Home Resources Blog

At a time when the popular look on the web is one of simplicity, it’s tempting to apply the same ideas to email. Less is more. The fewer prompts the better. But does this result in more click throughs, or are we simply indulging in form, over function?

I started asking these questions after attending a talk on essentialism, or the disciplined pursuit of less, but better. Championed by Plato, Gandhi and Steve Jobs, it’s an idea loaded with promise and apparently, few downsides; the essentialist spends their time on meaningful projects, travels light and naturally, designs the most spartan of email campaigns.

The problem, of course, is that while it’s attractive to think that less links in an email campaign (and less surrounding content clutter) should result in higher engagement, we’ve never looked at the numbers. That is, until now.

A case for more

With the hypothesis that HTML email campaigns with less unique URLs and/or links would have a higher click rate (% click/open) than those with more, we looked at the behavior of over 500 million email recipients, from email campaigns sent to over 500 subscribers in recent months. Each email campaign was categorized by unique URLs and links therein, to allow us to determine whether having less to click on resulted in a more engaging email experience.

Here are how click rates perform as the number of links increase in email campaigns:

Links vs Click Rate

For comparison, here are how click rates perform as unique URLs increase:

Unique URLs vs Click Rate

Increasing the number of unique links and URLs has a strong positive effect on click rate until you reach 11 or so – after which, it see-saws around the 17% mark, but never really takes a dive. So overall, it seems that the more links, the higher the click rate. This is upsetting for anyone who dotes on the single call-to-action approach to email design, or worse still, has a client intent on packing in an extreme amount of content.

The other thing worth noting is the relationship between links and unsubscribes. Email campaigns with no links at all receive an unsubscribe rate of 0.45%, or roughly 55% more unsubscribes on average than those with 1 or more links. Unsubscribe links were excluded from our link total.

Did we just bury simplicity?

Like with many things, if you narrow your focus down to simply the numbers, you can come to some rather unappealing conclusions. For example, having 10-20 links in an email campaign may at face value seem like it can “boost” your click rate, but potentially at the expense of having an unfocused message. For example, if you have a newsletter with say, 6 different articles and a dozen links to choose from, you may be increasing the likelihood that there’s something that any given recipient may find relevant. But if you want everyone on your list to buy a specific product or sign up to a webinar, then having an overly-complicated email won’t get you any closer to your goal. HubSpot emphasised finding a “middle ground” in their blog:

It’s never a bad idea to include multiple links in an email, since each link is a call-to-action that could reconvert your email recipient. That said, you don’t want those calls-to-action to compete with one another, which is why it’s crucial that you decide exactly what it is you want your email recipient to do upon receiving your email. Answers to your top 11 questions about email marketing“, HubSpot

So, our takeaway here is, use links. Use multiple links. But stay focused on your message and whatever you do, don’t provide just one link to use – being your unsubscribe.

Now that you’ve seen the numbers, what do you think? Should senders be deliberate about how many links they use? Let us know your thoughts in the comments below.

  • Harry Brignull

    Perhaps there’s another factor underlying the “link count – CTR” correlation?

    e.g. if the high link count emails are all ecommerce sales emails and if all the low link count emails are non-sales newsletters…

  • Vladimir Bosanac

    That seems kind of intuitive because different people respond to different things, so one should expect certain links to work better for some people and other links for others. One simple link may only work for a subset of the population. Great to see metrics around this. Thank you for providing that.

  • Jordie van Rijn

    While the question of more versus less links is valid, the way it is investigated (all emails, all links, general outcomes) here isn’t. There are so many different factors and reasons that could influence these numbers, any real insights would be at least two levels deeper. Or.. don’t take avarage for an answer: http://www.emailmonday.com/dont-settle-for-average-email-marketing-statistics

  • Chris Batter

    I have over 20 links on most of my campaigns, your research certainly confirms in our strategy that more links produce higher clicks. Works very well for us but we have little or no competition in our sector, so this may not be representative of most? Thank you for the stats ;)

  • Gareth

    Very interesting article Ros. Completely agree about staying focused with the message. I always aim to have a primary aim (link) with a splash of secondary links.
    I think the way you present the links is more important than volume. I’ve found that very similar emails containing large call-to-action button worded slightly different will have different open rates. Changing the standard ‘More Info’ to something tailored to the audience like ‘Get Stuck In!’ often works better.
    So, I’d prefer to see one well designed link compared to many poor engaging links.

  • Tim Watson

    Great piece of research and well worth doing. Would be even better if analysis included conversion too.

