Do good-better-best estimate options actually help, and how would I know?

What the mechanism is supposed to do

Presenting tiers changes the question a customer is answering. Instead of yes or no on one number, they are choosing among three. The claimed benefits are a higher average sold value from customers who step up, and a preserved sale from customers who would have declined the single price but accept the entry tier.

Those are two different effects pulling in opposite directions on average ticket, which is why a single blended number cannot tell you whether options are working.

The three measurements, in order

• Presentation rate. Of eligible estimates, what share actually included multiple options. This is usually far lower than management believes, and it is the first thing to fix.
• Tier mix. The distribution across entry, middle and top. If nearly everyone takes the entry tier, the tiers are anchored wrong or the middle is not differentiated. If nobody ever takes the top, it may still be doing its job as an anchor — check whether middle-tier selection is above what single-price quotes achieved.
• Close rate and sold value, within job type. Compare option quotes to single-price quotes for the same category of work. Comparing across categories measures job mix, not presentation.

The bias that ruins the naive comparison

Technicians do not present options at random. They present them when the customer seems engaged, the system is old enough to justify a replacement conversation, and there is time in the day. Those are exactly the jobs most likely to close at a high value regardless of presentation.

So an uncontrolled comparison attributes the quality of the opportunity to the format of the quote. To narrow it, restrict the comparison to a single job type in a single season, and check whether the option and single-price groups look similar on observable characteristics like equipment age or property type. If they do not, say so in the finding. A cleaner approach is to change the policy for a period — options required on every eligible job — and compare that period against the prior one for the same job types.

This is where quote data alone runs out. Whether options were genuinely presented, and how, lives in the conversation, which is what call and visit analysis reads.

Turn the finding into coaching, not a mandate

If presentation rate is the gap, the useful output is a per-technician view of eligible jobs where options were and were not offered, given to that technician's manager. If tier mix is the gap, the useful output is what the middle option contains and whether it is described in terms of outcomes the customer cares about.

Either way the analysis supports a human coaching decision rather than replacing it — which is how we structure rubric evaluation generally. Your standards, applied consistently, surfaced to the people who manage the work.