[Estimated read time: 10 minutes]
The traditional ways of measuring the success or failure of content are broken. We can’t just rely on metrics like the number of pageviews/visits or bounce rate to determine whether what we’re creating has performed well.
“The primary thing we look for with news is impact, not traffic,” says Jonah Peretti, Founder of BuzzFeed. One of the ways that BuzzFeed have mastered this is with the development of their proprietary analytics platform, POUND.
POUND enables BuzzFeed to predict the potential reach of a story based on its content, understand how effective specific promotions are based on the downstream sharing and traffic, and power A/B tests — and that’s just a few examples.
Just because you’ve managed to get more eyeballs onto your content doesn’t mean it’s actually achieved anything. If that were the case then I’d just take a few hundred dollars and buy some paid StumbleUpon traffic every time.
Yeah, I’d generate traffic, but it’s highly unlikely to result in me achieving some of my actual business goals. Not only that, but I’d have no real indication of whether my content was satisfying the needs of my visitors.
The scary thing is that the majority of content marketing campaigns are measured this way. I hear statements like “it’s too difficult to measure the performance of individual pieces of content” far too often. The reality is that it’s pretty easy to measure content marketing campaigns on a micro level — a lot of the time people don’t want to do it.
Engagement over entrances
Within any commercial content marketing campaign that you’re running, measurement should be business goal-centric. By that I mean that you should be determining the overall success of your campaign based on the achievement of core business goals.
If your primary business goal is to generate 300 leads each month from the content that you’re publishing, you’ll need to have a reporting mechanism in place to track this information.
On a more micro-level, you’ll want to be tracking and using engagement metrics to enable you to influence the achievement of your business goals. In my opinion, all content campaigns should have robust, engagement-driven reporting behind them.
Total Time Reading (TTR)
One metric that Medium uses, which I think adds a lot more value than pageviews, is "Total Time Reading (TTR)." This is a cumulative metric that quantifies the total number of minutes spent reading a piece of content. For example, if I had 10 visitors to one of my blog articles and they each stayed reading the article for 1 minute each, the total reading time would be 10 minutes.
“We measure every user interaction with every post. Most of this is done by periodically recording scroll positions. We pipe this data into our data warehouse, where offline processing aggregates the time spent reading (or our best guess of it): we infer when a reader started reading, when they paused, and when they stopped altogether. The methodology allows us to correct for periods of inactivity (such as having a post open in a different tab, walking the dog, or checking your phone).” (source)
The reason why this is more powerful than just pageviews is because it takes into account how engaged your readers are to give a more accurate representation of its visibility. You could have an article with 1,000 pageviews that has a greater TTR than one with 10,000 pageviews.
Scroll depth & time on page
A related and simpler metric to acquire is the average time on page (available within Google Analytics). The average time spent on your webpage will give a general indication of how long your visitors are staying on the page. Combining this with ‘scroll depth’ (i.e. how far down the page has a visitor scrolled) will help paint a better picture of how ‘engaged’ your visitors are. You’ll be able to get the answer to the following:
“How much of this article are my visitors actually reading?”
“Is the length of my content putting visitors off?”
“Are my readers remaining on the page for a long time?”
Having the answers to these questions is really important when it comes to determining which types of content are resonating more with your visitors.
Social Lift
BuzzFeed’s “Social Lift” metric is a particularly good way of understanding the ‘virality’ of your content (you can see this when you publish a post to BuzzFeed). BuzzFeed calculates “Social Lift” as follows:
((Social Views)/(Seed Views)+1)
Social Views: Traffic that’s come from outside BuzzFeed; for example, referral traffic, email, social media, etc.
Seed Views: Owned traffic that’s come from within the BuzzFeed platform; e.g. from appearing in BuzzFeed’s newsfeed.
This is a great metric to use when you’re a platform publisher as it helps separate out traffic that’s coming from outside of the properties that you own, thus determining its "viral potential."
There are ways to use this kind of approach within your own content marketing campaigns (without being a huge publisher platform) to help get a better idea of its "viral potential."
One simple calculation can just involve the following:
((social shares)/(pageviews)+1)
This simple stat can be used to determine which content is likely to perform better on social media, and as a result it will enable you to prioritize certain content over others for paid social promotion. The higher the score, the higher its "viral potential." This is exactly what BuzzFeed does to understand which pieces of content they should put more weight behind from a very early stage.
You can even take this to the next level by replacing pageviews with TTR to get a more representative view of engagement to sharing behavior.
The bottom line
Alongside predicting "viral potential" and "TTR," you’ll want to know how your content is performing against your bottom line. For most businesses, that’s the main reason why they’re creating content.
This isn’t always easy and a lot of people get this wrong by looking for a silver bullet that doesn’t exist. Every sales process is different, but let’s look at the typical process that we have at HubSpot for our free CRM product:
- Visitor comes through to our blog content from organic search.
