Skip to content

How open and click tracking works

JungleMail 365 counts opens with a tiny image in each email and clicks with links that pass through JungleMail on their way to their destination. Both are added when the newsletter is sent, one set per recipient, so every open and every click can be tied to the email it came from. How they work tells you how far to trust each figure: the open rate is an estimate, usually too low, and the click rate is the firmer signal.

Each email carries an invisible image, one pixel in size, that is unique to that email. When the recipient’s email program loads the newsletter’s images, it also asks JungleMail for that image, and JungleMail records an open. The image comes from your organization’s own JungleMail pages domain, such as enovapoint.eu.junglemailpages.com.

An email program that doesn’t load images never asks for the image, so the open isn’t counted, even when the recipient reads the whole newsletter. JungleMail makes up for part of that: if the recipient clicks a link, answers a survey or unsubscribes, the email is counted as opened too.

The links in the newsletter point to JungleMail first. When a recipient selects one, JungleMail records the click and sends the recipient on to the link’s real address at once. Recipients see the page they expected; the detour isn’t visible to them.

  • Unique counts each email once. A recipient who opens the newsletter three times and clicks two links is one unique open and one unique click.
  • Total counts every time: three total opens and two total clicks for the same recipient.

Rates are worked out from the emails sent. The open rate is the unique opens divided by the emails sent, and the click rate the unique clicks divided by the emails sent. Emails that bounce stay in the emails sent, so they lower both rates slightly. The click-to-open rate on the click map is the unique clicks divided by the unique opens.

With each open, the recipient’s email program says what it is and what it runs on. JungleMail uses that to sort opens into desktops, phones and tablets, which the reports show as Reading Devices.

  • Images that aren’t loaded. Classic Outlook doesn’t download pictures from the internet until the recipient allows them, partly so that senders can’t see who read their email (Block or unblock automatic picture downloads in classic Outlook email messages), so many opens are missed there. An administrator can let Outlook load the newsletter’s images for everyone (Load newsletter images automatically in Outlook).
  • Automated checks. Email security scanners and image proxies can load images and follow links on a recipient’s behalf. JungleMail leaves out the requests it recognizes as automated, by the program and the network they come from, but some look like people and are counted.
  • Tracking turned off. Opens are counted only if the newsletter was sent with Track email opens on, and clicks only with Track clicked links on. Both are on for a new newsletter (Delivery, tracking and report options).

A site administrator can turn on Enable anonymous tracking of email opens and link clicks (Security and privacy settings). JungleMail then still counts opens and clicks, but stores no email addresses or other data that tell who opened or clicked: lists of recipients show pseudonyms, such as Pseudonym 12, instead of addresses. The setting applies to newsletters sent after it is turned on.

  • Treat the open rate as a minimum. Some recipients read the newsletter without being counted. Comparing newsletters sent to the same people is more reliable than comparing with another organization’s figures. If the open rate looks low, see Images don’t load, or open rates look low.
  • Trust clicks more than opens. Apart from the scanners JungleMail doesn’t recognize, a click is made by someone who read the newsletter and wanted more, whatever their email program does with images. The click rate is the firmer signal.
  • Give each newsletter a link worth selecting. A newsletter without links can be measured by its opens only. One clear next step, such as “Read the full policy on the intranet”, gives you a click rate and counts the opens of readers whose pictures stay blocked.
  • Set up automatic image downloads before you follow results over time. A change in how many opens can be counted looks like a change in readers: if your IT team makes Outlook load the images in March, the open rates from March on aren’t comparable with those before.
  • Mark trial sends as tests. Newsletters you send to try things out can then be left out of your results (Find and compare newsletter results).
  • Know whether bounces are reported for your sending account. Bounces are reported only for some sending accounts; see Bounces and rejections.