Matt Smith was managing director of eCommerce at Future Plc, leading their pace-setting global affiliate revenue strategy for six years. He’s led publishing teams that have driven over £350mn in revenue, and claims to have helped retailers to achieve more than £3bn in online sales. In 2022, he founded New Model Media, a UK-based digital consultancy that advises media companies on strategies to build audience and revenue growth. In this occasional series of Teach-In with industry specialists, Smith answers our questions about the prospects and pitfalls of the revenue stream that all publishers are chasing…
How important a revenue stream can eCommerce be for a B2C media company? Why should established media bother with it? Does eCommerce have a role in B2B?
eCommerce revenue isn’t a fit for every media brand. But, for the vast majority of B2C media businesses, it should play an important role in a diversified digital revenue strategy.
How important it is within that mix really differs on a brand-by-brand basis, and depends on things like the content vertical that the brand is in, its ability to offer trusted and authoritative product recommendations and the investment cost / benefit vs other revenue streams. For some brands, that might mean there’s an opportunity for commerce revenue to be as important if not more important than advertising. But, for most brands, commerce is likely to be complementary and 10-25% of total revenue at best.
I also think it’s important not just to assess eCommerce revenue through its bottom line gross contribution because it can bring many other benefits. Commerce revenue can provide a much higher margin and net contribution than other revenue streams and requires smaller teams to manage. If done well, it can provide a predictable and recurring revenue model that allows for accurate forecasting and investment, something that lots of publishers struggle with!
The biggest benefit to commerce as a revenue stream – and one of the reasons I’m so passionate about it – is that (if it is done well) it is a mutually-beneficial way to make money that serves both the readers and the brand, and there is a true value exchange at play. Ultimately, media brands won’t make money from eCommerce if readers don’t have a good experience, get the advice they were looking for and purchase a product. In a digital media world still dominated by advertising and the constant push/pull between ad yields and UX, publishers should be excited about revenue streams that grow as the content quality and user experience improves.
That’s not to say it’s by any means easy though, and there is a lot more that goes into driving commerce revenue than simply conducting some product reviews and creating buying guides. Publishers need to invest in high-quality editorial and best practices for content templates, design as well as building data teams and reporting that can drive on-page optimsiations and real time feedback on reader engagement.
There are many more examples of “bad practice” right now than “best practice”. Publishers with poor quality and rushed content that clearly hasn’t been thoroughly reviewed, with product recommendations that don’t fit the user intent, are out of date, and often out of stock.
The other big mistake I see media brands make is over-extending themselves beyond their site’s core offering, to the detriment of the whole brand. I was recently looking at a tech site that had a “best cat beds” article. Why? They wouldn’t normally cover pets, or pet product releases, so why have this article? Short-term it might provide a small boost in revenue, but long-term I’ve seen multiple sites punished for thin irrelevant content that isn’t their expertise and core, and it’s likely that – even if Google rankings remain ok – this content won’t drive any meaningful conversion and revenue.
At its best, eCommerce content strategy should be complementary to the rest of the brand, with high-quality, well-researched content, and concise buying advice that caters to a wide range of use cases.The best sites are ones that have passionate editorial teams and who love the product categories they are testing: that passion is reflected in the quality of the content and the revenue follows. One such site I’ve loved seeing grow over the last few years is Golf Monthly, which was acquired by Future as part of TI Media. The team there really quickly embraced buying advice content, had an obvious core base of products to review, and have gone from strength to strength over the last few years with truely expert advice. Early this year, they released data that they’d driven around $15mn worth of sales, which would likely be around $1mn in affiliate commissions – a brilliant outcome for a site of their size, and perfectly sustainable.
I think the UK is behind the US in terms of best practice on the whole, but other sites that fit this mould well are sites such as Good Housekeeping (in the US and UK), Sleep Foundation (US), The Spruce (US) and Who What Wear (US/UK).
In terms of B2B, the answer is “it depends”!
B2B content targeted at an enterprise level doesn’t fit the standard eCommerce mould where you are looking to drive affiliate sales for a product or service on a mostly last-click attribution model. The B2B sales funnel isn’t that linear, and the most similar to eCommerce model that works for enterprise B2B content is lead generation. However, there is still a good opportunity in B2B eCommerce content where smaller businesses operate in similar ways to consumers: they generally search out the products and services their businesses need and buy these online. These can be products such as accountancy and HR software, which are all lucrative affiliate channels, through to web hosting and consumer tech such as laptops and monitors.
How far does eCommerce work better with non-content media (i.e. listings and recommendation content rather than comprehensive media brands)? Does it conflict with the editorial role of established media brands?
