Your statement shows another five-dollar jump on a video app or a home internet plan. You wonder whether the government's inflation numbers caught it. They might not, at least not the way you expect. Before you read any official figure against your own receipts, run four quick checks.

  1. The category: Does the index track one service, or does it pool cable, satellite, and live streaming into one group?
  2. The base period: Is the number an index level or a percent change? A level only means something next to another level.
  3. The seasonal treatment: Has the series been adjusted to remove regular calendar patterns?
  4. The publication stage: Is this a preliminary estimate or a final release posted in agency tables?
A worn kitchen table in a sunny home with an open folder of household bills, a mug and a pencil, illustrating a careful review of monthly subscription costs.
Checking monthly bills against the official streaming and cable price series before reading a single jump as a trend.

Where subscription prices enter the official data

Subscriptions do not sit in one catch-all bucket. For cell phone plans and home internet, the Bureau of Labor Statistics describes how it prices services in a fact sheet on residential telecommunications services. The sources reviewed here give only the title and purpose of that document. Anyone who wants the actual sampling rules should read it directly.

Home video works differently. The Consumer Price Index tracks cable, satellite, and live streaming television service as one joint category. The series is published by BLS and carried in the St. Louis Fed's FRED database. It is seasonally adjusted and set to 100 in December 1983. That matters for a simple reason. A price hike at one major streaming service gets blended with every other provider in the group. A big jump at one company can shrink to a small move in the index. A change in legacy cable fees can offset it.

What the recent streaming figures show

Here is the FRED series for that category, seasonally adjusted: 604.484 in May 2026, 608.640 in June, 610.951 in July, 610.807 in August, and 614.756 in September. These are index levels, not prices in dollars and not percentages. To get a rate of change, subtract the earlier level from the later one, divide by the earlier level, and multiply by 100.

Take May to September. The gap is 614.756 minus 604.484, or 10.272. Divide by 604.484 and you get about 0.017, or roughly 1.7% over four months. Month by month, the moves are smaller still. June rose about 0.7% from May. July rose about 0.4%. August slipped by a hair. September rose about 0.6%. None of this is dramatic, and that is the point. A blended index moves slowly.

Now compare that with a CBS News video summary. It said BLS data showed streaming service prices up nearly 20% between November and December. The summary gives no year in the material reviewed here, and the underlying BLS tables are not included. The two numbers may describe different things, such as a narrower set of services against the full blended category. They may also come from different periods or different treatments. The sources do not settle it, so treat the 20% figure as unverified until you find the release it came from.

How to read the numbers without being misled

Start with seasonal adjustment. Adjusted series remove regular calendar patterns, such as price changes that tend to land in the same month each year. Compare an adjusted number from one month with an unadjusted number from another and you create a gap that is not real. Check that both points use the same treatment.

Next, use Table 2 of the CPI release. It reports index levels by detailed expenditure category for the U.S. city average. Find the row for your item, then read the column headings before you read any number. Confirm which columns hold index levels and which hold changes, and whether the figures are adjusted. Then look at neighboring rows, such as rent or food, to see how the item compares. The sources reviewed here do not say how much weight subscriptions carry in the overall index, so do not assume a big jump in one row moves the headline rate much.

Names change too. BLS published a notice about upcoming CPI series title changes. If you compare data across several years, confirm that the series title has not shifted. The agency's CPI frequently asked questions cover general method.

Questions to ask about a subscription price headline

Ask what the number measures. The Federal Reserve Bank of Chicago describes Flash PCE as a mid-month inflation nowcast, an early estimate made before the full monthly data arrive. Early figures can be revised. A nowcast is not the same as a final CPI table, and a news clip is not the same as either.

Ask whether the reported jump reflects list prices, promotional prices, or the way the index handles them. The sources reviewed here do not describe how BLS treats introductory discounts, bundles, or retention offers. So an official index may not match what you actually paid after a promo ended.

Finally, ask about scope. A headline percentage for one category over one month does not describe overall inflation. Your own bill is a real data point. The index is a different one, built to track the average across many households and many providers. Use the receipt to see your costs and the table to see the pattern.