Case study
Eleven partners, 815 opportunities, and a scorecard the board could read
An enterprise sales performance management software company had a real partner program: more than ten active System Integrators and hundreds of millions of dollars in partner-influenced pipeline. What it did not have was any structured way to measure that program, track certifications, or report partner contribution to the board. We built the measurement layer from the raw Salesforce data up.
Josh Weckesser · May 2025 · 7 min read
The activity was real. The measurement was not.
When we started, the partner team could describe its program in stories: this SI ran that implementation, that partner brought in a marquee logo. What no one could do was answer the question leadership kept asking, which partners actually drive revenue, and by how much. The data to answer it sat in Salesforce, but it had never been segmented, attributed, or rolled up into anything a board could read.
The gap was not effort. The partner team worked hard and the partners delivered. The gap was structural: partner-sourced and partner-influenced pipeline were never separated, certifications lived in a learning system nobody reported from, and there was no single document that said how the program was performing this quarter.
The symptom was a board deck that took days to assemble and still felt anecdotal. The root cause was that the program had never been instrumented.
The build, step by step
This was not a data science project. It was a sequence of specific steps that turned raw Salesforce exports into a scorecard, a set of segmented analyses, and the enablement tracking the program had been missing.
Step 01
Define the partner taxonomy
Before any number meant anything, we had to agree on what we were counting. We separated partner-sourced pipeline (the partner brought the deal) from partner-influenced pipeline (the partner shaped a deal that arrived another way), and we standardized partner type so a global SI and a regional SI were never added together by accident. Most partner reporting goes wrong right here, by folding sourced and influenced into one inflated number nobody trusts.
Step 02
Build the scorecard
We built a single scorecard tracking eleven named partners across the metrics leadership actually cared about: partner-sourced and partner-influenced pipeline, created and won, technical experience ratings, implementation project counts, certification levels, and cost structure. One row per partner, one view of the whole program.
Step 03
Run the pipeline analysis from the raw export
We pulled the raw Salesforce opportunity export, 815 opportunities across every stage, and segmented it by partner, geography, industry, and year. That let us compare global versus regional SI performance, average contract value by partner, and year-to-date trend. Working from the raw export rather than an existing dashboard meant every number was ours to verify, line by line.
Step 04
Stand up the enablement layer
Measurement is only half of a partner program. We built the partner learning path in both visual and document form, set up certification tracking from the learning system’s badge export, and created an implementation tracker covering every active partner-led project. Then we wrote the Partner Welcome Guide, the external onboarding document a new partner gets on day one.
What changed
For the first time, leadership had one document that answered the revenue question with numbers instead of anecdotes. At the time of delivery, the scorecard showed:
The qualitative change mattered as much as the figures. The quarterly partner review stopped being a scramble to assemble and started being a document the team could update and defend. Certification gaps were visible. Underperforming partners were visible. So were the partners quietly driving the most revenue, who had never been recognized for it.
What made it work
Separate sourced from influenced, always
The most common reason partner numbers are not trusted is that sourced and influenced pipeline get added together. We kept them apart in every view. A partner that influences forty million dollars and a partner that sources five million are both valuable, but they are not the same thing, and a board can tell when a deck pretends they are.
Work from the raw export
We rebuilt the pipeline analysis from the raw Salesforce export rather than an existing report. It was slower, and it was the right call. Every number on the scorecard could be traced back to specific opportunities, which is what made leadership willing to put it in front of the board.
A scorecard is a living system, not a one-time deck
We designed the scorecard so the partner team could refresh it each quarter from the same export and tracker, not so it would look good once and then rot. The value of partner measurement is the trend, and the trend only exists if the measurement repeats.
What we would do differently
Instrument attribution at deal creation
Most of the work in the pipeline analysis was reconstructing partner attribution after the fact from messy opportunity data. The better fix is upstream: capture partner type and sourced-or-influenced on the opportunity when it is created. The cleanup gets cheaper when the inputs are clean.
Automate the certification pull from day one
We exported certification status from the learning system by hand for the first version. It should have been an automated pull from the start, so the scorecard never goes stale between quarters because someone forgot to update a column.
Model the cost side earlier
We tracked partner cost structure, but we added it late. Partner profitability, not just partner pipeline, is the question leadership asks next. Building the cost view alongside the revenue view from the beginning would have answered it a quarter sooner.
The takeaway
Most partner programs are measured in stories because the data was never structured to be measured any other way. The activity is real and the pipeline is real, but the program cannot be managed because no one can see it. The fix is rarely more partner activity. It is a measurement layer that turns the activity already happening into something leadership can read, compare, and act on.
If you are running a partner program on spreadsheets, memory, and good intentions, the Partnerships & Ecosystem workflows we have productized at NVG8 came directly from engagements like this one. The Workflow Design Hour is the fastest way to scope whether one of them fits your program.