Start with your workload, not a rate
Two projects with the same monthly traffic can have different bills. One delivers large cached archives; another serves millions of small images. A useful estimate includes request counts, object sizes, audience countries and peak patterns. Gather a representative month of data and identify seasonal events separately.
Geography changes the calculation
Delivery costs may vary between regions. An average rate tells you little if much of your audience is in higher-cost markets. Compare proposals against the same traffic distribution: which countries generate most of your volume, and where is consistent delivery particularly important?
Include your origin costs
Cache misses send requests to your origin server, so your hosting bill can also be affected. Caching public files appropriately can reduce repeated transfers. Personalised responses and private data, however, must not be indiscriminately cached as public content. Your delivery model needs to respect those boundaries.
Check what the proposal includes
Ask about request charges, minimum commitments, overage, optional features and peak-related billing. Compare the complete budget for your workload rather than the most prominent per-gigabyte rate. A test or price in one region should not be treated as representative of your entire audience.
Where pooled demand can help
An aggregator combines commercial demand from several customers. That may improve purchasing terms without requiring every project to make a large independent commitment. Savings should be supported by a calculation. Share your traffic distribution and current costs, removing confidential billing details where appropriate.
Let’s apply this to your project.
Tell us what you deliver and where your audience is. We will discuss a suitable configuration.
Discuss your project