Google Preferred Source See Bitcoin.now more often.

Add Bitcoin.now to your preferred sources so Google can surface more of our Bitcoin reporting and explainers in your results.

Make Bitcoin.now preferred Opens Google’s source preferences. You stay in control.
Google Preferred Source
Block-space history

Bitcoin historical fee explorer: transaction fees over time

See what fee estimates mean, compare confirmation targets, and preserve recent observations so the page remains useful during source outages.

Current confirmation targets

Loading cached fee observations…

Next block -
About 30 minutes -
About one hour -
Economy -

Fee-rate curve by confirmation target

The site retains last-good public observations. As more samples accumulate, this page can compare the current curve with earlier network conditions.

Fees buy block space, not bitcoin value

A transaction moving 0.01 BTC can cost more than one moving 10 BTC if it has more inputs or bids a higher sat/vB rate. Miners generally compare fee rate because blocks have limited weight.

Historical averages can hide spikes

Demand can change within minutes. A daily median may hide a short rush, while a peak may describe only a few blocks. Always pair the time range with the statistic.

Estimate your own transaction in the UTXO planner or inspect the live backlog in the mempool explorer.

Why fee estimates disagree

  • Nodes can observe different mempools.
  • Models use different recent blocks and probability targets.
  • Wallets choose different safety margins.
  • Child-pays-for-parent packages change effective fee rates.
  • Replace-by-fee can raise a transaction’s bid later.

No target is a promise

Miners choose transactions independently. A quoted target should be read as an estimate under current conditions, not a guaranteed service level.

Method and boundary

How a fee observation becomes useful history

Every retained point needs a target, timestamp, unit, and source condition.

01

Capture the curve

Several confirmation targets are retained together so one urgent estimate is not mistaken for the whole fee market.

02

Keep the outage visible

The most recent valid observation can remain available with its age shown. A cached point is never relabelled as newly observed.

Questions for comparing Bitcoin fee conditions over time

What is sat/vB?

Satoshis per virtual byte is the transaction's fee rate. It lets miners compare the reward offered for the block space consumed rather than the bitcoin value transferred.

Can a low-fee transaction be accelerated?

Some transactions can use replace-by-fee or a child-pays-for-parent package. Availability depends on transaction structure, wallet support and node policy.

Why retain cached observations?

A history is built from labelled point-in-time samples. Keeping the last valid sample also prevents a temporary upstream outage from turning the entire page blank.

Block-space history

See what the fee market has actually charged

The explorer plots fee rates over time so a reader can tell an ordinary week from an exceptional one. Fees are a market for block space, and their history is the best guide to what counts as expensive.

Reading the chart

MetricDefinitionTypical calm rangeWhat a high value means
Median fee rateMiddle rate of confirmed transactions1–10 sat/vBSustained demand for block space
Average fee rateTotal fees ÷ total vsize2–20 sat/vBA few large payers, or genuine congestion
Fees per blockBTC paid to miners in fees0.05–0.5 BTCFees rivalling the subsidy
Fee share of rewardFees ÷ (subsidy + fees)1–10 %The network paying for security through fees
Reading note 1

Median tells a truer story than average

A handful of very high-fee transactions pull the average up while most users paid far less. The median is the rate a typical confirmed transaction paid, which is what a sender needs to know.

  • Read the median for planning, the average for what miners earned.
  • Check whether the series is per day or per block.
  • Rates are in sat/vB; multiply by size for a cost.
Bitcoin fees explained →
Reading note 2

Spikes have causes you can name

Late 2017 saw exchange-driven congestion; spring 2021 followed a price surge; May 2023 and April 2024 were driven by inscriptions and token protocols competing for block space. Each spike fell back within weeks.

  • Compare a spike with the price chart before assuming a link.
  • Watch the mempool backlog that preceded each rise.
  • Note that SegWit and batching lowered the size of typical payments.
Open the live mempool explorer →