Where Gmail trackers differ once the pixel loads

If you are choosing between Mailtrack and Streak on accuracy, start from this: they detect an open in the same way, so neither one sees something the other cannot. Both put a tiny image in your email and record a view when that image is fetched. Where they differ is in the rules applied after the fetch arrives. Which loads get thrown away as your own? Which person in a group gets the credit? What happens when the email travels further than you sent it? Both companies answer these questions in their own help centres, more candidly than most comparison pages do, and the answers are what decide whether the number in your Sent folder means anything.

One pixel under every tracker

Mailtrack, which now sits inside the Mailsuite product family, describes the mechanism in its guide to tracking opens: it adds a very small image to each email, and when the recipient opens it, the image is downloaded and Mailsuite's servers are told. Streak's explanation is the same idea in different words, and adds that if the image is never fetched, Streak receives nothing at all. Its own tracking overview calls email tracking fallible. That is an honest word, and it applies to every product in this category, including ours.

Your own rereads

The most common false open is the sender. You send a proposal from your laptop, then reread it on your phone in the lift. Both vendors try to filter this, and both say where the filter stops. Mailsuite's self open article says the detection works when you view the sent email in a computer browser with the extension installed, or in Mailsuite's own mobile app. Open your sent copy in the Gmail app on Android instead, and the open is recorded as if the recipient had read it. Streak's equivalent page is narrower still: your activity is blocked only when you view sent mail in Chrome with Streak installed, and any other computer, browser, tablet or phone registers a view.

This matters more than it sounds. The moment a salesperson most often rereads a sent email is on a phone, shortly before a call, which is exactly the moment both filters are documented not to cover.

Groups and forwards

Send one email to three people and the pixel is shared, so the tracker has to decide whom to credit. Streak's answer is that it cannot: it reports the total number of views, and the approximate location may hint at who it was. Mailsuite's group tracking page says its paid Mailsuite plan will name who in the group opened, with listed exceptions: scheduled emails, emails to more than 50 people, emails sent from your phone, team addresses, and anything sent before you upgraded.

Forwards are the same for everyone. Mailsuite's forwarding article says an open by the new reader is attributed to the original recipient, and suggests reading an unusually high open count as a sign the email was passed around. Streak's page on inaccurate view totals says forwarded views are counted too, and describes a stranger case: a user copied a signature out of a tracked email, carried the invisible pixel into every message they sent afterwards, and made one old email look as if it had been read hundreds of times.

Opens nobody caused

Mailsuite publishes a short list of why an open alert can arrive immediately after sending: the recipient's email service fetching the pixel automatically, a CRM that processes copies of your mail, and an email program that previews your sent messages. It adds that a sync or refresh can make old conversations fire several alerts at once. Its advice for the first case is sensible: ignore the instant open and watch for a later one from the same person. Streak notes the opposite failure: a reader who blocks images can read and reply without any view being recorded.

A ten minute test before you choose

You do not need to take anyone's word for this, ours included. Install the tracker you are considering and run five sends. Use a second address you own as the recipient, because Mailsuite does not track mail sent to the address you send from, and Streak's test guide suggests opening the test in a private browser window.

• Send to your second address and open it once on a desktop with images on. You should see exactly one open.
• Reread the sent copy in the Gmail app on your phone. If the count moves, the tracker is counting you.
• Send one email to two of your addresses and open only one. Note whether the tracker names the right address or only says someone opened.
• Forward that email to a third address and open it there. Note who gets the credit.
• If you have one, send to a work address behind a corporate security gateway and note whether an open appears within seconds of sending.

Write down what each step showed. Ten minutes of this tells you more about a tracker's accuracy than any feature table, because it tests the filter on your own devices.

How Fast Mail Tracker handles the same cases

We built the sender case differently, because a filter that depends on where you reread an email will always have gaps. When you send through Fast Mail Tracker with Gmail connected, the copy in your Sent folder is written separately from the copies your recipients get. It carries its own pixel and plain, unwrapped links. A follow up check then reads your mailbox and replaces any copy that turns out to carry a recipient's tracking. So when you reread the email on your phone, the hit is logged as Sender copy and never counted as an open, whichever device or app you use.

Fetches from known security scanners, and any open within three seconds of sending, are kept out of open counts and notifications. For a group, you can send each person their own copy while the To line still shows everyone; the default is one shared copy, and an open on a shared copy belongs to the group rather than a named person. The limits that remain are the ones above: a forwarded copy still looks like its original reader, and a scanner that imitates a normal browser and waits a few seconds can still be counted.

Fast Mail Tracker is in early access while Google's security review completes. The install button below this article is the place to start, and the five sends above are a fair way to judge it.