Is My Email Newsletter Actually Bringing In Business?

You send the newsletter on a Thursday morning. By Friday afternoon your email tool has produced two numbers: an open rate that looks respectable and a click rate that looks less so. You glance at them, feel vaguely reassured or vaguely disappointed, and get on with your day. The question you actually had when you sat down to write the thing — did this bring me any work — is exactly as unanswered as it was before you pressed send.

This is not your fault. The email industry has spent twenty years reporting on the email and almost nothing on what happens afterwards, which is the only part that pays for anything. The good news is that the fix is small, one-off, and takes about twenty minutes. You put a short label on the end of the links in your newsletter, and from then on your website can tell you how many people arrived because of that email, what they looked at, and how many of them got in touch.

Here is how an open is counted. Your email contains one invisible image, roughly the size of a full stop, hosted on your email provider's servers. When someone's email programme downloads that image, the provider records an open. That was a reasonable proxy for attention in about 2005.

It stopped being one when Apple's Mail app started fetching those images automatically for everyone who uses it, whether or not they read a word — or ever will. Corporate spam filters do something similar, and some of them click every link in the email to check where it goes. Between those two, a decent chunk of your reported opens and a meaningful slice of your reported clicks belong to software, not customers. In my experience the distortion is bad enough that a rise in open rate from one month to the next tells you more about which email programmes your list happens to use than about whether the content landed.

The practical consequence: stop setting targets against open rate. Keep an eye on it only as a rough alarm — if it falls off a cliff, something has broken, like your emails landing in spam folders. Otherwise ignore it.

Clicks are more honest than opens, because a real click usually requires a real finger. But your email tool's knowledge ends the instant someone leaves the email. It cannot tell the difference between a person who clicked your link, saw a page that did not match what they expected, and left in four seconds, and a person who clicked the same link, read three pages, checked your prices, and filled in the contact form.

Both of those are one click. If you judge the newsletter on click rate, you are treating those two people as identical, which they emphatically are not. What you want to know is what happened on the other side of the door — and only your website can tell you that.

When someone lands on your site, your analytics can usually see where they came from — a search engine, another website, a social post. But a click from an email often arrives with no source at all, because the person's email programme does not pass that information along. Those visits get lumped in with everyone who typed your address in directly, and your newsletter's work disappears into a pile marked "direct".

The fix is a campaign tag: a short label you stick on the end of a link so the visit identifies itself when it arrives. A tagged link looks like this: yoursite.co.uk/spring-menu?utm_source=newsletter&utm_medium=email&utm_campaign=march-scones. Everything after the question mark is the label. It changes nothing about the page — the visitor sees the same thing they would have seen anyway. It simply means your analytics can now say "this person came from the March scones email" instead of shrugging.

You do not need to build these by hand. Most email platforms have a setting, usually called link tracking or campaign tracking, where you type the campaign name once and it tags every link in that email automatically. If yours does not, our guide to campaign links walks through building them in a couple of minutes. The one rule that matters more than any other: be boringly consistent with your spelling. "newsletter", "Newsletter" and "news-letter" will be reported as three separate sources, and you will spend a happy afternoon wondering why your numbers do not add up.

You also need somewhere for the tags to land. Any competent analytics tool reads them; if you are still deciding what to use, our plans and what each one includes are worth two minutes before you set anything up, because campaign reporting is a feature some tools tuck behind a higher tier.

Visits alone will flatter you. A hundred people arriving from an email feels like success right up until you notice none of them did anything. So pick, in advance, the one or two actions on your site that genuinely mean business is coming: the contact form being submitted, the booking page being reached, someone tapping the phone number on their mobile, a quote request going through.

In analytics these are usually called goals or conversions — a plain word for "the thing I wanted them to do". You set one up once and from then on every visit is measured against it. Our notes on setting up goals and what counts as a conversion cover the mechanics, and neither takes long.

Choose something that a real customer would have to work for. "Visited the services page" is not a result. "Filled in the enquiry form" is.

Suppose you are a solicitor with a list of eight hundred people, and you send a monthly note about something practical — the paperwork people forget when they buy a flat, say. Your email tool reports a healthy-looking open rate that you have now decided to ignore, and twenty-nine clicks.

With tags on the links, your site shows something more useful: twenty-four visits from that campaign, of which nineteen were people who had not visited in the previous month, an average of just over two pages each, and three enquiry forms submitted. Three enquiries from eight hundred emails sounds thin until you work out what a single conveyancing instruction is worth to you, and how many hours the newsletter actually took to write.

Now compare that with the previous month's send, which went out on a Sunday evening and produced forty-one visits but only one enquiry. Same list, similar effort, very different outcome. That contrast is a real piece of information — it points at the subject, the timing, or the page you sent people to — and none of it is visible in an open rate.

It is worth writing a one-line note in your analytics on the day each newsletter goes out, so that in four months you can look at a spike and remember what caused it. Annotations exist for exactly this, and they save you a great deal of squinting at calendars later.

Small numbers wobble. If one newsletter produces four enquiries and the next produces one, that difference may be entirely down to who happened to be moving house that fortnight. Three or four sends in a row moving in the same direction is a signal. One send is an anecdote.

So do the sum quarterly. Add up the visits and the enquiries attributed to the newsletter across three months, and set that against the honest cost of producing it — usually your own time, which is the most expensive thing you own. If the answer is clearly positive, send more often. If it is clearly negative after three or four honest attempts with different subjects and different landing pages, stop, and put the hours into whichever channel is actually working.

If you would rather not remember to do this, a weekly summary with ranked suggestions will surface it for you, including the awkward observation that your newsletter traffic has been falling for a month.

Some of your newsletter's effect is invisible and always will be, and anyone who tells you otherwise is selling something. Someone reads it on their phone, does nothing, and rings you three weeks later — that call will show up as a phone enquiry with no source. Someone forwards it to a friend, who clicks the tagged link, and the friend gets counted as newsletter traffic even though they were never on your list. Someone reads it, remembers your name, and searches for you on Google a month on; that visit is recorded as a search visit, which is technically true and completely misleading.

This is the general problem of delayed and word-of-mouth business, and it means your measured numbers are an undercount rather than an overcount. That is a comfortable direction for the error to run in. If the newsletter looks worthwhile on the numbers you can see, it is probably a little better than that. If it looks like nothing at all on measured visits and enquiries across a whole quarter, the invisible portion is unlikely to be rescuing it.

One small habit closes part of the gap at no cost: ask new enquirers how they heard about you, and write the answer down. It is rough, people misremember, and it is still the only sight you will get of the offline half.

Turn on campaign tagging in your email platform and give the campaign a consistent, lowercase name. Decide the one action on your site that counts as a result and set it up as a goal. Send the next newsletter as you normally would. Then, the following week, look at three numbers only: how many people arrived from it, how many pages they looked at on average, and how many completed the action you chose.

After four sends you will know more about your newsletter than the open rate would have told you in four years. And you will find you start writing differently — sending people to one specific, useful page rather than a general "read more on our website", because you can finally see which approach produces enquiries and which produces polite scrolling.