopper only gets the deal by taking several packs, so less of the discount goes to people buying their usual single pack.
Then build the deep and moderate‑uplift trap: discount 35%, uplift 1.8x, net share of uplift 50%. Deep means a big cut in price. A straight price cut with no condition attached typically lands around 50% on the slider, which is what this setting reflects.
Expected outcome: Shallow and high‑uplift: Net Incremental Profit +$3,280 on an $11,000 investment, a return of +29.8%, with incrementality at 42.0% of everything sold on deal and 70% on the net share of uplift slider. Grid Scale. Deep and moderate‑uplift: -$27,017 on a $32,457 investment, a return of -83.2%, incrementality 22.2%. Grid Cut. A $30K gap on the SAME SKU in the SAME 2‑week slot, driven entirely by the depth, uplift and net share of uplift combination. Read the investment figure as well as the profit figure. The deep template spent nearly three times as much and lost most of it. Every extra point of discount costs you margin on every pack sold on deal. Extra uplift only pays back on the share of it that is genuinely new demand.
Add a realistic post‑promo dip
Reset to default. Now set Post‑Promo Dip to 30% and Dip Duration to 3 weeks. That is the realistic shape for a category shoppers can store at home. Deep events run one after another, so sales rarely get a clean week to recover. The dip cost grows sixfold, from $1,020 (default) to $6,120.
Expected outcome: Net Incremental Profit at default plus a heavy dip = -$20,640, a return of -79.1%, still Cut. The Post‑Promo Dip bar in the waterfall is now $6,120. That is more than the $5,440 of gross profit the new units earned, so the lost sales after the event cost more on their own than everything the event made. This is the profit killer nobody puts in the post‑event presentation. Store sales data often shows a good part of a promotion week's gain given back over the weeks that follow. The bridge is where you would have seen it coming.
Stretch the duration to see the loss multiply
Reset, then drag Duration from 2 to 4 weeks, holding everything else at default. Compare the Net Profit, ROI, and waterfall against the 2‑week baseline.
Expected outcome: At 4 weeks: Net Profit = -$31,080 on a $41,960 investment, ROI = -74.1%. The dollar loss is exactly twice the 2‑week loss, and the percentage return does not move, because every line of the bridge grows with the weeks: both discount costs double, and so does the dip, since a longer deal lets more shoppers stock up. The damage to the P&L doubles all the same. Running a weak template for longer multiplies the loss it was already making, because each extra week repeats the same bad economics. Worse, longer events tend to lose genuinely new volume as they run, because shoppers learn the cycle and wait for the next deal. This calculator holds the net share of uplift steady whatever the duration, so it is being kind to the long event.
Build a counter‑proposal, then see how much better it can get
Reset. Build a counter‑proposal you can defend in front of the retailer, starting from the shallow template in step 3. Set discount 8%, uplift 2.5x and net share of uplift 70%, over 2 weeks, with the default 15% dip lasting 1 week. A 3‑for-$X multi‑buy at about 8 percent off, where the shopper pays one set price for three packs, is the kind of design that can reach 70% on that slider. What you put to the retailer is the mechanic and how much you will fund. The deal price on the shelf stays the retailer's decision.
Read the main card. Then set the depth to 10%, uplift to 3.0x AND net share of uplift to 80% to see what a genuinely strong template looks like: a 10 percent cut with a multi‑buy condition and display support. Eighty is above every band on the lesson's mechanic table, which is why a template like this is rare.
Expected outcome: Counter‑proposal at 8% / 2.5x / 70 slider: Net Incremental Profit = +$3,280 on an $11,000 investment, a return of +29.8%, with incrementality at 42% of everything sold on deal. Grid = Scale. That is a loss turned into a gain, on the same SKU and in the same promotional slot on the trade calendar. Push to 10% / 3.0x / 80 slider: +$5,770 on a $15,990 investment, a return of +36.1%, and just over half of everything sold (53.3%) was new. Compare the two. The stronger template spends about $5,000 more and earns about 1.8 times as much. Its two extra points of depth cost you on every pack sold on deal, and the display and the tighter condition win enough genuinely new volume to pay for them and more. The calculator books no display fee, so take yours off the +$5,770 before you count the gain. Templates this strong exist in real calendars and are rare, so the realistic ambition is a plan with several like the first and one or two like the second.
Land on the Thursday recommendation
Settle on your preferred template, then test it against three things that could go wrong. The net share of uplift drops 10 points because the promotion type performs worse than you assumed. The dip runs a week longer because another event overlaps it on the calendar. And the depth creeps up 3 points because the retailer pushes you for more. Record the profit, the investment and the return for each.
Then scale it up. Say this one SKU earns +$3.3K an event on the counter‑proposal, or +$5.8K on the stronger template. Work out what the whole calendar comes to at 8 SKUs across 6 events a year.
Expected outcome: Your recommendation should sound like this. 'The retailer's proposal, a 20% discount at 2.0x uplift, lands in Cut on the grid. It destroys $15.5K an event on a $21K investment. The counter‑proposal, an 8% cut at 2.5x uplift with a multi‑buy condition, delivers +$3.3K an event on an $11K investment and lands in Scale. A 10% cut at 3.0x, with a tighter condition and display support, delivers +$5.8K on a $16K investment. The counter‑proposal holds up if the multi‑buy works less well than planned: ten points off the net share of uplift slider still earns +$1,240, still in Scale. It holds up if the dip runs a second week, at +$2,260. It does not hold up if the depth creeps 3 points to 11%: it loses $463 and drops into Fix. The stronger template stays in profit through all three tests. Across a full year, at 8 SKUs and 6 events each, dropping the retailer's proposal for the counter‑proposal is worth roughly $900K. Moving on to the stronger template adds about $120K on top of that.' One paragraph, every number from the calculator, decision‑ready. Note which test the counter‑proposal fails. Three points of depth on a pack that earns you $1.36 eat its whole margin of safety, so the depth is the number to hold firm on, and the most you will fund belongs in the plan in writing. Keep one test beside every number: a trade dollar you do not spend is one dollar of profit kept, so a return above zero already beats holding price. The Power of 1% from Pricing Lesson 2 says what is at stake, about a ninth of a thin‑margin business's operating profit for one point of price, which is why every trade dollar gets counted. It is not the bar.

