Why one number
Anyone following inflation already has more numbers than they can use: headline CPI, core CPI, PCE, core PCE, trimmed mean, median, sticky-price, and a dozen regional variants. Each one is defensible. Together they are difficult to hold in your head from one month to the next.
The index is not an attempt to be more accurate than any of these. It is an attempt to be followable. One number, one scale, one direction of travel.
The trade-off
Collapsing many series into one throws away detail, and detail is often where the interesting story is. That is a real cost, and it is why the breakdown will always be published alongside the headline figure.
What you get in return is a reading you can actually track over time without re-learning the context every month. A number that sat at 44 last autumn and sits at 62 now tells you something immediately, in a way that a table of year-over-year percentage changes does not.
What it is not
It is not a forecast of next month’s CPI print, and it is not a recommendation about anything. It measures pressure that is building, which is related to but distinct from prices that have already moved.