My view is that there is very little space left for generic sites solely built on “recommendation content” only, and that most brands which try to be a ‘one size fits all’ review site across 100s of topic areas will, inevitably, fail. It’s hard to convince both readers and Google that you can be an expert in everything from headphones to yoga mats.
There are some notable exceptions to this, such as the New York Times’ brilliant Wirecutter, which has built a reputation for high-quality reviews and recommendations, and has slowly expanded its category offering over time, applying its trademarked rigour to every new category it enters. But most media simply doesn’t have the brand equity to be this broad.
eCommerce content should be one element of a brand’s broader content strategy and can, therefore, fit harmoniously into it. This is especially true in specialist media, where there is often a natural and logical extension of a brand’s existing content strategy into eCommerce.
I’d argue that if a specialist brand isn’t doing product reviews in its sector, then it’s doing a bad job for its audience and missing a significant opportunity to add value to both their audiences and their business. In these instances, commerce should be one in the same as the editorial role of the sites, it serves a need and adds value for readers. I’ve seen publishers approach this differently. But I believe the core editorial teams on these sites should be the ones writing commerce content.
That doesn’t mean that, within specialist media, there aren’t conflicts that can arise between established editorial strategy and commerce content. Mostly this is around what products to cover and for whom buying guides are being written. The vast majority of “best X” buying guides receive 90%+ of their traffic from search. The challenge is that these audiences aren’t always the “typical” readership of that brand. More often than not, these readers don’t want the literal top of the range “best” product, they want the best product for their own needs and/or budget, both of which can be hugely varied and often entirely different to a loyal brand reader.
Often for the larger intent terms within a category, the audience is newer or is buying for someone else, and so their starting knowledge base is very different to a regular reader of that specialist website. This can create tension on the editorial teams about how to cover these articles, and what range of products to include that helps the most people whilst also maintaining the brand. Often specialist editorial teams are so deep into their areas of specialism that misunderstand the core reader needs and what level of buying advice they should be sharing. The key is to make sure your recommendations cater for a wide variety of readers, both your core audience and new readers into the category.
eCommerce content can be harder for larger news sites, where there isn’t an obvious associated product category that you can lean into, but eCommerce content can still work with the right strategy. Trading events such as Prime Day and Black Friday are obvious areas for news brands to have success with “newsworthy” content on deals, and there are also examples of news brands, such as The UK’s Independent with the Indy Best, building good eCommerce revenue streams. The challenge is not to treat these articles like the rest of news sites, but to invest in thorough and real reviews and testing, going deeper into specific product areas, rather than trying to cover everything at a high level. Commerce content will be killed by the high volume ad model that lots of news sites adopt, so it needs to be siloed away from this and given time to bed-in and establish.
How much ‘risk’ is involved in eCommerce that is wholly or partly dependent on Amazon retailing and Google search? What are the implications of such relationships?
To some extent, eCommerce helps mitigate areas of risk within media, it diversifies revenue, decreases reliance on ads, and provides revenue that should be beneficial to site user experience. However, much like most other areas of media, there is still unfortunately a large amount of risk within eCommerce itself and you’re right to point out that is mostly due to the reliance on both Amazon and Google. It’s probably easiest to deal with them both separately because they present differing risks and prospects.
With Google, there’s no escaping the fact that search currently provides the majority of eCommerce revenue. To be successful commerce content relies on a steady traffic of high intent audiences, and no other platform currently provides this like search does.
Even if you can generate traffic from elsewhere, it won’t perform as well. I regularly see search traffic drive anywhere between 20-100x the yield of other traffic sources e.g. email, social, direct etc on commerce content. This is especially true in sectors like Tech, Gaming, and Home where you make considered purchase decisions of medium-high value products, and each purchase might be 2-3 years apart. It’s unlikely that sharing a buying guide on social or in an email newsletter is going to reach someone who just happens to be looking to buy that product. But, with search, you know they are in-market.
Short and medium term the reliance on Google for commerce revenue is hard to escape for most but we’ve all seen publishers severely impacted by algorithm changes, with many small companies struggling to compete with the media goliaths (Glen Allsopp’s Detailed.com has some fantastic analysis here). The recent vaulting of Reddit and other forums to the top of search rankings for many product related terms has been a stark reminder for publishers of the precarious nature of some of these high-value terms.
I know some publishers whicj have experimented with paid social ads and an arbitrage model, but often RoI has been hard to maintain and the investment required for meaningful returns makes it a risky prospect. Others are looking to build more direct and owned audiences, but this takes time and, crucially, an established and recognisable brand – again The Wirecutter and sites like Hearst’s Good Housekeeping do as good a job as anyone.
