The pocket price waterfall: finding the 20% you never knew you gave away
There are two prices in every deal you close: the one you negotiated and the one you actually kept. The pocket price waterfall shows every step between them, and in Indian B2B about half the gap sits below the invoice line where no discount report can see it. Twenty contracts and one afternoon is the whole method.
Summary
- Everything above the invoice line is visible and governed. Everything below it, freight, waived implementation, unbilled scope and payment terms, leaves the business silently.
- Twenty contracts, every step from list price to the cash you kept, worked out from the contracts and the ledger rather than the CRM: that is the whole build.
- Give every step an owner, then lift the bottom of the price band and narrow its spread. Neither move requires a price increase.
There are two prices in every deal you close.
The one you negotiated. And the one you actually kept.
Most companies report the first and live on the second, and the gap between them is usually much larger than anyone in the business believes. The tool for seeing it is nearly forty years old: it came out of McKinsey's pricing work in the 1990s, and it's called the pocket price waterfall. It's the single most useful thing you can build if you want to know what your pricing is really doing.
What it looks like
Take a list price of 100 and follow it down through a typical Indian B2B deal.
- List price: 100
- Volume discount: -8. Running price 92.
- Competitive discount to win the deal: -7. Running price 85.
- Invoice price: 85
- Early payment discount: -2. Running price 83.
- Freight you absorbed: -1.5. Running price 81.5.
- Implementation you waived: -6. Running price 75.5.
- Support and scope you never billed: -3. Running price 72.5.
- Cost of waiting 90 days to get paid: -3.5. Running price 69.
- Pocket price: 69
You negotiated 85. You kept 69. That's 31% gone from list, and only half of it appears anywhere in your reporting.
The line that matters
Look at where the invoice sits in that table.
Everything above it (the volume discount, the competitive discount) is visible. It's on the quote. It's on the invoice. It's in the CRM. It shows up in the discount report. Somebody probably had to approve it.
Everything below it leaves the business silently. Freight sits in operations. Implementation sits in the delivery team's time. Payment terms sit nowhere at all. None of it appears in a discount report, because technically no discount was given.
So your sales team genuinely believes they held the price at 85. By every number available to them, they did.
The measurement just stops at the invoice line.
In the original McKinsey work, the money going out below the invoice was often about the same size as the discounts above it. In our experience with Indian companies it's usually worse, because free implementation, absorbed freight and long payment terms are so normal here that nobody counts them as concessions at all.
Build yours in an afternoon
This is not a big project. Here's the whole method.
Take twenty recent contracts. Not your best ones, and not your worst. A normal spread.
For each, list every step between the price you quoted and the cash you kept. Write each as a percentage of list price. The usual suspects: volume discount, negotiated discount, early payment discount, freight, implementation or onboarding waived, extra seats or units thrown in, unbilled support, scope that grew after signing, credit notes, and the cost of however long you waited to get paid.
Work it out from contracts and the ledger, not the CRM. The CRM has what people entered. The contracts and the bank have what actually happened. They are rarely the same.
On the payment terms line, use your real cost of money. At 14% a year, ninety days costs you about 3.5% of the invoice. At 120 days, about 4.6%. It's a genuine cost and it belongs in the picture. Then average across the twenty and you have your waterfall.
Give every step an owner
This is the part that turns a chart into a change.
A leak with no name against it stays a leak.
So assign each step:
- Discount depth: sales
- Freight: operations
- Implementation and onboarding: delivery
- Unbilled support and scope: customer success or delivery
- Payment terms: finance
- Credit notes and settlements: finance
Most companies find three or four steps that literally nobody knew existed. Those are the cheapest wins you will ever get, because fixing them doesn't require winning a single new customer. The complimentary ones are usually the largest.
The second half: the price band
The waterfall tells you what you give away on average. The band tells you where.
Plot every deal as a dot. Price you actually kept on one axis, size of the customer on the other. You'd expect a neat downward slope: bigger customers pay a bit less. What you get is a cloud. A customer buying 300 units at 95. Another buying 380 units at 66. One at 6,500 units paying 76, and one at 7,500 paying 49.
Three things fall out of that picture immediately.
How wide the band is, is how much money is available. If your deals span 49 to 95 on the same product, roughly half your pricing is being decided by something other than policy, usually which salesperson handled it and how hard the buyer pushed. That spread is money you can recover without winning anything new.
The upside-down ones are the scandal. Every case where a small customer pays more than a large one is a negotiation outcome, not a commercial decision. These are the deals to look at by name in your first review. There's always a story. Some of the stories are fair. Most aren't.
The bottom of the band tells you where to start. A small number of accounts hold most of the damage. You don't need to reprice your whole base. You need to fix about a dozen customers.
The two moves
Once you can see the band, there are only two things to do with it.
Lift the bottom
Bring the worst-priced deals up toward the rest. This is a rules problem, not a pricing problem: it means a discount approval ladder and a floor nobody crosses. It doesn't require you to announce anything to anyone.
Narrow the spread
Make price a function of volume, segment and contract length rather than of who negotiated. This usually delivers more than lifting the bottom, and it's easier politically: you're enforcing your own discount rules, not raising prices.
Neither one needs a price increase. That's the part founders find surprising.
One condition
The waterfall only works if you can calculate the price you actually kept, per deal. That means your contract data and your ledger have to be reliable enough to work from.
If they aren't, fix that first. There's no point building a careful picture on numbers you don't trust. You'll just be precisely wrong, and clean data is the cheaper problem to solve first.
Where to start
Twenty contracts. One afternoon. Every step from list price down to the cash you kept, with a name against each step.
Most founders have never seen that number for their own business. It is, in our experience, the single most uncomfortable and most useful afternoon in the whole pricing exercise.
How useful was this article?
One tap. It tells us what to write more of.
The next one
Get what we publish next, by email.
Working notes on raising, borrowing, protecting, growing and structuring capital in India. One email a week at most, and you can leave any time.
We use your address only to send this. See our privacy policy.
We store your address to send you these emails and nothing else. See our privacy policy.
Related reading