- Visitor clicks on a CTA within the blog post.
- Visitor downloads a gated offer in exchange for their email address and other data.
- Prospect goes into a nurturing workflow.
- Prospect goes through to a BOFU landing page and signs up to the CRM.
- Registered user activates and invites in members of their team.
This is a simple process, but it can still be tricky sometimes to get a dollar value on each piece of content we produce. To do this, you’ve got to understand what the value of a visitor is, and this is done by working backwards through the process.
The first question to answer is, “what’s the lifetime value (LTV) of an activated user?” In other words, “how much will this customer spend in their lifetime with us?”
For e-commerce businesses, you should be able to get this information by analyzing historical sales data to understand the average order value that someone makes and multiply that by the average number of orders an individual will make with you in their lifetime.
For the purposes of this example, let’s say each of our activated CRM users has an LTV of $100. It’s now time to work backwards from that figure (all the below figures are theoretical)…
Question 1: “What’s the conversion rate of new CRM activations from our email workflow(s)?”
Answer 1: “5%”
Question 2: “How many people download our gated offers after coming through to the blog content?”
Answer 2: “3%”
Knowing this would help me to start putting a monetary value against each visitor to the blog content, as well as each lead (someone that downloads a gated offer).
Let’s say we generate 500,000 visitors to our blog content each month. Using the average conversion rates from above, we’d convert 15,000 of those into email leads. From there we’d nurture 750 of them into activated CRM users. Multiply that by the LTV of a CRM user ($100) and we’ve got $75,000 (again, these figures are all just made up).
Using this final figure of $75,000, we could work backwards to understand the value of a single visitor to our blog content:
((75,000)/(500,000))
Single Visitor Value: $0.15
We can do the same for email leads using the following calculation:
(($75,000)/(15,000))
Individual Lead Value: $5.00
Knowing these figures will help you be able to determine the bottom-line value of each of your pieces of content, as well as calculating a rough return on investment (ROI) figure.
Let’s say one of the blog posts we’re creating to encourage CRM signups generated 500 new email leads; we’d see a $2,500 return. We could then go and evaluate the cost of producing that blog post (let’s say it takes 6 hours at $100 per hour – $600) to calculate a ROI figure of 316%.
ROI in its simplest form is calculated as:
(((($return)-($investment))/($investment))*100)
You don’t necessarily need to follow these figures religiously when it comes to content performance on a broader level, especially when you consider that some content just doesn’t have the primary goal of lead generation. That said, for the content that does have this goal, it makes sense to pay attention to this.
The link between engagement and ROI
So far I’ve talked about two very different forms of measurement:
- Engagement
- Return on investment
What you’ll want to avoid is actually thinking about these as isolated variables. Return on investment metrics (for example, lead conversion rate) are heavily influenced by engagement metrics, such as TTR.
The key is to understand exactly which engagement metrics have the greatest impact on your ROI. This way you can use engagement metrics to form the basis of your optimization tests in order to make the biggest impact on your bottom line.
Let’s take the following scenario that I faced within my own blog as an example…
The average length of the content across my website is around 5,000 words. Some of my content way surpasses 10,000 words in length, taking an estimated hour to read (my recent SEO tips guide is a perfect example of this). As a result, the bounce rate on my content is quite high, especially from mobile visitors.
Keeping people engaged within a 10,000-word article when they haven’t got a lot of time on their hands is a challenge. Needless to say, it makes it even more difficult to ensure my CTAs (aimed at newsletter subscriptions) stand out.
From some testing, I found that adding my CTAs closer to the top of my content was helping to improve conversion rates. The main issue I needed to tackle was how to keep people on the page for longer, even when they’re in a hurry.
To do this, I worked on the following solution: give visitors a concise summary of the blog post that takes under 30 seconds to read. Once they’ve read this, show them a CTA that will give them something to read in more detail in their own time.
All this involved was the addition of a "Summary" button at the top of my blog post that, when clicked, hides the content and displays a short summary with a custom CTA.
This has not only helped to reduce the number of people bouncing from my long-form content, but it also increased the number of subscribers generated from my content whilst improving user experience at the same time (which is pretty rare).
I’ve thought that more of you might find this quite a useful feature on your own websites, so I packaged it up as a free WordPress plugin that you can download here.
Final thoughts
The above example is just one example of a way to impact the ROI of your content by improving engagement. My advice is to get a robust measurement process in place so that you’re able to first of all identify opportunities, and then go through with experiments to take advantage of the opportunity.
More than anything, I'd recommend that you take a step back and re-evaluate the way that you're measuring your content campaigns to see if what you're doing really aligns with the fundamental goals of your business. You can invest in endless tools that help you measure things better, but if core metrics that you're looking for are wrong, then this is all for nothing.
