Unit Economics: Track Cost Per Business Metric

Unit Economics

Unit economics lets you answer questions like “How much does it cost to serve one customer?” or “What’s our cost per API call this month?” Instead of looking at raw cloud spend, you track cost relative to business output.

Why Unit Economics

Raw cloud cost going up isn’t necessarily bad. It might mean your business is growing. What matters is whether your cost per unit is stable, improving, or degrading.

Scenario	Raw Cost	Customers	Cost/Customer	Verdict
January	$8,000	200	$40	Baseline
February	$12,000	350	$34	Improving
March	$15,000	360	$42	Degrading

In this example, March’s cost increase is a real problem: you’re spending more per customer. February’s increase was healthy growth.

How It Works

1. Define a metric. On the Unit Economics page, click New metric. Give it a name (for example API requests), a unit (request), the cost basis to divide (Effective (amortized), Billed or List (on demand)), a scope (accounts, providers, services, regions, tags, or one virtual tag), and how daily values roll up into weeks and months (Sum, Average, Maximum or Last value). Click Create metric.
2. Add daily values. Upload them by CSV, or store one value across a date range (see below).
3. Xplorr matches them with cost. Each day’s value is joined with that day’s scoped cloud cost, and cost per unit is scoped_cost / units.
4. Trends are tracked. The Trend card compares the second half of the selected range against the first half, so you can see whether cost per unit is rising or falling.

Adding Values

Select a metric and use the Add values card. A new value for a date replaces the old one, so re-uploading a corrected file is safe.

CSV upload. Two columns, date (YYYY-MM-DD) and value. A header row is optional. Paste the rows and click Upload pasted CSV, or click Choose file to upload a .csv file up to 900 KB.

date,value
2026-09-01,120000
2026-09-02,118400

Same value for a date range. For slow moving metrics, such as a customer count that holds for a week, pick a start date, an end date and a value, then click Store for each day.

Reading Cost Per Unit

Once a metric has values, the page shows four cards for the selected range:

Card	What it shows
Cost per unit	Scoped cost divided by units, over complete days in the range
Scoped cost	Cloud cost for the metric’s scope in the range
Total units	Units in the range, and how many periods have values
Trend	Second half of the range vs the first half

The Cost per unit over time chart plots the daily figure. Days without a value are left out, so gaps in the chart mean missing uploads rather than zero cost.

MCP & Slack Access

Unit economics data is also accessible via:

• MCP Server: use the get_cost_summary tool and correlate with your own metrics.
• Slack Bot: ask questions like “What’s my cost per customer this month?” (if you’ve uploaded customer metrics).

Best Practices

• Keep values daily. Unit economics works best with daily granularity. Export a daily CSV from your application database, analytics platform, or billing system and upload it on a regular schedule.
• Pick 2-3 metrics. Don’t track everything. Pick the metrics that matter most to your business: customers, API calls, or transactions.
• Choose the right roll up. Use Sum for counts like orders and Average or Last value for gauges like active customers, or weekly and monthly figures will be wrong.
• Review weekly. Check the Trend card weekly. A gradual increase in cost-per-unit is easy to miss if you only look at raw spend.
• Correlate with anomalies. If cost-per-unit spikes, check the anomaly detection page to see if a specific service is the cause.