Most downtime calculators do "annual revenue ÷ 8,760 × hours offline" and stop. That misses 60-80% of the real cost. A real outage hits four lines on your P&L: direct revenue lost during the window, customer LTV erosion (a fraction of affected users churn permanently), brand cost (the implicit CAC for re-acquiring trust), and engineering recovery overhead (the team-hours spent triaging, fixing, and writing the postmortem). This calculator layers all four with industry-calibrated weights from Gartner, ITIC 2024, Forrester TEI, and Atlassian / Statuspage research.
Layer 1 — Direct revenue: your annual revenue divided by hours in the year, weighted by industry (fintech 3×, e-commerce 1×, SaaS B2B 0.7×) and time-of-day (peak 2.5×, overnight 0.3×). Layer 2 — Customer LTV erosion: a fraction of users who tried to access your site during the outage will churn permanently; for fintech this is 8%, for restaurants 2%, multiplied by your customer LTV. Layer 3 — Engineering recovery: the team-hours spent on triage, fix, postmortem, and customer comms × fully-loaded hourly rate. Layer 4 — Brand cost: affected users × industry brand-sensitivity multiplier × estimated CAC. Total bleed is usually 3-5× the naive direct-revenue number.
Gartner's widely-cited $5,600/min average is enterprise-IT-wide — every system, including internal ERPs, supply chain, CRM, and customer-facing. For SMB e-commerce and solo entrepreneur scenarios, that number is usually too high; bottom-up modeling from your actual revenue and customer count is more accurate. ITIC's 2024 survey shows 91% of mid-large enterprises report >$300K/hr cost and 44% report >$1M/hr — also enterprise-skewed. Use this calculator for SMB-to-mid-market sizing; cite Gartner for enterprise narratives.
A SaaS B2B platform that's down for an hour mostly delays work — users come back. An e-commerce site during BFCM permanently loses the cart. A fintech that's down during payroll day has direct customer financial harm. The weight reflects how much of an outage hour translates to permanent revenue loss versus deferred revenue. Calibrated from ITIC/Forrester vertical data.
Hand-wavy if you use a flat number; defensible if you tie it to user behavior. Our model treats brand cost as the implicit CAC for re-acquiring trust from users who experienced the outage but didn't churn. We use an industry brand-sensitivity multiplier (5-50%) × estimated CAC × affected users. Conservative — major incidents like the 2017 AWS S3 outage created brand cost in the $100M+ range for affected customers.
If your site is down >24 consecutive hours, Googlebot starts treating it as a soft 404. Recovery typically takes 7-30 days of degraded rankings. We don't model this because it's heavily site-specific and most outages are <4 hours. If you've had a multi-day outage, expect 10-30% organic traffic loss for ~2 weeks. Use Search Console's Coverage report to monitor.
Because 99.9% SLA = 8.76 hours/year of downtime. If your incidents are typical, that's a lot of outages. The 99.99% projection assumes 52 min/year — much closer to acceptable for revenue-critical sites. Use these numbers to justify uptime engineering investment to leadership.
Yes — the SaaS B2B preset weights direct revenue lower (0.7×) since most downtime delays rather than destroys value, but raises engineering brand cost (0.30 multiplier) since enterprise customers track SLA breach against renewal probability. Total cost still accumulates meaningfully on multi-hour or repeated outages.
Sources: Gartner · ITIC Hourly Cost of Downtime 2024 · Forrester TEI 2022 · Atlassian / Statuspage 2023 · Akamai · Baymard Institute · OpenView 2023