Who does it and how:
The CTO or a lead developer owns email infrastructure at most Indian SaaS companies below Series B. The work looks like one of these:
- The EC2 Postfix route: Spins up an AWS EC2 t2.micro, installs Postfix, configures SPF/DKIM/DMARC, then fights Spamhaus and SORBS blocklistings every few months. Time cost: 2–4 hours per incident, 3–6 incidents per year.
- The Gmail API hack: Developer tunnels transactional email through Gmail SMTP for small-volume apps (under 500 emails/day). Hits Google rate limits, gets the account flagged, wastes half a day resolving it with Google Workspace support.
- The shared hosting SMTP trap: SMB SaaS companies on GoDaddy, BigRock, or Hostinger shared hosting use the hosting provider's SMTP. Deliverability to Jio, Airtel, and BSNL email addresses is poor because shared IPs are already flagged. Founders don't know why their "password reset" emails never reach customers.
- The SendGrid/Mailgun onboarding tax: Company signs up for SendGrid, spends 3–5 hours configuring the API, writing webhook handlers for bounces and complaints, setting up suppression lists. Then a DevOps engineer leaves and institutional knowledge walks out the door.
- DevOps time on email infrastructure: estimated 4–8 hours/month for a team under 20 people. At ₹1,500/hour fully-loaded cost, that's ₹6,000–₹12,000/month in engineering time, invisibly.
- Lost emails that never reach customers: password resets, order confirmations, payment receipts. Founders rarely measure this — they only notice when a customer complains.
- Emergency firefighting when SMTP gets blocklisted: a t2.micro on Spamhaus means 100% of emails fail until delisting, which takes 2–48 hours. For a SaaS doing 50+ orders/day, this is a revenue event.
- Multiple SaaS tools each doing their own SMTP: the marketing tool uses one SMTP, the product uses another, the billing tool uses a third. No unified reputation management.