The Economics of Per-Byte Billing in Serverless and Edge
Serverless and edge platforms bill storage, reads, and egress per byte. Small-string compression turns a ratio into a recurring discount.
Per-Byte Is the Default
Modern serverless and edge platforms almost universally bill on bytes: bytes stored, bytes read, bytes transferred between regions. Unlike a fixed server you rent, these bills scale directly with your data volume — which means a compression ratio translates straight into a recurring discount.
Where the Bytes Live
Most of the byte volume in these systems is not large files; it is enormous quantities of small records:
- KV and object entries holding JSON state
- Queue messages
- Per-request logs and traces
- Analytics and telemetry events
Each item is tiny, but you have billions of them. This is precisely smoltext's target band.
Turning a Ratio into Money
Suppose smoltext compresses your records to 40% of their original size. Then, to a first approximation:
- Storage at rest drops by 60%.
- Read egress on those records drops by 60%.
- Cross-region transfer of those records drops by 60%.
On a six- or seven-figure annual data bill, that is a material line item, and it recurs every month without further work once the pipeline is in place.
Where It Does Not Apply
Be honest about scope. Compute time, request counts, and per-invocation charges are unaffected by payload size. smoltext reduces byte-denominated costs, not request-denominated ones. If your bill is dominated by invocation count rather than data volume, compression moves a smaller fraction of it.
The Latency Budget
Edge compression runs in sub-millisecond time, so it rarely threatens latency SLAs. But every external call has a round trip. Co-locate compression with the edge, or batch records, so the API call does not become a new latency source. The deep dive on batching covers this.
Modeling Your Savings
Before committing, model it: take your monthly byte volume for small records, multiply by your expected compression ratio (measure a sample against https://api.smoltext.sprapp.com/v1/compress), and apply your per-byte rates. That gives a defensible savings estimate to justify the integration work.
The Compounding Effect
Per-byte savings compound across the data lifecycle. A record compressed once at ingestion stays small through the queue, through storage, through every read, and through every cross-region hop. You pay the compression cost once and collect the discount repeatedly.
When to Prioritize It
The payoff is largest for high-volume, retention-heavy, read-heavy workloads on per-byte platforms. If that describes your system, short-string compression is one of the highest-leverage cost optimizations available — precisely because it targets the bytes the rest of your stack ignores.