Research
Where the money would go
The same ₹9.7 Cr as every other screen, sorted by the system the work sits on rather than by the finding. ₹6.2 Cr of it is on processes no software does today, which is the part that needs building; the rest is inside software they have already paid for, which is configuration and process.
The one split that decides what a first phase costs. 6 of the twelve findings sit in software the client already owns and are worth ₹3.5 Cr a year: cheaper to fix, harder to sell, because nothing new arrives and the work reads as configuration. The other 6 sit in the space between systems and are worth ₹6.2 Cr, close to twice as much. That is where a build goes and where the money is. Read the two figures against each other rather than adding them: together they are the same ₹9.7 Cr every other direction ties to, sorted by what the work would run on instead of by what it costs.
₹6.2 Cr a year on work no software touches, biggest first. Spend analytics is the largest at ₹2.1 Cr and falls to nobody at all, which is precisely why no price on this page is measured from the client's own data. Demand planning falls to one spreadsheet, and it is why stock cover runs at 38 days against a sector best of 22. Invoice capture falls to nine people keying about 96,000 invoices a year from PDF and paper. Naming who absorbs the work matters more than naming the system, because the person doing it by hand is the one who feels the case and the one who will say so on the call.
The three SAP modules already running, and what the findings on each are worth. Nothing in this section needs buying: ₹3.5 Cr of the total sits inside software that has already been paid for, so the work is configuration and process rather than a purchase. That makes it the easier half to get agreement on and the harder half to sell, because the client cannot see anything arrive. It also sets the integration boundary, which is the first thing an engineer asks: anything built for the section above has to read from and write back to these.