There are sectors that are slightly more insulated from Google, fashion and beauty for example, where purchase behaviours are often driven by inspiration – and impulse has been able to grow commerce revenue outside of just search, driving performance from social, direct traffic and owned newsletter audiences. The US-based Who What Wear, acquired by Future in 2022, is a good example of this.
Longer-term, it’s been interesting to see the growth of social media platforms as search engines, especially TikTok. At the moment, these searches are more focused on “how to” and informational style queries, less so on product review and “best” terms in categories that are driving a lot of publisher commerce revenue. Nevertheless, brands are already seeing success within fashion and food categories here and it’ll be fascinating to see if this starts to extend to other areas.
On Amazon, it’s a slightly different situation to Google. They have had a firm grip on publisher eCommerce performance for the last almost decade and – even now – I see publishers with 80-90% of commerce revenue with Amazon, which seems an extremely high and risky number.
There are a number of reasons for this, Amazon can cover more products than any other retailer, has generally high conversion rates, and decent commission percentages. For various reasons, hwoever, publishers have found it hard to do real-time price comparison at scale, and so often publishers choose only to link to one retailer for most products; inevitably, that mostly becomes Amazon.
However it’s easier for publishers to diversify away from Amazon than it is Google. In the last several years, US retailers such as Best Buy and Walmart have become equals to Amazon, significantly improving conversion rates and product catalogues as well as paying far higher CPAs than Amazon to level the playing field.
The UK market has been much harder with large retailers like John Lewis and Currys not always being competitive on a per-click basis. But savvy publishers are able to find categories where these retailers can significantly outperform Amazon. Specialist retailers have also started to understand the value in affiliate marketing and are becoming increasingly competitive options with publishers able to build strong direct relationships with them – in stark contrast to what, for many media companies, is an impersonal and one-way relationship with Amazon.
Better tech is also coming along for price comparison, with Skimlinks and Sovrn as well as other independent tech companies building tech solutions for publishers to use -and more publishers investing in owned and operated price comparison tech as Future did a long time ago with Hawk.
Despite all of the above, Amazon will still inevitably play an over-sized role in the commerce ecosystem, and even publishers’ best efforts at diversifying will still see Amazon around the 30-40% of revenue mark. The underlying risk and worry amongst publishers is that Amazon will do something drastic like completely shutting down its Associates programme, or reducing CPAs to a small nominal amount. It seems far-fetched, but they have ‘form’: in 2023 they closed their On-Site Associates program at very short and abrupt notice after publishers had invested millions in building teams around it.
Interestingly, there is a growing ecosystem of companies building around Amazon associates that is gaining traction. Platforms like Levanta allow Amazon sellers to pay additional CPAs to publishers for sales of their products, essentially allowing publishers to double dip on commission rates. It’s still early days but it’ll be interesting to see how this develops – and if it points to a future where Amazon commissions aren’t set or paid by Amazon but, instead, determined by the individual sellers.
What are the implications of AI for eCommerce?
I probably hold a relatively boring and mainstream view on AI in publishing in general and that applies to publisher eCommerce too.
AI has the ability to build amazing tools that automate and optimise key tasks in the publishing workflow. It can be used to aid high level research, to improve data and reporting, to help summarise complex workflows and processes. Within media eCommerce, there are applications to use natural language processing, such as Google NLP, for better product and article matching and real-time retailer optimisation.
I don’t see AI, in any way, as a replacement for writers in the traditional sense, and I think the best commerce content will be original, authentic and human, relying on real testing and insight that AI can’t replicate. As AI use spreads across the internet, particularly with AI generated “reviews”, I think publishers have an important role and duty to act as the source of trusted and authentic information. Publishers should see human researched and written content as a differentiator that allows them to be trusted and valued.
The generative AI deal revealed this week as part of the Google News Initiative certainly concerns me. The local publishers it’s targeting will be hard pressed to refuse extra cash and extra content output, but I worry about the implications for reader trust. It feels like Google is squarely talking out of both sides of its mouth by trying to “reward” human written stories in search AND funding the creation of AI driven content. Even if they currently justify these approaches by the differing use cases, it’s only going to get murkier and feels like Google is at a philosophical crossroads with regards to AI, content and what’s best for their users.
I’m particularly interested in the use of AI for data and reporting automation. At Future, one of the best changes we made to the eCommerce team was to build-out a trading function, that looked at weekly performance of all our content and identified areas for improvement and optimisation. There are lots of possibilities for AI to improve and automate areas of this process and deliver a greater flow of actionable insights .