    Look at slide 27 from these slides
    (scroll down to get to the slide deck).

    It shows that more choice (links) can increase engagement but may decrease conversion. Perhaps providing more links for less engaged people is the answer.

    And there are examples of course when split testing fewer links shows more clicks & conversion.

    What the data does perhaps help set is some guide on the number of different links to try split testing… as the rest is down to offer, brands, audience, objectives.

    Don’t build a strategy on being average.

  • Shayra Quezada

    Great article -thanks for this! It’s always a battle for finding the right balance isn’t it? We deploy about 4 MM emails monthly across multiple client-brands and markets. Our findings in A/B tests tell us that having multiple links perform best for the reason you stated here: increasing the number of interest-specific pathways to an otherwise diverse audience – a catch-all of sorts. However, we have found that the best way to do this is through the use of a navigational bar with carefully selected pathways vs. diluting our message with too many CTAs. This allows us to send focused messages with very direct call-to-actions a strategy that consistently works best with engaged users hence allowing for higher conversion velocities on those campaigns. On the same token, messages with a diversity of links/messages in the body content works best in converting unengaged users where we may not necessarily be as familiar with their hot-button interests.

  • Jim Clarkson

    It seems most prudent to test your markets, create narrative and vary approaches. We are fickle, we are pushed for time and our reactions, emotions and demands are not a constant. Being average is an occasional thing, our users are often above and sometimes below.
    Our content, service, skill or product will be acted upon due to way more than a lot or a little CTAs
    Research and evidence coupled with useful appropriate design will increase results.

  • Tom

    “…we looked at the behavior of over 500 million email recipients, from email campaigns sent to over 500 subscribers in recent months.”

    Just to clarify, was the sample size 500 — or 500 million?

  • JohnP

    Hi Ros – How about having multiple links/buttons to the same CTA? Do you have any data on how that improves the overall click rate for that primary CTA?

    Seems like a good way to maximize the clicks without diluting the end goal…

  • Seppo Sinkkonen

    JohnP just asked the same question I was thinking.

  • Justus

    Very interesting article and subsequent discussion. What I take from it is encouragement to ‘test’ different strategies with my particular readers always keeping an ‘essentialist mindset’: the disciplined pursuit of less but better.

  • WP Learner

    Yes, the more links, the more chance recipients will find something useful to them. It also means that the content is richer in order to contain more links. Thanks for the study.

  • Matthijs

    Correlation is not causation! It’s so frustrating to see this mistake being made time and again. Such a waste of the time and effort put into this “research” (no offense). It’s nice to find such a correlation but you absolutely don’t know anything about what caused what. There could be many other variables causing the correlation. The real causal relationships could go every way. And, most probably, will differ depending on a whole bunch of other factors. Maybe for some newsletters increasing the amount of links leads to higher click rates. For other newsletters it could reduce click rates. Etc etc. Only when you do an A/B test on your own specific campaign can you draw any conclusions.

  • Nicki

    I agree with Matthijs, especially “Only when you do an A/B test on your own specific campaign can you draw any conclusions.” I send emails for a variety of clients on a variety of topics, and I’ve seen click rates varying from 75% to just a few percent, depending on the topic and the audience.

    That said, though … thanks, Ros, for launching a good discussion!

  • JohnP

    @Matthijs That is why CM took a sample of over 500 million recipients.

    It is impossible to pinpoint a cause for interaction, simply because we are not inside every recipients head. We can’t pull stats on people who clicked a button “because it was pretty” vs. people seeking valuable content. We can’t pull stats on people who link to “poor quality” content (as that is subjective). A large sample like 500 million helps to balance out those less likely scenarios and offers a general trend that is applicable to the majority of email marketers.

    I disagree with your statement that this research is a “waste of time”. Acknowledging trends is good food for thought, not a hard fast rule where you should ignore your own testing, experience or scenario.

  • Matthijs

    @JohnP: the fact that the correlation was found in a large sample doesn’t matter. The problem remains that there’s not any proof of a causal relationship. It’s a real possibility that there are other variables responsible for the correlation. That is just basic statistics/science 101.

    I can show a very strong correlation between the amount of storks and the amount of births in a region. The larger the sample I take, the stronger that correlation might get. But that doesn’t proof that storks are responsible for bringing babies to families.

