This site exists to document bill creep and help readers audit and reduce it. But not every price increase is rent-seeking. Some recurring-charge increases reflect genuine cost increases, genuine feature additions, or genuine inflation. An honest consumer-advocacy site names both sides. Here is the case for when creep is legitimate.
Legitimate driver 1: Content licensing costs
Netflix's annual content spend grew from approximately $8 billion in 2017 to over $17 billion in 2023. Live sports rights, which Netflix began acquiring in 2023 (WWE Raw, NFL Christmas Day games, boxing), command multi-year rights fees in the $100-500 million range per year. Disney's acquisition of sports rights for ESPN+ is similarly expensive. A streaming service that is paying more for content it genuinely delivers to subscribers is passing through a real cost, not inventing a charge.
The caveat: content-cost pass-through is legitimate when the content is genuinely better or more extensive. It is less legitimate when a service uses a cost-argument to justify eliminating a lower-cost tier that was adequate for most subscribers. Netflix's elimination of the Basic ad-free tier in July 2023 was not purely cost-driven; it was a deliberate move to increase average revenue per user by removing the $9.99 tier that a significant portion of price-sensitive subscribers occupied.
Legitimate driver 2: Cloud infrastructure investment
The marginal cost of compute, storage, and bandwidth at the hyperscaler level has generally declined over time, driven by hardware efficiency improvements. AWS, Azure, and GCP have consistently reduced prices for standard compute over the past decade. However, GPU compute for AI workloads is genuinely expensive: the capital cost of H100 and A100 GPUs has not fallen at the same rate as CPU prices, and energy costs for large data centres in constrained power markets have increased.
For enterprise software vendors that use significant AI inference in their products, a price increase that reflects genuine inference costs is more defensible than a price increase on legacy functionality. Microsoft's Copilot add-on is positioned as a new capability layer; whether the $30/user/month enterprise add-on or the $21/user/month Copilot Business price reflects the actual incremental inference cost is hard to verify externally, but the argument that AI features have genuine marginal costs is sound.
Legitimate driver 3: General inflation
US CPI increased approximately 20% cumulatively between January 2020 and January 2024. Labour costs for customer service, engineering, and content creation have risen faster than CPI in most markets. A service that raises prices by CPI once every two to three years is not exploiting its subscribers; it is maintaining its real revenue against inflation. The Xfinity Broadcast TV Fee, which reflects genuinely increasing retransmission consent costs, has some legitimate basis.
The test is whether the increase tracks inflation or significantly exceeds it. Netflix Premium from $7.99 (2011) to $25.00 (2025) is a 213% increase over 14 years. CPI over the same period was approximately 55%. The delta between 213% and 55% is harder to justify on inflation grounds alone. Some of it reflects content investment; some of it is above-inflation margin expansion. Consumers can and should evaluate whether the service delivers value at the current price rather than at the 2011 price.
Legitimate driver 4: Features genuinely added
A price increase that accompanies a genuine new capability with clear user value is often legitimate. When Apple Music raised its individual subscription from $9.99 to $10.99, it had simultaneously introduced spatial audio and lossless audio at no additional charge, while also expanding its catalogue. The net price per song or feature arguably improved. A SaaS tool that raises prices by 15% but simultaneously adds integrations, increases storage limits, and improves reliability has a credible case.
The heuristic for telling them apart
A simple test: if the new price tier comes with a measurable new capability and the old tier remains available at the old price, the increase is often legitimate. If the old tier is removed or renamed so that staying in the same tier costs more without new features, it is typically rent-seeking.
Netflix Basic killed in July 2023: the option of a lower-priced ad-free tier was removed, forcing subscribers up to Standard ($15.49 at the time). That is rent-seeking.
Apple Music adding spatial audio at the same price for 2 years before raising it by $1: that is closer to legitimate.
Microsoft adding Copilot features as an optional add-on at additional cost rather than forcing all E3 subscribers to pay for it: that is legitimate, though the upcoming E3 base price increase from $36 to $39 in July 2026 requires a separate justification beyond the optional Copilot add-on.
The consumer response
The consumer's job is not to adjudicate whether every price increase is legitimate. The consumer's job is to pay for what they use and not pay for what they don't, and to periodically re-evaluate whether each service is delivering value at its current price rather than at the price when they signed up. The audit methodology on this site, the 50-item checklist, the negotiation scripts, and the tools comparison, applies equally to legitimate and illegitimate price increases. If Netflix at $25/month delivers more value to you than the alternatives, keep it. If it does not, cancel or downgrade.
The editorial position of this site is that the mechanism of bill creep, the combination of autopay, fine-print price changes, tier eliminations, and quiet fee additions, is designed to exploit the inertia of existing subscribers. That is different from saying that every price increase is unjustified. Some are unjustified; some are not. The audit tells you which services you are paying for and what they cost. The decision of whether to keep paying is yours.