WhatsApp and email cannot be ranked by a single "open rate", because the two channels do not measure the same event. Email reports an open when a tracking image loads, which a privacy feature can trigger without a human. WhatsApp reports a message as delivered, read or answered through status webhooks, and a person can switch read receipts off. Any headline figure that puts one channel's "opened" next to the other's "read" compares two different things, and the figures that circulate most widely come with no primary source behind them.
This guide covers what a brand can actually measure on each channel, why delivered, read and replied are three different questions, how cost and consent change the comparison, and the cases where email remains the better choice.
What each channel can really report
Email: sent, delivered, opened, clicked. A delivered email means the receiving server accepted it, not that it reached the inbox rather than the spam folder. An "open" is inferred from a tiny remote image loading. Apple's Mail Privacy Protection downloads remote content in the background "regardless of whether you engage with the email", and hides the recipient's IP address behind relays. Apple Mail users who enable it can therefore register as opens without reading anything, which inflates open counts and removes reliable location and timing data.
WhatsApp: sent, delivered, read, failed. The Cloud API reports message statuses through webhooks, and a message moves from sent to delivered and then, if the recipient opens the chat, to read. The read status depends on the recipient's own read receipt setting, so a missing "read" does not prove the message went unseen. Unlike an email pixel, though, a read event on WhatsApp is generated by the chat being opened on the person's device, not by an automated download.
Replies are the cleanest signal on both. A human reply cannot be produced by a privacy proxy. On WhatsApp it also opens the customer service window, so the reply is both a measurement and a commercial event. Clicks sit in between: useful, but link scanners at some mail providers can fire them too.
Delivered, read, replied: three different questions
Delivered asks whether the channel did its job. On WhatsApp you pay only when a message is delivered, so this number ties directly to your invoice. On email it is a technical acceptance and says little about attention.
Read asks whether the message was seen. It is the number everyone wants and the one least comparable across channels. A WhatsApp read is a conservative lower bound, because receipts can be off. An email open is an upper bound of uncertain size, because machines can trigger it.
Replied asks whether the message mattered. This is the only metric worth putting in a cross-channel report, together with the downstream result you control: orders, bookings or unsubscribes attributed to the send. Measure each channel against its own previous sends first, then compare outcomes per message sent and per euro spent.
Why the comparison tables mislead
A bound is not a rate. Comparing a lower bound on one side with an inflated upper bound on the other produces a gap that says nothing about attention. Audiences differ too. People who gave a phone number and agreed to WhatsApp marketing are a smaller, more selected group than an email list built over years, so a better result may reflect who is on the list, not the channel. If a vendor quotes a single percentage for either channel without naming how it was measured, on which list and over what period, treat it as marketing copy and leave it out of your own planning.
Cost and consent change the maths
WhatsApp has a unit cost. Since 1 July 2025 Meta bills each delivered template message, priced by category and recipient country, as covered in WhatsApp per-message pricing explained. From 1 October 2026 Meta also charges for service messages, with a free tier of 1,000 service messages per business phone number per month. Email's marginal cost per send is close to zero once you pay for the platform.
WhatsApp needs its own consent. In Italy a marketing opt-in collected for email does not cover WhatsApp, and the collection has to be recorded and revocable, as set out in WhatsApp opt-in and GDPR in Italy. Starting a WhatsApp programme therefore begins with building a new, smaller list, while the email list already exists.
Templates constrain the message. Outside the customer's window you can only send approved templates, explained in WhatsApp template messages. That is a different creative format from a long, designed email.
When email is the better choice
Long or visual content. A newsletter, a catalogue, a detailed receipt with attachments or a multi-section announcement fits email. WhatsApp templates are short by design, and a long chat message is easy to ignore.
Large lists at near-zero marginal cost. If the goal is broad, low-urgency reach, paying per delivered message is hard to justify when email costs almost nothing to repeat.
Contacts who never gave a WhatsApp consent. Without a separate, documented opt-in you cannot send marketing there at all. For those contacts email is not the second-best channel, it is the only lawful one.
Archival and forwarding. Invoices, contracts and anything people search for months later are easier to find in an inbox. In practice most brands do best using both: WhatsApp for time-sensitive, conversational moments and email for depth, each judged by replies and outcomes.
Putting it into practice
Pick one outcome, such as recovered orders or booked appointments, and log delivered, read and replied for WhatsApp next to sent, clicked and converted for email. Ignore opens as a decision metric. TikJoy runs the official WhatsApp Business API and reports message statuses and replies alongside the conversation. Check Meta's pricing documentation for current rates before modelling costs (guidance as of October 2026).