Hey Matthew, it's great to see the numbers behing the content. Most of the time, we just create the content, make some SEO and after some time forget about them. But, the key is interpreting on statistics and changing your strategy with respect to these numbers properly.
Thanks for opening our minds.
Thanks for great post Matthew .
Never knew that websites like medium and buzzfeed check their audience behaviour as I see majority of bloggers (who want to earn fast and easy) writing for sake of writing (based in keyword) thinking that somehow with good backlinks they might get them on top (which usually is true) but with time and user experience they eventually fall time with time.
Adding to your points:
Quality matters more than quantity.
ROI comes with good content.
Thanks and regards
Pulkit Thakur
A really good post. When you mentioned time on page as a metric I started feeling skeptical - bouncing visits showing times of 0:00:00 skewing your mean value and TTR calculation and all that - but that Scroll Depth plugin took away my skepticism and introduced me to a great new tool.
That Summary button on your blog is pretty interesting and an unusual approach. Did you test that against the more standard "use your introduction to sum up the post" approach or just go for the button? If you did test, how big was the difference in engagement?
I liked it, althought some of the people don't. Lol. Thanks Matthew!
Totally agree that ROI should always be measured, but also common sense. I believe that as a rule if the cost of analysing eats up the revenue ROI of the action analysed then it is not worth the measurement... Having a proper process in place to analyse can save time on it, plus be able to focus on producing quality engaging content and avoid the "from analysis to paralysis" problem I believe.
I think many people do not measure the results, it is very important to know the data say to act correctly
I really loved your post.
Measuring content is important but knowing that they are doing things the same point incorrectmante .
It's funny, I consider the summary of the text would entail a faster website abandonment and hurt us in the bounce rate . I'll try it in my next article to see what effects have respect for others.
I completely see your point here, but here's the thought process that I started with when creating the tool:
Users that visit my content but don't have time to read long-form content are going to leave. How can I match their preference of getting some quick information and expose them to my content without having to completely rewrite my whole post?
From this came TLDR. The testing that I've done so far has definitely seen a big inprovement on the engagement within my longer content.
Good stuff Mathew. Makes me think about getting smarter and more efficient at this kinda stuff!
I must say I hear all too often about people creating blog posts almost just for the sake of it, in the hope they might pick up some traffic. A well researched and highly topical/relevant/informative article that you can then measure the impact of is worth 20+ random 'just because' articles. Proof is in the pudding as they say; we've got a blog in the Health niche that has a growing amount of well researched and quite lengthy articles, and even the old ones still pull i decent traffic that sticks around on the post for 3-4 minutes plus on average. We then use basic metrics in GA to measure what users that first heard about the blog via a certain post went on to do. If it's nothing - then the purpose is questionable... if they check out more content, sign up, come back again etc. then it's worked.
I've got an issue with all this analysis. I mean yeah it is useful but I say just get on with it. Use common sense to realise what people in your field want to read about, and spend your time thinking about how best to serve this need in an exciting way. Better to put your energy into this than worrying too much about how your last article did. Robust measurement process? I think that's just wasted energy. Spend that time producing 2 or 3 more blog articles a week and paying attention to the comments.
Respectfully, I strongly disagree here. If you're not measuring things correctly then how will you know that "producing 2 or 3 more blog articles a week" is even worthwhile? How will you prioritise resources against specific task, and how will you ever know if what your doing is delivering ROI? This is particularly important if you're doing this for clients where they're paying money in and want to know how much is coming out that can be directly attributed to what your doing. That's the way I look at it - I know it can sound tedious, but a lot of this can be automated.
Mcparland_Copy, I totally disagree with your comment for a number of reasons.
Firstly, you've essentially said is that it is better to make 'more' content than to take the time to measure what makes better 'quality' content. The over-saturation of content on the web is a huge problem and it is partly due to statements like 'just get on with it' and measurement being 'wasted energy' that has caused this. In my humble opinion, your attitude personifies a cancerous problem in our industry - lazy marketing. Just because measurement is time-consuming does not make it optional.
Secondly, I have never met a client who has not demanded an ROI on the content they produce. While I personally feel ROI is a terrible metric to measure content success, it is something all content creators need to be able to measure in order to satisfy a client's demand to their superiors. And Matt's article is a good way to do this. If you are not investing in your content's ability to affect the bottom line of the business, explain to me how are you justify the strategy of creating more content in terms of business cost and time?
Finally, kudos to Matt. This is an fantastic article on how to measure ROI and engagement. At our agency, we also measure Visibility (how much reach the content gets via fresh links or organic shares and impressions on social) and a more ambiguous report on how the audience affects future business decision. This last one is about change management. We listen to where the audience says our business is most valuable. It's a 'hands off the handle bars' approach that lets the audience, not the marketing team, steer. It provides some amazing insight into opportunities around creating new revenue streams and reshaping the silos within a business.
