Why we do not track cold email open rates: a metric that distorts what it measures

Open rate is the number most often shown in cold email reports. A high figure looks good on a slide, which is why many agencies open the results conversation with it. We made the opposite decision: at Konvertia we deliberately do not track open rate and we do not report it to clients.

This is a methodological choice, not an oversight, and two arguments sit behind it. The first is technical: measuring opens requires a tracking pixel, which can reduce deliverability by itself, so the measurement damages the result it is meant to describe. The second is commercial: an open says nothing about what the recipient intends to do, so even a perfectly measured rate gets you no closer to a sales meeting. Below we take both arguments apart and show what we measure instead.

Open rate does not appear out of thin air. It depends on a tracking pixel, an invisible image added to the body of the email. The sending tool inserts a tiny image, usually one pixel across, hosted on a tracking server. When the recipient opens the message, their email client downloads that image, and the server logs the download as an "open". The number of those events, divided by the number of delivered emails, gives the open rate.

One consequence matters for everything that follows: to count opens, you have to modify every message you send. A plain, personal note gets a foreign element attached to it that calls out to an outside server. The email stops looking like ordinary correspondence between two people and carries the technical signature of a marketing campaign. Spam filters are built to notice exactly that signature.

A tracking pixel has a cost that campaign reports rarely mention: it can lower the deliverability of the very email it is attached to. Some spam filters recognise tracking elements and read them as a sign of bulk marketing mail rather than personal correspondence. A message carrying a pixel is more likely to land in spam than the same message without one.

That creates a paradox which undermines open rate as a measurement. The harder you try to measure opens, the fewer emails reach the inbox, and the worse the campaign you are trying to describe performs. The act of measuring changes the thing being measured. In our cold email campaigns dropping the pixel is one of the pillars of deliverability, alongside sending infrastructure and personalisation of every message.

A mailbox that sends clean messages with no tracking elements builds a better sender reputation over time. So giving up open rate is not a loss of data. It is an investment in the outcome: more of your emails arrive where they are supposed to.

Even if the pixel cost nothing, open rate would still tell you very little, because opening an email is not a statement of interest. A recipient can open a message and delete it in the same second. They can open it on the move between two meetings. From the open alone you cannot tell whether the message landed or was dismissed by reflex.

There is a deeper problem too: some "opens" do not come from people at all. Some email clients, Apple Mail with Mail Privacy Protection among them, can download images in advance, before the recipient has decided whether the message is worth reading. Corporate security systems scan incoming mail and can trigger the pixel as they do so. The rate goes up even though nothing meaningful has happened on the other side.

It is a textbook vanity metric: it looks good in a report and gives a sense of progress, but it does not move you one step closer to revenue. Pipeline does not grow from opens. It grows from conversations.

The success of a cold email campaign is measured in conversations, not impressions. So we report three numbers, each of which needs a deliberate action from the recipient:

Together these three numbers form a funnel that an automatic image download cannot inflate. In our campaign for a B2B services company it looked like this: 800 prospects, a 14% reply rate, almost half of the replies positive, and more than 20 booked meetings from a single sequence sent over two weeks.

How measurement fits into the whole process is described on our how it works page.

If an agency opens its report with the open rate, ask one question: how many of those opens turned into conversations? It shows within seconds whether there is pipeline behind the impressive number or only a pixel. An agency that tracks replies and meetings will answer with specifics. One that leans on open rates will start explaining methodology.

A high open rate does not mean a campaign is bad. It only means someone chose to track it and accept what that costs.

Open rate looks like a hard metric, but it is really a count of image downloads logged by a sending tool. Measuring it can lower deliverability, which worsens the very thing it is meant to describe, and an open says nothing about the recipient's intent. That is why we deliberately do not track it and report only numbers that stand for a human decision: replies, positive replies and booked sales meetings. If you would like to see how that looks in practice, talk to us about your pipeline.