Twelve questions that only become answerable with the data together
It is not that these questions are hard. It is that the answer lives in three different tools, and none of them knows the other two exist.
There is a simple test for whether you need a unified data layer, and it involves no vendor language: take a question that crosses two tools and try to answer it in five minutes.
Below are twelve we hear often. All of them look basic, and none has an answer when the site, the checkout, the CRM and the ad platform each keep their own slice.
About origin and money
- 01Of yesterday's sales, how many came from someone who clicked a paid ad at some point — not necessarily the last one?
- 02Which ads spent in the last seven days and produced no contact at all?
- 03Does this month's revenue match what the platform credits? If not, is the gap matching or double counting?
- 04How much of this month's revenue came from existing customers?
The fourth surprises subscription businesses most. In a product we measured, 40% of the month's approved transactions were renewals — and they were being counted alongside acquisition in the return calculation.
About time
- 01How long passes, on average, between the ad click and the purchase?
- 02What percentage of sales happen after the platform's attribution window?
- 03How long does the sales team take to reach a new lead, per rep?
The sixth is the one that changes budget decisions. In a business we measured, 92.2% of sales closed within seven days of first contact — and around 8% escaped the standard window. Knowing that number tells you whether the platform is underestimating your result or whether the problem is elsewhere.
About people
- 01Who bought today and never appeared in the CRM?
- 02How many people visited the capture page and never left an email?
- 03Which campaign brings the lead that replies, not just the lead that arrives?
- 04Is the same person entering the base twice with different emails?
- 05Which rep handled the person who bought, when the sale happens outside the CRM?
The tenth is the most valuable and the rarest. Lead volume is easy to buy; leads that reply are not. In one real measurement, a top-of-funnel campaign brought 1,442 leads, of which 128 replied, 123 reached a conversation and 37 bought. Without crossing campaign with conversation history, that whole chain is invisible.
Notice the pattern: none of these questions is about "how many". All of them are about "who" — and that is where traditional analytics stops, because it works with aggregates.
Why AI alone does not solve it
It is tempting to think a clever assistant works around this. It does not: if the CRM record and the sale record never met, no well phrased question joins them. The model will answer with what it has, and what it has is a slice.
What AI does change, once the data is together, is the cost of asking. Each of the twelve above is a query somebody would have to write, schedule and maintain. With unified data and a connector with rules, they become conversation.
The five minute test
Pick question 2 — which ads spent and produced no contact. It is the cheapest to attempt and the most revealing, because it needs only two sources: the ad platform and your contact events.
If the answer requires exporting two spreadsheets and joining them by hand, you already know the size of the problem. And if it requires that every week, you already know its cost.