The problem: the pipeline lives in fragments
A sourcing event is one object, but you edit it as a dozen disconnected screens. The requirements sit in one document. The invitations and bids sit in another. The weighted matrix is usually a spreadsheet that nobody dares touch after the scoring meeting, because touching one weight means retyping the ranking, then reworking the margin, then rewriting the evaluation memo to match. So the analysis freezes early. You pick weights once, defend them once, and the moment a stakeholder asks what happens if we care more about implementation than list price, you are back in the spreadsheet doing arithmetic by hand.
That fragmentation is the same failure we described when the budget number was approved before anyone checked the market. Decisions get locked before the analysis is live, because keeping the analysis live is too much manual work. The desk removes the manual work.
app.isvcosell.com/sourcing
The floor shows every purchase on one runway with today as the gold line.
THE SAME JOB, TWICE
TODAY, BY HAND
Read the requirements document and the current bid list in two separate tabs
Rebuild the weighted scoring matrix in a spreadsheet and retype every vendor score
Recalculate the margin and the close price by hand after each weight change
Redraft the evaluation memo and the award letter so the narrative matches the new numbers
Roughly 10 hours, spread across the week before the award meeting
WITH ISVCOSELL
Open the sourcing desk over the pipeline you already run
Drag a category weight and watch the ranking, the margin and the memo rederive together
Hold a target discount against the leader's stated list to move the close and award scenarios
Read the live evaluation memo, which already reads the updated rows
About 25 minutes of your attention
What changes: 10 hours becomes about 25 minutes. For a team running, for example, two sourcing events a month, that is roughly 20 hours a month of retyping recovered, or close to a full working week each quarter that goes back into negotiating rather than reconciling.
PART TWO
The floor and the rooms
The desk opens on the floor. Every purchase sits on one runway with today drawn as the gold line, so you see what is early, what is late, and what is crowding the same approval window. From there you walk fourteen rooms in the order the work actually happens. Intake prices the overlap before anyone approves it, which is the discipline we argued for in the request that never states the cost of doing nothing. Then requirements, the field one of our six specialist agents researched, the invitations, the bids, the weighted matrix, the pricing walk, the strategy, the drafted letters, the live evaluation memo, the close, and the award.
The important part is what the rooms do not do. They do not own separate copies of the deal. They read the same rows the stage pages read, and the stage pages keep the heavy edits. So the desk is a reading and steering surface, not a second system of record. Nothing you do in it diverges from the pipeline underneath. This is the same principle behind working a whole contract without leaving the archive: one source, many views.
"The rooms do not own copies of the deal. They read the same rows the stage pages read, so nothing you steer diverges from the pipeline underneath."
PART THREE
Two levers that rewrite the analysis without writing anything
The desk runs on two levers, and neither asks you to type. The first is category weight. Drag a weight and the ranking, the margin and the memo rederive together. If you decide implementation risk matters more than headline list price, you move the weight and the leaderboard reorders, the margin math follows, and the evaluation memo already reads the new order. There is no retyping and no stale narrative.
The second lever is the target discount. Hold a target discount against the leader's stated list and the close price and the award scenarios move with it. You are testing what a given concession does to the shortlist in real time, against the leader's own quoted list, not against a number you invented. This is where the desk earns its keep in a live negotiation, because you can walk into the room already knowing which discount flips which vendor into the lead.
app.isvcosell.com/sourcing/strategy
The strategy room and the pricing walk rederive as you move a weight or a target discount.
PART FOUR
A gate is not a weight, and the desk knows it
There is one rule the levers cannot override, and it matters. A bid that misses a mandatory gate stays struck at every weight, because no weight cures a gate. You cannot weight your way past a vendor who failed a must-have. This is deliberate. It is the antidote to the failure we described in everything is mandatory, so nothing is negotiable, where must-have and nice-to-have collapse into one flat list and the scoring quietly rewards a vendor who never met the floor. Here the gate holds. A struck bid stays struck no matter how you drag the weights, and the memo says why.
The other thing worth knowing on day one: an empty pipeline does not open empty. It offers the fully worked sample estate, a complete sourcing event you can walk end to end. So the desk is worth opening before you have loaded a single real deal, because you can see exactly how the levers behave against a finished example first.
1 The pipeline reads as one object. Every purchase sits on one runway with today as the gold line, so overlap and crowding are visible before you approve anything into a busy window.
2 Weight changes rederive the analysis. Drag a category weight and the ranking, the margin and the evaluation memo move together, with no retyping and no stale narrative to reconcile.
3 Discount scenarios run against real list. Hold a target discount against the leader's stated list and the close price and award scenarios update, so you know which concession flips which vendor before the call.
4 Gates outrank weights, always. A bid that misses a mandatory gate stays struck at every weight, so you cannot accidentally weight a non-compliant vendor into the lead.
5 The stage pages keep the heavy edits. The desk reads the same rows it steers, so it is a view over your pipeline, not a second system of record that can drift out of sync.
6 Day one is not empty. An empty pipeline offers the fully worked sample estate, so you can learn the levers against a finished event before loading a real one.
PART FIVE
Honest limits
The desk steers analysis. It does not invent inputs. If your requirements are vague, the weighted matrix will faithfully rank vague inputs and the memo will read exactly as vague as you wrote it. The levers make the consequences of a decision visible, they do not make the decision for you. That remains a judgment call, and it should.
Two more boundaries. The target discount lever works against the leader's stated list, so its quality depends on having a real quoted list to hold it against. When a vendor is coy about list price, the scenario is only as honest as the number you can pin down. And the gate logic only strikes what you have marked as a mandatory gate. If a genuine must-have is filed as a preference, the desk will treat it as one. It enforces your structure precisely, which means it also inherits any looseness in your structure. The desk is a faithful mirror, not a corrective, and on a well-specified event that is exactly what you want it to be.
About the author
Fredrik Filipsson, Cofounder, ISVCOSELL
Fredrik has spent more than twenty years in enterprise software, with time at Oracle, IBM, SAP, and Salesforce before moving to the buy side. He structured and priced the kind of large agreements most buyers only see once or twice in a career, which taught him where the leverage sits and how far a vendor will actually move. He started ISVCOSELL to hand that knowledge to every sourcing team.
More posts by Fredrik Connect on LinkedIn →
See it in the product
How benchmarking works → Browse the use cases → Every feature → Calculate your time saved →
FREE TRIAL · FULL PLATFORM · NO CARD REQUIRED
Open the desk on an empty pipeline and see how it reads
The free trial opens the benchmarking database, 1,483 vendors deep, plus the negotiation guides, playbooks, and talking points for your own renewals. No card needed, a corporate email is all it takes.
Start your free trial → Or decode a contract free, no account
Free for 30 days, no card needed. Your data stays isolated at the database, and you can export or delete it any time.
Watch it in action
ISVCOSELL: the three minute demo What discount should we expect? One question, every agreement
Browse the full demo library →
V ISVCOSELL
A ISVCOSELL product · © 2026
PLATFORM Benchmarking Negotiations Contract management AI workflows Document search Renewal calendar
PRODUCT Use cases Features Security Pricing Deployment options About
RESOURCES Getting started ROI calculator Agent protocol FAQ Blog Request access