A/B Testing Bandwidth Cost Formula
Learn how annual A/B test data-transfer cost is estimated from traffic, page weight, traffic allocation, and CDN pricing.
This calculation estimates the annual delivery bandwidth and CDN cost for page views included in an A/B test. It also compares the mixed control-and-variant cost with a control-only baseline, helping quantify the delivery impact of a page-weight change.
- 100% Free
- No Sign-Up Required
- Private & Secure
- Mobile Friendly
Annual A/B Test Bandwidth Cost
Where:
Estimate monthly test page views, split them between control and variant, multiply each group by its transferred page weight, convert KB to decimal GB, annualize the amount, and multiply by the CDN price per GB.
Variables Explained
| Variable | What It Means | Unit |
|---|---|---|
| monthlyVisitors - Monthly visitors | Average unique visitors to the site per month. | visitors |
| pagesPerVisitor - Pages per visitor | Average page views generated by each visitor. | pages |
| testTrafficPercent - Traffic included in test | Percentage of all page-view traffic eligible for the experiment. | percent |
| controlPageWeightKb - Control page weight | Average transferred page weight of the control experience. | KB |
| variantPageWeightKb - Variant page weight | Average transferred page weight of the variant experience. | KB |
| variantTrafficPercent - Variant traffic split | Share of test traffic receiving the variant. | percent |
| cdnCostPerGb - CDN cost per GB | Effective data-transfer charge applied to each decimal GB. | currency |
Step-by-Step Calculation
Estimate monthly page views
Convert monthly visitors into estimated monthly page views.
monthlyPageViews = monthlyVisitors * pagesPerVisitor
Apply the test allocation
Keep only the page views that participate in the experiment.
monthlyTestPageViews = monthlyPageViews * (testTrafficPercent / 100)
Split traffic between experiences
Assign test page views to the variant and control.
monthlyVariantPageViews = monthlyTestPageViews * (variantTrafficPercent / 100); monthlyControlPageViews = monthlyTestPageViews - monthlyVariantPageViews
Calculate monthly test bandwidth
Multiply views by page weight and convert KB to decimal GB.
monthlyTestBandwidthGb = ((monthlyControlPageViews * controlPageWeightKb) + (monthlyVariantPageViews * variantPageWeightKb)) / 1000000
Annualize bandwidth
Assume the monthly traffic pattern continues for 12 months.
annualTestBandwidthGb = monthlyTestBandwidthGb * 12
Price the annual transfer
Apply the effective CDN or data-transfer price.
annualTestBandwidthCost = annualTestBandwidthGb * cdnCostPerGb
Calculate incremental cost versus control
Compare the mixed experiment delivery cost with serving the control to all test traffic.
annualIncrementalCost = annualTestBandwidthCost - ((monthlyTestPageViews * controlPageWeightKb * 12 / 1000000) * cdnCostPerGb)
Example: 100,000 monthly visitors and a heavier variant
Monthly page views
100,000 × 3
300,000 page views
Monthly test page views
300,000 × 50%
150,000 page views
Control and variant views
150,000 × 50% for each experience
75,000 control and 75,000 variant views
Monthly transfer
((75,000 × 1,500) + (75,000 × 1,700)) ÷ 1,000,000
240 GB
Annual transfer
240 × 12
2,880 GB
Annual test cost
2,880 × $0.08
$230.40
Incremental cost
$230.40 - $216.00
$14.40
Final Result
The estimated annual A/B test bandwidth cost is $230.40 for 2,880 GB of transfer, or $14.40 more than the control-only baseline.
Assumptions
- ✓Monthly visitors, pages per visitor, test allocation, and traffic split remain stable for a full year.
- ✓One GB equals 1,000,000 KB, reflecting decimal data-transfer billing.
- ✓The page-weight values represent average transferred bytes per relevant page view.
- ✓The entered CDN price is an effective single rate for all experiment transfer.
Limitations
- !Actual traffic can vary by season, campaign, device mix, and experiment duration.
- !CDN bills may use regional prices, tiers, included allowances, minimum charges, or other pricing rules.
- !Cache behavior, compression, retries, and protocol overhead can cause real transfer volumes to differ.
- !The calculation excludes A/B platform subscriptions, event tracking, API calls, and third-party service costs unless included in page weights.
Common Mistakes to Avoid
Entering an uncompressed design-file size instead of measured transferred page weight.
Using total site traffic when only a subset of pages or visitors can enter the test.
Treating the variant split as a percentage of all site traffic rather than of test traffic.
Mixing binary GB assumptions with a CDN rate billed in decimal GB.
Assuming the incremental result includes testing-platform or analytics fees.
Related Formulas
Frequently Asked Questions
How is A/B testing bandwidth cost calculated?
The calculation multiplies test page views by the weighted average of control and variant page weights, converts KB to decimal GB, annualizes the result, and applies the CDN cost per GB.
What does annual incremental cost mean?
It is the estimated difference between the test's mixed control-and-variant delivery cost and the cost of serving the control page to the same test traffic.
Can an A/B test reduce bandwidth cost?
Yes. If the variant is lighter than the control, the incremental cost can be negative, indicating an estimated delivery-cost reduction for the tested traffic.
Why are pages per visitor included?
Visitors can produce multiple page views. Pages per visitor converts visitor volume into the page-view estimate used to calculate transferred data.
Does a 50/50 test double bandwidth?
No. Each eligible page view receives either the control or the variant, not both. The total depends on the weighted average page weight of the two experiences.
Ready to calculate your result?
Use the calculator to get instant results with your own inputs.