    For example, it could be possible that email campaigns with fewer links are a different kind of campaign then the campaigns with more links. Say the first ones are the shorter, more informative emails (a company explaining something) with only a few obligatory links. And say the second group are the commercial campaigns with all kinds of discount offers on products. With more links then the first group. Now when you group both groups together, you might find a correlation between amount of links and click rate. But that doesn’t proof a causal relation. Maybe increasing the amount of links within either group itself even reduces the click rate. You just don’t know. The only way to find out is to design your campaign and do an A/B test in which you keep everything the same but change the amount of links between them. Maybe you discover that the more links the higher click rate. Or you discover -reducing- the amount improves click rate, maybe because it makes your emails look less spammy.

    Maybe my wording was a bit harsh (sorry Ros). Doing this kind of research is not a waste of time. The problem I have is with the conclusions being drawn and the scientific mistakes being made.

  • carl.michael

    Nice insight, though I believe that there are just too many variables consider here. In this situation I think each email needs to be looked at subjectively. Comparing 500m emails is great to get a broad baseline figure and show some link, but each email exec will need to do their own research on their client base.

  • Dan

    Great article but it’s not only number of CTAs that counts but type as well i.e. Buy now and More info. Also, this would be valid across “spray and pray” type of campaigns only. When you’re trying to sell something to a person that is not interested in a product you’re advertising in email, that recipient will be more likely to click on More info rather than Buy now. Following this thought, we shouldn’t be comparing number of CTAs only when these are different. In theory, email with 10 “More info” CTAs shouldn’t perform as well as email with 20 “Buy now” buttons, but it will because it’s more user friendly CTA. Conversion rate is a different story and buy now will probably convert better.
    In targeted campaigns type or number of CTAs is not that important if your target audience is right.
    It was very interesting read though

  • JohnP

    @Matthijs Even under a somewhat controlled scenario of changing one thing and A/B testing that way, we are still making many assumptions.

    It is impossible to eliminate all variable factors. Maybe people like more links in the afternoons, or maybe the links in the next campaign perform very differently – we just don’t know for sure. All we can do is acknowledge the (limited) accuracy of the data we obtain.

    This is similar to the industry wide email client % chart Litmus publishes. Sure it may mean nothing to you, but it might just inspire you to run the numbers on your own list.

  • Scot

    Thank you for the article. I will use more links in my next campaign.

    Please note that there is a difference between the words “less” and “fewer”

  • Ros Hodgekiss

    A huge thanks to everyone who’s chimed in here, I’ve been really enjoying the discussion so far. Double thanks to everyone who have turned our blunt conclusions into opportunities to ask more questions about testing, data sets and more. Distinguishing and isolating all the factors that go into any given click rate is quite the task – and so I’d love to take things to the next level and spend some time looking into the impacts of multiple CTAs and more. Hopefully this is something that we’ll be able to pull data on in the near future – you’ll see the results in this blog.

    And yes, you’re right Scot – “less” and “fewer” are indeed different. We had an epic discussion this that internally (hat tip to our English geeks in Support). I changed the headline to “fewer”, but looks like “less” still managed to sneak its way in – my apologies :)

  • Shaun Dutton

    We recently saw this with a client that too their email strategy in house (thinking it was easy) we saw the click % half over 6 months due to them believing that just having one button means more clicks… The campaign was for Nambiti Hills, feel free to use it in a case study. It’s under our FirstView account.

  • Dr Frances Follin

    Emails for my online art magazine, Cassone, include links to every article in the new issue. It is up to recipients what catches their interest and what they want to click on – the most popular things are often not what I would have expected. As long as recipients go to some part of the magazine that is fine by me – I have no interest in directing them towards one thing in particular. Once they are ‘in’ they can navigate round themselves, or go back to the email and click on something else (which a number of them do)

  • Big Jason

    As a direct response email marketer, I find this article (in its current form and assumptions) complete garbage. Different story if this article was for email markets who don’t care about sales…

  • Mathew Patterson

    Thanks for all the great comments. Jason, did you have any specific issue that you’d like to discuss? We always appreciate hearing other viewpoints elaborated.

This blog provides general information and discussion about email marketing and related subjects. The content provided in this blog ("Content”), should not be construed as and is not intended to constitute financial, legal or tax advice. You should seek the advice of professionals prior to acting upon any information contained in the Content. All Content is provided strictly “as is” and we make no warranty or representation of any kind regarding the Content.
Straight to your inbox

Get the best email and digital marketing content delivered.

Join 250,000 in-the-know marketers and get the latest marketing tips, tactics, and news right in your inbox.


See why 200,000 companies worldwide love Campaign Monitor.

From Australia to Zimbabwe, and everywhere in between, companies count on Campaign Monitor for email campaigns that boost the bottom line.

Get started for free