I'm going to have to side with Matt and Daniel on this, I'm afraid. Data generally trumps your gut: even if it doesn't then being able to say to your client or boss "this is what we've done before, these were the exact results, and this is why we think it will work again" is a much more convincing argument than "I reckon this will be good".
A good example of this is in the podcast I listened to while reading this post, with the freelance copywriter host talking to a social media consultant. The consultant described her old job at an ad agency, where her research into how social media could benefit clients - including data, reports, and statistics - was rejected by her boss because he had a hunch radio was going to come back. She got fired for focusing on Facebook instead of traditional methods. The host's conclusion based on his writing experience? If she'd been talking about a client then he would have called her being fired "dodging a bullet" because failing to listen to data is one of the cardinal sins of business.
Also, producing "2 or 3 more blog articles a week" might work for some people but won't for others. Matt's blog is a great example: tens of thousands of people, agencies and companies produce several useful posts a week about online marketing but Matt's stand out because they're so in-depth. If he was to be one of those thousands producing average length posts several times a week he probably wouldn't have such a successful blog because despite his writing talents he would just blend into the crowd.
I don't know but I'd hazard a guess that Matt started using this approach after he wrote a really long post, looked at the engagement data, and had an epiphany.
I stopped reading at "You're doing it wrong" - Great, someone that knows how I am doing it and assumes they are doing it better. The SEO & Analytics world is full of two types of people. Those that listen to everyone else and improve the way they do it, and those that assume they are doing it better than everyone else. This annoys me when they say this in conferences & in blog posts.
Hey Gerry, this isn't about me coming out and saying, "Hey, I'm doing things better than you." Instead, this is about me saying, "I've tested some things out and here's what I've found." This model won't work for everyone but I've done my best to show a specific use case where this can be really valuable.
The title was "you're doing it wrong" doesn't that explicitly says that your way is better? Maybe I think that moz should do better than clickbait titles & people whether or not this is a good article, we maybe should switch to better more user friendly, SEO friendly title tags?
What if their data shows that clickbait titles drive traffic, engagement, and subscriptions? ;-)
Fair point :-)
Should Moz be doing it? Nathalie Nahai did an awesome presentation about this at BrightonSEO on clickbait & how to do it (right), and in fairness since listening to her, I have been improving the titles of many blog posts at work to try and improve CTR... It isn't just this article, it is the rise of this sort of naming that I think is not a positive move, and most "10 GA features you are not using" articles are 10 things that most people ARE using.
"Great, someone that knows how I am doing it and assumes they are doing it better"
Although that's not the case, it's a fair assumption.
That someone is Matthew Barby, who happens to be one of the top marketing experts in the world.
Don't take my word for it, check out his blog, here is my challenge: try to find one piece of content that's not spectacular (you'll come up short).
I urge you to read his blog, that's not an easy assignment, but here is my promise - you'll get better.
Yes, it's that simple.
In fact afterwards, when you'll negotiate a new salary you'll feel obligated to send him a check.
Why am I telling you this with such confidence? Because that's my story.
Employers & colleagues always keep asking me: "WTF... how do you know so much about ..."
Matthew deserves recognition for his work, he helped me become a better marketer and I'm grateful for it.
Hey Gerry, I think you misunderstands the mean of Mathew's wordings. It was just his experience sharing with a compelling TITLE. Isnt it?
Hi Matthew
Indeed, the primary think we look for With news is impact, not traffic. We are used to give importance to some details that are not as important as the ultimate goal, which is none other than sales or impacts.
Congratulations for the post !!
I also think that people use to focus too much on the final goal and they get so lost in the process that forget how to get to it. Details are the ones that make the difference :)
Thanks, Carlos, glad you found it useful.
Hey Matthew, great insights there! Interesting how Buzzfeed and Medium measure their audience engagement. For most businesses in the content marketing terrain, conversion is what counts the most. You're right - when the audience is unlikely to convert, engagement metrics are but vanity. Thanks for sharing your research!
Hi Matthew!!
It is obvious that for us the most important is that you are able to visit us achieve goals (usually, sales ...), but absolutely all metrics we control them precisely to achieve these goals.
That was a fantastic post Matthew--I have to say that I got a lot out of that and am going to rededicate myself to going deeper in analyzing our content and the value it brings to the company.
Quick question for you--can you provide (if you are willing and can) what kind of percentage change you experienced when adding in the "Summary" option?
One of the challenges we've come across is being able to compare article performance side by side from day 1 of publication. Yes, we can measure scrolling, time on article, page views, social views and interactions. But it takes a really long time to pull together this info in a single report, for each article, over a consistent time frame since publication.
With out measuring the right metrics, and improving based on data, you are never gonna improve by much.
We should avoid the spun & SEO tools to create content for website promotion.
Such a great post Matthew .