Ask what has already been learned about older adults.

Straight answers

The questions people actually ask
in the first meeting.

Do we have to hand over data to start?

Usually not. If your conversations already run through infrastructure we hold, the first read needs no new data agreement at all. That is the fastest start available and it is the one we prefer. If not, the marketing itself is enough to begin. If they do not, we can work from your public creative. Sharing your own de-identified data is a later step that sharpens the read, never a condition of getting one.

How is this different from a panel or a copy test?

It sits upstream of them. A panel grades what an older adult says about a message. We start from what older adults did after real messages, then tell you which question is worth putting in front of a panel at all. We are not a replacement for copy testing and we say so.

We already own a call-analytics seat. Why this?

Because that playbook was learned selling business software. We put twenty-nine of its published findings to the tape. Thirteen could be graded against settled enrollment: seven held, five inverted on older adults, one split by leg. Your seat is right about a lot. The problem is that nothing in the language of the advice tells you which half you are looking at. Play it yourself.

What if we already have a clean outcome feed?

Better. We grade against yours. If you do not have one, or it arrives weeks late with gaps, we read the outcome out of the interaction and tell you exactly how far that reader is certified, including the populations where it refuses to score.

Will you write our copy?

No. We flag where a message loses people and rank what to test. We do not propose replacement wording, and for regulated advertisers we never touch required-safety language, anything adjacent to it routes to your medical and regulatory review rather than to media.

Does any of this work off the phone?

Yes. The instrument that reads an advertisement needs no conversation at all: give it a television spot and it returns pace, placement and where the safety-bearing copy sits, validated against a published FDA content analysis. The measurement changes by channel; the method does not.

How long before we know something?

Days for a read against the existing record, often that is the whole answer, because the question turns out to be already settled. Where it genuinely needs a test, the first state reads out in about four weeks on the screener leg of a book running about 76,000 conversations a month where the customer speaks, and in four to five months on a transfer leg of about 8,000, scaling with your volume.

What does it cost?

A fixed fee for the first engagement, refundable unless you decide to continue. We would rather carry the risk of the first read than argue about its value afterwards. Anything beyond that is a conversation about scope, not a price list.

What happens if the answer is that you do not know?

You get told that, with the interval, and what it would take to resolve it. Most of what we froze from stated evidence alone did not survive contact with the outcome. A system that returns an answer to every question is not measuring anything. It is generating.

What happens next

We will not force every question
into a pilot.

You get one of four answers, and three of them do not involve buying anything.