5.4Reading the output

Every KPI, the formula behind it, and how to interpret a positive or negative value.

KPI	Formula	How to read it
Net Profit	(Net incremental units x full margin) - (Units not new x discount per unit) - (Net incremental units x discount per unit) - (Dip volume x full margin)	The number Finance cares about. Negative means the promo destroyed gross profit; positive means it created it. Read this BEFORE ROI. A small ROI % on a large investment can be a much bigger absolute loss than a large negative ROI % on a small investment. On the main card it carries a minus sign when the event loses money.
Promo ROI	Net Profit / (Discount on Units Not New + Discount on New Units + Dip Cost) x 100	ROI as a ratio of net incremental profit to total promotional cost. Useful for comparing events of different absolute sizes, but read alongside Net Profit. A +35% ROI on a $10K event is a $3.5K win; a +35% ROI on a $200K event is a $70K win. The percentage is identical, so only the absolute profit tells you how much money is actually at stake.
Net Incremental Units	(Total promo volume - Baseline volume in promo period) x Net share of uplift	Units the promotion ACTUALLY added, over and above what the pack would have sold anyway. This is the only volume the event can credibly claim as its own. Divided by total promo volume it gives incrementality (net incremental volume / promoted volume), the figure on the tile. The rest of promo volume is units not new: the subsidized base, loyal buyers who would have bought at full price and now get the discount too, plus the part of the uplift that only moved from elsewhere. Subtract Net Incremental Units from Total Promo Volume and you have the units whose discount bought nothing new.
Units Not New	Total promo volume - Net incremental units	The baseline that would have been bought at full price anyway (the subsidized volume), plus the part of the uplift that was not new. Every unit here took the discount and added nothing for your brand. The default scenario funds a discount on 16,000 of these units to win 4,000 genuinely new ones, four to one. That ratio is why net profit comes out at -$15,540.
Performance Grid Verdict	Two questions and no others. Did the money come back, meaning the return is above 0. And was the volume real, meaning Net share of uplift (net incremental volume / uplift) is above 50, a rule of thumb. Scale passes both. Hold passes the money question only. Fix passes the volume question only. Cut fails both.	The call to take into a trade review. The volume question reads the SLIDER, net share of uplift: of the extra units, the share that is genuinely new. TPO Lesson 4 uses the same number and the same cut‑off at 50, so an event scored here gets the same verdict there. The Net Incremental Units tile also shows incrementality (net incremental volume / promoted volume), which is the smaller number whenever the uplift is above 1.0x. The grid does not use it for the verdict. It is the diagnostic that explains why a template loses money. Its own pass mark is break‑even: discount per unit / regular gross profit per unit. Scale means replicate it and give it more weeks and more customers. Hold means the event pays, but mostly on volume that was coming anyway. Spend the next dollar on reaching more shoppers rather than on cutting the price further. Fix means the event is bringing in genuinely new shoppers and still does not make its money back. Change how it is funded rather than cancel a template that is already doing the hard part. Cut means cancel it and move the budget. Beyond the zero line there are no further bands. If somebody hands you a table of green, yellow and red cut‑offs, that is your company's own convention rather than an industry standard.

