Quantify the load-time saving from reducing HTTP request count.
Requests are served in parallel batches, so eliminating requests only saves time when it removes a whole round trip. Header overhead saving is a separate and usually smaller benefit. Under HTTP/2 multiplexing the round-trip penalty largely disappears, which is why aggressive bundling matters far less than it once did.
HTTP Request Reduction
Round trips = ⌈requests ÷ concurrency⌉; time saved = round trips saved × latency
Round trips = ⌈requests ÷ concurrency⌉; time saved = round trips saved × latency Requests are served in parallel batches, so eliminating requests only saves time when it removes a whole round trip. Header overhead saving is a separate and usually smaller benefit.
Under HTTP/2 multiplexing the round-trip penalty largely disappears, which is why aggressive bundling matters far less than it once did.
This calculator takes 5 inputs: Current request count, Target request count, Round-trip latency, Concurrent connections, Per-request header overhead. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.