What are the best eCommerce ideas you have seen?
Sadly, I think a lot of the industry has been in copycat mode the last two years. When you consider publisher traffic struggles, post-Covid dips in consumer spend, and widespread layoffs, its been tough for media companies to invest in innovation. I think there’s been a lot of catch-up happening across the industry and the net result is that media companies are deploying the same eCommerce strategy, chasing the same intent search terms, with similar content plans, templates and broadly similar commercial operations.
The ones that have been winning are those that have been executing that strategy best. The innovation has mostly come around the periphery of this, and outside of the media companies themselves. As I mentioned earlier, there’s been some progress on ‘white label’ price comparison services, allowing publishers to move away from Amazon reliance, and there’s tech being pitched that allows readers to checkout directly on page, like that offered by Instagram and other social apps. But I have my doubt about the benefits for both readers and publishers. There’s also been some great new tools, such as Affilimate, that help publishers aggregate eCommerce reporting across the whole affiliate ecosystem.
The best innovation from within media companies is coming from those who are attempting to build direct relationships with audiences, via email, direct and subscription models, where product recommendations, via inspiration rather than intent, can form one part of that relationship. As mentioned earlier, I’ve been impressed by fashion brands that have been able to diversify away from search. But, in my opinion, there’s a lot more to be untapped here and in other verticals too, especially for specialist media brands that operate in verticals with high consumer product spend and passionate, highly-engaged enthusiasts – think sports and hobbies.
I’m not, however, as optimistic as many others about direct eCommerce for publishers. It certainly can be a successful strategy if you have the brand recognition and cache to support it, such as Elle has done in Japan. But the threshold for successful entry is very high and only brands that truly transcend media have a real chance of building long-term successful stores. Perhaps that is what Hearst and Conde Nast are thinking. To be successful, it also requires a lot of capabilities that most digital media companies aren’t as good at, such as fulfilment and customer service, which can create significant additional costs that eat into any margin gains vs affiliates.
What’s your best advice to media owners on how to maximise eCommerce?
Firstly, it’s a cliche, but there’s no better advice than to ‘think audience first’. Unlike advertising, you can’t drive eCommerce revenue without doing a brilliant job for the reader and helping answer their questions on any given product or service. I’m continuously surprised at the number of media brands which lose track of what the reader actually wants from their product content and over-complicate what should be a quite simple process.
Writers need to ensure that their content caters for as wide a variety of use cases and price points as possible, and not just the “best” products. Often the products that perform best on page are the ones that have flaws but are cheaper and more attainable for most. In a box-ticking approach to E-EAT, I also think brands are over-complicating content in order to cram in “expertise”, often to the detriment of the reader, who doesn’t want 10 paragraphs on how you tested, or what your expertise is, before they reach a single product recommendation. You wouldn’t expect to walk into a shop and ask a sales assistant for a recommendation, only for them to spend the first two minutesa talking about their expertise and how they tested the products; the same applies when you click on a buying guide. By all means, you should be including this information, but do it succinctly and just don’t do it at the expense of getting to the point of the article, ie the product recommendation. This also means making pages easy to read and navigate. Make sure you have the ability for readers to jump to the product or use case they care about the most, and don’t overload the page with ads. eCommerce yields will always outperform any incremental ad revenue on these pages.
Secondly, heavily invest in your data and reporting capabilities. It’s probably the biggest blind spot I see with publishers, which have very little insight about what’s actually happening across their content. A huge portion of the continual year-on-year affiliate revenue growth we achieved at Future was driven by improvements in page yield. This included building a trading team, optimising product and price point recommendations based on reader purchase date, optimising page and category level retailer mix amongst lots of other things. None of this would have been possible without a significant multi-year investment in eCommerce data and reporting capabilities.
Finally, investment in eCommerce needs to come from the top, and eCommerce needs to be given space to breathe. You can’t just decide to “do eCommerce” and expect immediate returns. When I think back as to the reasons why eCommerce grew so successfully at Future in the past decade, much of it came down to the buy-in and support we had from the CEO down. We had investment in technology with Hawk, investment in reporting and, crucially, an understanding throughout the organisation that this was a strategic focus. All editorial teams had to prioritise it, we kept the pages purposefully ad lite – despite inventory pressures – and, commercially, eCommerce was able to build its own partnerships without existing advertising partnerships dominating the selection of retailers. This buy-in across the whole organisation was paramount to us being able to flourish and grow as quickly as we did.