Read the four figures and the grid verdict as a stack. Net Incremental Profit answers did this make money. Promo ROI answers how much money it took to find out, and it divides by the three middle bars so you can check it yourself. Net Incremental Units against Units Not New answers where the money went. It tells you which of three things to change: the depth of the cut, the uplift you are buying, or the type of promotion you run. The grid verdict answers what to do next.

Before the trade review, read all five together. Then test how firm the answer is by sliding the net share of uplift 10 points either side of your assumption. If the verdict holds through both, the recommendation is committee‑ready. If a 10‑point move changes the verdict, check your assumption against scanner data before signing the JBP.

Benchmark to keep handy: McKinsey, on Nielsen data, puts the share of trade promotions that lose money at around 59 percent. The bridge is how you tell which of your own events sit in that 59 percent, and which ones only looked good because the volume went up.

5.55 common mistakes to avoid

Diagnostic patterns that catch the most common misuse of this calculator.

Mistake 1Reporting 'ROI' on total volume instead of incremental volume
Symptom: The headline read '+150% ROI' on a 20% TPR. Six months later the post‑event analysis was rerun, this time separating the genuinely new volume from the volume that was coming anyway. The same event came back with a negative return and a recommendation to cancel it. The underlying sales data never changed. What changed was the top half of the calculation, the profit, and the bottom half, the cost it divides by. That alone turned a headline win into a loss.
Fix: The return is net incremental profit over everything that left the business, and both halves have to be right. The top half counts only genuinely new volume, valued at the REGULAR gross profit per unit, the profit you make with no deal on the pack. The bottom half carries three things: the discount funded on EVERY promoted unit, any fee money you paid the retailer, and the margin the post‑promotion dip takes back. Counting only the discount on the units that were not new leaves out the rest. The default 20% / 2.0x scenario here: -$15,540 over $20,980 is -74.1%, which is what the event actually returned. Leave the discount on the new units out of the bottom half, and the events most worth fixing look better than they are. Those are the ones where most of the extra volume is genuinely new. And the old shortcut of multiplying the sales spike by the margin per unit overstates the return by a factor of two to five.
Mistake 2Confusing volume uplift with profit success
Symptom: Sales celebrated a 2.5x uplift on the Q2 event. Trade marketing celebrated the sell‑in number being hit. Three months later the bridge showed about -$40K of net profit per event. The net share of uplift had come in at 25%, not the 50% the plan assumed. And the 35% cut gave away too much margin to loyal buyers who were coming anyway.
Fix: Volume uplift is necessary and nowhere near sufficient. The grid needs the money back AND more than half the uplift genuinely new, a net share of uplift (net incremental volume / uplift) above 50, and uplift on its own answers neither. A 2.5x uplift at a 25% net share of uplift is one to cancel. The same 2.5x at 60%, on a shallower cut, is a template worth running again elsewhere. Pair the uplift number with the net share of uplift estimate before anyone celebrates.
Mistake 3Ignoring the post‑promo dip in the bridge
Symptom: The promo‑week analysis showed +$8K net profit. The retailer called it a win. Looking 4 weeks past the event, the number was -$15K. The shoppers who had stocked up stayed away for 3 weeks afterwards, and nobody had built that loss into the bridge in the first place.
