Lessons from a decade of monetising content with affiliate marketing
We've been around for over a decade offering users price comparison for telecoms and technology products. As a business, our commercial model is entirely based around affiliate marketing and a large part of this is producing content that delivers value to our users and generates revenues for the business. Over the past 14 years we’ve learned a lot when it comes to driving affiliate revenues from content. Here we share some of the lessons we’ve learned along the way.

Reporting is everything
The affiliate (and digital) worlds have the beauty of being extremely data rich. Despite this, it’s amazing how many sites (some of these being major, major news brands) are not able to measure commercial performance of their content. When it comes to improving revenues, it’s absolutely essential to know what content is driving what sales.
This is so important that we spent over a year developing technology (which we now make available to others at Fullsight.tech) that would allow us to track affiliate revenue back to page. Crucially, Fullsight pushes this data into GA so both our commercial and content teams can instantly see what articles are driving what sales and revenue.
Value added is essential
This may seem like an obvious point, but it’s amazing how many sites don’t understand this. Back when we started, we used to write-up every deal that every advertiser launched with 300 words of filler and a basic quote. It took us a while but eventually we started experimenting with content that addresses question and adds value for someone researching a product/looking to buy. Users are people, and if they’re searching for an answer, it’s because they have a question.
Users understand the value exchange
We are very transparent about our commercial model with our users and make it clear on our site that we have commercial agreements with some of the brands we mention/link to. On the whole, users understand and are happy with the fact that we are a business and have to make money. What’s more, and this may be controversial, I think users are a lot happier about a site that is making money through affiliate links than they are with a site overloaded with ads and/or a site that sells every aspect of their data
Be wary of sponsored placements
We’re living in the age of influencer marketing. One of the things that comes up again and again from influencers is the importance to them of only promoting brands that are a good fit, and in a way that is right for their followers. The same should always ring true of content sites – a brazen promotion will jar with your users and will not get the same levels of engagement. Not all sponsored placements are bad, but these need to be done in the right way.
Get your metrics in order
This isn’t just a lesson learned applicable to understanding affiliate performance of content but certainly does apply here. Over the past decade we’ve changed our golden metrics several times, from traffic to bounce rate, to scroll depth to click through rate to registration.
After a lot of trial and error, we’ve found a secret-sauce combination of three key metrics that work for us. One of these is page value – by using Fullsight to pull real affiliate revenue into GA we can see the real page value for each article. The beauty of this metric is that GA attributes revenue to all pages in the user journey, so we can see the value of pages that weren’t just first click or last click. This has helped us to really understand the importance of pages that increase conversation rate in the middle of user journeys – for example review pages – which might not be the entry or exit point for the user but can really help with increased conversion.
These might not be the same for every business but the key point with this is to think through and understand what the ultimate goal is for your content, and then exclusively measure performance, success or failure against these metrics.
Dominic Baliszewski is head of new ventures at Decision Tech
Content by The Drum Network member:
