The report had three new submissions. The example inbox had one person.
That person had sent the same request, waited, tried again with a shorter message and finally added a note asking whether anything had arrived.
This was a fictional review exercise, but it captured a very ordinary ambiguity. A submission is an event. A new conversation is something we need to understand from the request.
Count the events by all means. Just be careful about what the label implies.
The repeated message might have been an accidental second submission. It might have described a separate job. It might have reflected a broken confirmation or a customer unsure what would happen next.
A larger total cannot tell those stories apart on its own.
Follow the uncertainty
In the sample journey, the form worked. The customer received a short confirmation on screen. The message reached the inbox.
The confirmation didn't explain when someone would respond, and the receiving arrangement was unclear. The second submission was trying to solve that uncertainty.
The advertising account looked busy. The customer experience looked rather less satisfying.
My instinct used to be to start with the counting definition. That still matters. But the human question can reveal why the repeated event exists in the first place.
If the same request appears again, read the timing and the context through the business's appropriate process. Keep personal details private. We can learn from the pattern without copying somebody's message into a presentation.
Then check the confirmation and the response arrangement. Does the page acknowledge receipt? Does it describe the next step accurately? Can the receiving team maintain the response expectation?
Define the labels together
I would want the person handling the requests involved in the reporting definition. They can explain whether two messages concern the same conversation and when a request becomes useful to the business.
A clean event count is still useful for checking the form. A count of distinct, suitable conversations answers another question. The team should be able to discuss both without treating the difference as a mysterious disagreement between marketing and sales.
This is especially relevant when a report uses a word such as “lead” for several actions. A button click, a submitted form and a qualified conversation may all matter. They are not interchangeable just because the dashboard has room for one headline.
I am slightly allergic to dashboards being described as the single source of truth. They can be a very useful source of carefully defined facts. The truth sometimes needs a colleague reading the damn request.
The improvement might be outside the account
For the fictional example, the sensible change was a clearer acknowledgement and a defined handover. The form confirmed receipt, named the next step and gave an honest response expectation. The receiving team knew who was handling it.
We would then watch whether repeated uncertainty decreased, allowing enough time for ordinary requests to arrive.
I wouldn't claim the campaign suddenly generated fewer interested people if the submission total fell. The changed experience may simply have stopped one person needing to ask three times.
Equally, I wouldn't assume every repeated form has the same cause. The definition and the review need to remain open to other explanations.
A good report leaves room for the customer to have a continuing conversation. It doesn't require them to become a brand-new prospect whenever they press the button.
The example inbox was much easier to understand once we gave the person a place in the story.
A thought worth keeping?
