How Billing Works

Bitping bills based on Units. One unit equals either 1,000 API requests (checks) or 1 GB of bandwidth, whichever is consumed first. Your plan includes a set number of units per month, and any usage beyond that is billed as overage.

Plan	Monthly Fee	Included Units	Overage Rate
Starter	$5	50 units (50K checks or 50 GB)	$0.10 / unit
Pro	$99	1,000 units (1M checks or 1 TB)	$0.10 / unit
Business	$999	10,000 units (10M checks or 10 TB)	$0.08 / unit
Enterprise	Custom	Custom	Negotiated

What Counts as a Check

Every time Bitping runs a monitoring job against one of your endpoints, that counts as one check. A single endpoint configured at 30-second frequency generates:

• 2 checks per minute
• 120 checks per hour
• 2,880 checks per day
• ~86,400 checks per month (86.4 units)

How Frequency Affects Usage

The check frequency you set for each endpoint is the biggest driver of unit consumption. Lower intervals mean more checks:

Frequency	Checks / Month	Units / Month
5 seconds	518,400	518.4
10 seconds	259,200	259.2
30 seconds	86,400	86.4
1 minute	43,200	43.2
5 minutes	8,640	8.6
15 minutes	2,880	2.9
30 minutes	1,440	1.4
1 hour	720	0.7

Changing frequency from 1 minute to 10 seconds multiplies your usage by 6x for that endpoint. The usage estimate on the config page helps you see the impact before saving.

Frequency and Data Quality

Higher frequency doesn’t just cost more — it produces fundamentally better data. The number of data points directly affects the reliability and usefulness of your metrics.

Why higher frequency matters:

• More accurate percentiles — P95 and P50 are statistical measures. With 2 data points per 15-minute window, you’re computing a “95th percentile” of almost nothing. With 30 data points per 30-second window, the percentiles actually represent the distribution of real user experience.
• Faster incident detection — At 15-minute intervals, a 10-minute outage might only appear as a single failed check. At 30-second intervals, you’d see 20 consecutive failures with clear start and end times.
• Meaningful min/max spread — The Response Time chart shows Min, Avg, P50, P95, and Max. With infrequent checks, these values collapse into a single line because there aren’t enough samples to show variance. Higher frequency reveals the true spread of performance.
• Better geographic coverage — Bitping distributes checks across nodes globally. More frequent checks mean more nodes get sampled, giving you a broader picture of how different regions experience your service.

Recommended frequencies by use case:

Use Case	Recommended Frequency	Why
Uptime monitoring (is it up?)	1 — 5 minutes	Enough to catch outages within a few minutes
Performance monitoring (how fast?)	10 — 30 seconds	Need enough samples for meaningful percentiles
SLA reporting	30 seconds — 1 minute	High-resolution data for accurate SLA calculations
Cost-conscious monitoring	5 — 15 minutes	Minimal unit usage, but limited insight into performance
Critical API endpoints	5 — 10 seconds	Near real-time visibility into latency and errors

Rule of thumb: If you care about response time, use 30 seconds or less. If you only care about uptime, 5 minutes is fine. 15-minute intervals are too slow to meaningfully measure website speed — by the time you notice a problem, it may have already resolved.

Usage Estimate Preview

The Estimated Usage panel on the config page shows a live calculation as you edit your configuration:

• Units / month — Total estimated units based on all endpoints and their frequencies, with your plan’s included units shown for reference
• Checks / month — Raw number of API requests
• Est. monthly cost — Your plan fee plus any overage, calculated from your current config
• Endpoints — Total number of monitored endpoints

This updates in real time as you add endpoints, change frequencies, or modify your setup. Use it to find the right balance between monitoring coverage and cost.

Active Series

Each check generates multiple Prometheus time series depending on the protocol. More series means richer data, but also more storage:

Protocol	Approximate Series per Endpoint
HTTP	~10 (success rate, response time percentiles, status codes)
DNS	~6 (success rate, lookup time percentiles)
ICMP	~8 (success rate, latency percentiles, packet loss)
HLS	~20 (buffer health, TTFB, segment durations, histograms)

Series count affects your Prometheus/Grafana Cloud storage costs if you use Remote Write to push metrics to an external destination. It does not affect your Bitping unit billing.

Tips to Manage Costs

• Start with longer intervals — 5-minute checks are sufficient for most uptime monitoring. Reserve 10-30 second intervals for critical endpoints.
• Use the usage estimate — Check the estimated cost on the config page before saving changes.
• Review endpoint count — Remove endpoints you no longer need to monitor.
• Consider your plan — If you consistently exceed your included units, upgrading to the next tier is usually cheaper than paying overage.

Viewing Your Usage

Navigate to Account > Billing to see your current billing period usage, included units remaining, and overage charges. The Account Overview dashboard also shows a usage summary.