Chapter 10: Your Quote Log Knows More Than Your Market Research
A product manager at a crane-equipment manufacturer wanted to know how the new generation of control systems was actually landing in the market. The official answer lived in quarterly revenue reports, which meant it was months old and aggregated into uselessness.
So she looked somewhere unusual: quote frequency. How often was the new system appearing in configurations, per region, per week? The answer was immediate and specific: adoption was healthy everywhere except one region, which was quoting the old generation as if the new one didn't exist. One conversation later, the cause surfaced (a local objection nobody had escalated) and the fix landed within the quarter. The revenue reports would have shown the problem six months later, as a mystery.
That story is the thesis of this chapter: CRM knows the narrative. ERP knows the money. Only CPQ knows the choices. Every configuration session is a customer telling you, in structured form, what they need, what they considered, what they rejected, and what they were willing to pay. Multiply by years and you're sitting on the largest market research study ever conducted on your products. Almost nobody reads it.
What's in the log
Start with what was quoted versus what was won versus what was built. The gaps between those three are strategy information. Options with high quote rates and low win rates are mispriced or misexplained. Options customers add and then remove late in negotiation are your discount pressure points. And the free-text field where salespeople type the "specials" is a product roadmap in disguise: if the same request shows up twelve times, it's not an exception. It's an unmodeled requirement, which is to say, Chapter 4's three-times rule, now automated.
Timing hides in there too. One analysis I keep coming back to: about a third of what sales called "price problems" were actually timing problems. The quote was fine; it arrived after the buyer's urgency peaked, and the discount was the apology. You met the time tax in Chapter 1. The quote log is where you catch it in the act.
Then it gets predictive. One manufacturer fed four years of quotes, wins and losses both, into a model that scores new quotes on win probability. The craft detail that separates useful from theatrical: calibration. A score of 0.70 must mean seven of ten similar quotes actually close, or the sales team will smell it and ignore it. Calibrated, it routes effort: fast-lane the likely winners, stop chasing the walking dead, and watch discount behavior improve because desperation has left the process.
The signal even flows upstream of the factory. Configured quotes are the earliest demand signal your company produces: a spike of a long-lead component in this quarter's pipeline, probability-weighted, gives sourcing a 90-day head start on the order book. Start with five critical parts, not fifty. And a port-equipment client computes sustainability numbers at configuration time, so "eco share of sales, by region, this week" is a query, not a project.
Anti-pattern: Dashboard Theater. Beautiful analytics nobody acts on. The test is brutal and worth adopting as policy: if a chart hasn't changed a decision in 30 days, delete it. Non-choices are signal too; most teams store only the final configuration and throw away the story of how the customer got there, which is like keeping the verdict and burning the trial.
A warning before the finale, because this chapter can sound like an analytics sales pitch: all of it depends on people actually using the system. Data from a tool the field routes around describes your workarounds, not your market. And the field routes around more of these systems than any vendor will admit.
Why they do it, and what the companies that fixed it did differently, is the next chapter.
You're sitting on years of market research you've never read. Your quote log is the earliest demand signal your company produces.