Fix: Always include Post‑Promo Dip x Dip Duration in the bridge. Heavy‑promotion categories (biscuits, snacks, canned goods, cleaning) see 15‑25% dips lasting 1‑3 weeks. Light‑promotion categories see 5‑10% lasting 1 week. The default in this calculator, a 15% dip lasting 1 week, is the middle of the range. Plenty of real events come out worse. Before you sign a trade plan with deep cuts in it, go back to your scanner data and check what sales actually did in the weeks after past promotions.
Mistake 4Comparing depth and uplift in isolation, without incrementality
Symptom: The trade plan set a shallow cut with low uplift against a deep cut with high uplift, and chose the deep one because it shipped more total volume. The bridge, run after the event, showed the shallow template delivered 2‑3x the net profit. It ran at a 60% net share of uplift because the deal only paid on a multi‑buy, against 25% for the deep template, which was a straight price cut.
Fix: The share of genuinely new volume is a lever, not a constant. How you design the promotion lifts the net share of uplift (net incremental volume / uplift) from the 35‑45% an untargeted price cut typically earns to 50‑65%, and the best‑built events reach 70 or more. On a pack that earns $1.36 at full price, a price promotion only pays near the top of that range, which is why so few price promotions on a pack like this one pay. What does the work is a condition the shopper has to meet. A multi‑buy, an offer that rewards a bigger basket, a slot aimed at shoppers who have never bought you, or a deal with a deadline on it. In this calculator, a small price cut with a high uplift and a high net share of uplift (8% / 2.5x / 70%) consistently beats a deep cut with only a moderate uplift. Depth costs you margin on every pack sold on deal. Uplift only pays back on the share of it that is genuinely new.
Mistake 5Repeating losing templates because the cosmetic ROI metric looks 'okay'
Symptom: The retailer scorecard showed an average return near break‑even and the Customer Marketing team read it as 'roughly fine, optimise around the edges'. Commercial Finance ran the bridge across the calendar and found the average was hiding two very different groups. Most events lost money. A thin group returned very well. The average sat between them and described neither. The losing events were destroying enough to cancel out everything the strong ones earned.
Fix: Track distribution, not average, because an average is the wrong statistic for a shape like this. Every quarter, summarize the calendar on the Performance Grid, and give every event its verdict: Scale, Hold, Fix or Cut. Treat the events to cancel as a list, rather than as a percentage. In the worked calendar on this lesson, 59% of events lose money, 18% pay back thinly, and 23% pay back properly, with a mean of +0.7% that would have told you to leave the whole thing alone. Read the verdicts and the medians. The Net Incremental Profit Bridge diagnoses one event at a time. The Performance Grid is where you decide what to do with all your events together.

Related concepts

Go deeper on the theory

Continue with the lessons

Go further inside Trade Promotion Optimization

This calculator is the sandbox slice of Lesson 1: Promo ROI Fundamentals. Each of the other 7 Trade Promotion Optimization lessons teaches a complementary concept that sharpens how you read the output above.

See Promo ROI Calculator inside the full lesson

RGM Academy lets you pull every commercial lever yourself inside a senior-practitioner simulator, with the AI RGM Strategist coaching every decision you make.

Unlock the full Promo ROI Fundamentals lesson

Or see what a team rollout includes