Active months come from the calculation, not the calendar

An active month is a month that received TRP once the calculation ran — not a month someone ticked on the input calendar. The two definitions agree in the middle of a flight and split at its tails. That split is where the "system lost a month" question comes from, and it decides what counts as a month of airtime, which makes it rather more than an interface detail.

A calendar tick is intent, not placement

Ticking a month says: I want weight here. Whether weight arrives depends on the curve that distributes it. The engine ships 17 flight-shape templates — Flat, Front-Loaded, Mid-Peak, Crescendo among them — and 5 intensity levels ranging from 0.5 to 1.7. Together they set the curve TRP follows across the period.

Front-Loaded at low intensity pushes the bulk of the weight into the opening and lets the curve fall away. By the final month it can reach zero. Nothing was dropped or forgotten; that month simply has no placement to report, so it stays out of the breakdown.

Why one definition changes every derived number

Take a 250 TRP plan. Split it across four calendar months and the monthly average is 62.5 TRP. Split it across the three months that actually carry weight and it is 83.3. Same plan, two lines in a deck, roughly a third apart. The question worth asking is which number describes the air.

We chose the second. Active months come out of the calculation itself, and monthly averages, budget breakdown and period splits all read from that same source. One source keeps those blocks locked together: there is no state where the average covers four months while the budget spreads over three. Keep two sources — the calendar tick and the actual TRP — and they desynchronize sooner or later, usually in a client meeting and not in a review.

Part of a wider rule: the engine does not complete your intent

This is one case of how the core behaves generally. It does not guess what the planner meant. If a value is missing, the calculation stops and names the specific field. If a value is present, that value is what gets computed. No silent substitutions, no "sensible" defaults. What you see is what was calculated.

The payoff is traceability: any figure in the report walks back to its input, a month later, by someone who never built the plan. The cost is that the report sometimes argues with memory. A planner recalls ticking four months and sees three. That is the moment to inspect the flight shape and the intensity level — the report is only showing the consequence of the chosen curve.

What should you check when choosing a tool?

Details like this rarely reach feature comparison tables. "Monthly breakdown" sits on every list; how a month is defined is an implementation question — and the implementation is what moves the number in the presentation. We take the comparison logic further in Media planning tools: what to compare before you buy

So, briefly. A calendar tick describes intent, realized TRP describes placement, and the report should carry the second. If monthly averages in your process are computed over ticked months, reconcile them once against the actual weight distribution. The gap at the tails of a flight is usually wider than you expect.

On measurement vocabulary and how ratings and TRP are defined, see Nielsen's published methodology material and the Media Rating Council. In TV Budgeting we reproduce Nielsen's numbers, we do not reinterpret them; and CPP inside the engine is a planning proxy, not a contracted price.