Why Using Traffic As A KPI Is Hurting Influencer Marketing Campaigns
By Matt Gibson
Traffic is a terrible indicator of potential readership for a specific article. Its use for that purpose probably wastes millions of PR and marketing dollars each year.
A website with a million posts might get a million page views per month. That could be made up of one view per post, or, it could have a more common traffic distribution and receive a million views to one post while all the others receive zero. Or it could be anywhere in between. Regardless, any story on that website will receive an average of one pageview per month. A blog that has 5 posts and gets 100 page views per month receives an average of 20 pageviews per post per month.
So, all other things being equal, a story placed on the blog with 100 pageviews per month should be expected to reach 20X as many readers as one placed on the website that gets a million pageviews per month. It’s not hard to see how a blog that 100 pageviews per month could realistically provide exponentially better value than the blog that gets a million. My example exaggerates the characteristics of the average website traffic distribution to make a point, but it’s not as exaggerated as it might first sound. Traffic is highly correlated with volume of content and is usually the result a small number of internal pages that rank well. Websites with more content get more traffic, that’s almost universally true. The vast majority of website traffic consists of readers who land on — and leave from — internal pages, rarely taking the time to browse around. I’d be willing to bet 95% of us here receive more than half our monthly traffic from our top ten posts. A lot of us probably get it from the top five.
My home page gets less than 6% of my website’s monthly pageviews. Only 6% of all total website traffic will even have the opportunity to stumble onto a new article I’ve written, and even then the article is competing with a variety of other content for their attention. Let’s say I have 100,000 monthly visitors. Less than 10% of them are landing on the homepage. That brings the number of potential readers for a new article down to 10,000. On the average blog homepage, readers are usually choosing between 10 and 20 content options. This brings us down to a realistic estimate that we may reach around 500-1000 readers through organic traffic; around 0.5 to 1% of total traffic.
Total website traffic has minimal effect on the amount of traffic that will visit any individual new post, yet that number is widely considered to be the best (if not only) predictor of potential traffic to a new individual post and every day people are making decisions in the 4-, 5-, and 6-digit range based largely on that (basically useless) metric.
Traffic is a good number to know in the context of understanding an influencer’s overall authority in a topic. But it’s never treated that way. Traffic numbers are wildly inaccurate predictors of outcomes and they’re so universally trusted and grossly overemphasized by the people who vet influencers and media that their use is probably responsible for more wasted PR and marketing money than any other single factor in the industry today.
That would be understandable if we didn’t have any better indicators to rely on, but we do have them. The problem is that they get ignored. These are few things that have a real and (and in some cases adjustable) effect on traffic to a new post:
This is the best indicator of a blog’s following and are the people most likely to actually want to read a new post on the blog out of interest.
Size and engagement of the blogger’s social following
This is a fuzzier indicator because engagement varies so widely, but also a good indicator of people who may be interested in reading a new blog post out of interest.
The quality of the post in question
This can make a huge difference in traffic. It’s a less specific indicator because there’s no objective measurement. But it’s adjustable because it’s related to the amount of time a blogger can spend working on it, which is in turn affected by the rate of pay (if there is any).
Time (and money) spent promoting the post
This is probably the easiest and cheapest way to reliably increase the potential readership of a blog post. Dropping $50 on Facebook boosted Facebook posts and targeted ads makes a huge difference. If a marketer wants to maximize the performance of a potential blog post they should look at these KPIs:
- Newsletter subscribers instead of traffic
- Average engagements per social post instead of followers
- Budget allocation that prioritizes quality instead of quantity
- Budget allocation for content promotion
A couple of years ago I started following the above practices, tracking results, and compiling case studies so I could give potential clients a clear picture of what they could reasonably expect from working with me. On the first project I set aside a promotions budget of about $50 per post. The first of two posts received nearly 20,000 readers who stayed on the page for an average of over 5 minutes. That post was super successful on Facebook, and I wouldn’t claim I could replicate that level of success. The second post received 2,500 readers for an average of 5 minutes each, which I’m confident I could easily repeat (and probably beat). This was at a time that my blog was getting around 20,000 visitors per month. (Just a reminder here that organic traffic on a website that gets 100,000 visits per month could be expected to send 500 – 1000 visitors to a new post.)
At that time (and still today, actually) a person with 5X more traffic than me could charge at least 3X as much money for the same work and results that would be — in many cases — comparable. Because of that, when I do paid campaigns I always provide a report that includes a breakdown of the key metrics (social reach, social interactions, pageviews, average time on page, etc.) as well as value-related KPIs, such the cost per second of consumer attention (price paid for blog content/total content viewing time in seconds) and cost per social interaction (price paid for social promo/total social interactions). That way they there is a clear understanding of cost of each outcome and the value of my work can be clearly compared to the value provided by others, dollar-for-dollar, regardless of the size of their following.
I highly recommend that others do that same. Showing exactly what you are able to deliver is an amazing sales tool, because it removes uncertainty from the equation. Breaking down the value of results to their most basic useful units (such as cost per social engagement) is an excellent way to show how your work compares to others’. And, by using these kinds of measurements when you discuss rates, value, and effectiveness, you’re giving your potential client the tools to effectively evaluate influencers and campaigns (which helps the whole industry) while at the same showing that you’re genuinely interested in — and capable of — providing value backed up by solid numbers.
This post originally appeared on the UpThink blog here.