My 20 Predictions for Content Marketing, PR and SEO in the Next 5 Years

As someone who has been doing some form of Content Marketing, PR, and SEO since 2006, I've seen the strategies and tactics change massively.

The pace of change on the Internet is reaching the knee of the exponential hockey stick and things are getting WEIRD, to say the least. Artificial intelligence is impacting nearly every aspect of our work as #contentmarketing , #PR , and #seo pros.

What follows are my predictions for the next 5 years for those creating content for the purpose of SEO/Linkbuilding/Earning Press.

Agree? Disagree? Let me know in the comments.

TLDR:

1. Middling Content Dies
2. AI content tools increase productivity/output volume - Leading to glut of mediocre content but increasing efficiency for those with a human in the loop.
3. Automation of content distribution improves (in terms of breaking up content, automating content modifications for the distribution channel, and posting automation).
4. Bots are the secret weapon of the most successful.
5. Text to Video/Audio and Audio/Video to Text become ubiquitous – Search indexes grow exponentially as a result.
6. Blog/Specific site readership declines as recommendation engines and aggregators continue to grow, subscription readership grows on Substack, Patreon, Etc. (but only for those providing massive value).
7. Alternative SEO grows massively (Tiktok, Amazon, Etsy, Shopify, Reddit, App Store, Etc.).
8. Content Agencies and internal content teams become more like newsrooms. Presenting entirely new information/value becomes the standard bar.
9. New AI and Data processing tools, alongside massive increases in open data availability, will democratize the ability to do data journalism for individuals and brands.
10. Brands increasingly becoming the drivers of investigative stories/news.
11. Outreach/PR will become harder, requiring good PR to be hyper-targeted, and provide real value (read: data-driven stories, actual news).
12. Deepfake representatives and fine-tuned AI "Oracles" will start to become brand/person face/advocates, taking on customer-facing roles in marketing and sales.
13. Longtail Shrinks, especially for high-intent terms (Google will capture with AI).
14. Google will fail to keep up with the AI glut. Dead Internet Theory is proving right. Community signals become more important because they are harder to fake.
15. Video/Audio/Image results will become much more common as Google improves the indexing of words said in video.
16. New Search Engines gain some of Google's share.
17. Reactive/Newsjacking campaigns from brands grow substantially.
18. Brand-based social media becomes much more reactive and voice shifts to meet the audience (less branded, more human).
19. Maintaining brand voice across all channels becomes very difficult (Brand Voice becomes more determined by channel than overall).
20. Facebook Groups lose influence, Linkedin Groups gain influence, and walled-garden chat groups share the best content.

1. Middling content dies.

For the last 15 years, there's been a quiet arms race happening between Google and marketers. SEOs and marketers have worked hard to become increasingly proficient at capturing traffic from the long tail by creating massive amounts of content (much of it made just to rank in Google, not necessarily to provide much actual value).?

Content farms, blogspam, Private Blog Networks, programmatically created content (Madlib style and article spinning), and now AI-created content has led to a massive and growing glut of useless, copycat, or otherwise poor-quality content. This glut makes Google's job harder, requiring it to become increasingly proficient at parsing quality from crap while having several orders of magnitude more to sift through than it did 5-10 years ago.

?Most major updates to Google's algorithm updates have been made to address "Quality," and Google is making big strides. At the same time, Google is also improving its ability to understand the goal and intent of the searcher, leveraging new AI in the form of Large Transformer Models like BERT to more accurately match query and result.

This progression points in a general direction, one we've seen firsthand at Fractl. Sites that on average have a low quality (read: not particularly useful, interesting, or unique), will not succeed in the long term when it comes to organic search rankings. Whether it is a past Google update, or a near future update, sites that fail to provide significant value and answer search queries successfully, will continue to decline in their ability to rank.

As the volume of mediocre content continues to increase exponentially while Google continues to improve its ability to recognize quality, vast volumes of content will be relegated to obscurity in search. The slow death of middling content will accelerate as these two trends collide.

2. AI content tools Increase productivity/output volume - Lead to a glut of mediocre content but increase efficiency for those with a human in the loop.

Highly proficient large transformer models like GPT-3 are fundamentally changing content creation. Previously bottlenecked by needing human writers, those intent on creating content at scale, can now do so.?

As of March 2021 (last time data was released), GPT-3 was generating 4.5Billion words per day.? Now, 18 months later, it's probably safe to assume the volume of words being generated by GPT-3 and similar transformer models is orders of magnitude greater, perhaps even nearing 100Billion words per day.

That's roughly 10-12 words per day for every single human on earth.

This means that we are nearing (or perhaps now crossing) an inflection point where there is more AI-generated text being generated each day than there is human-written content.

Unfortunately, AI-generated text is still highly prone to going severely off the rails. Without a "source of truth," GPT-3 (and all the Saas products that use GPT-3 as a backbone) will regularly generate outright falsehoods. These models produce inaccurate information as a rule. Without a human in the loop, there's little chance the content created by AI will provide significant value to searchers.? This is the core issue at play in the aforementioned trend towards a growing glut of crap content.

However, all is not lost when it comes to AI-generated content. When looked at as a support tool, AI-generated text can massively speed up human writing. GPT-3 and similar can be used with incredible efficacy. Tools like Jasper (Gpt-3 or similar tech stack at its core), claim improvements up to 10x in speed of writing when using their assistant. If you've ever used these tools, it's clear this is a plausible figure.

In the next five years, most content creation workflows will incorporate AI assistants, and they will do so at all phases of the writing process. Ideation, outlining, organization, the writing itself, copyediting, and adjusting tone/voice/style, will all be increasingly influenced by these massive AI language models. When utilized alongside a human, the potential for improving both quality and speed is profound.

3. Automation of the content distribution improves (in terms of breaking up content, automating content modifications for distribution channel, and posting automation)

Content distribution is a critical part of the content marketing process, and it is an area that is ripe for automation. In the past, content distribution has been a largely manual process, requiring a fair amount of time and effort to properly place content where it will be seen by the right people.

The next five years will see massive improvements to the automation of content distribution in two ways:

a. Automation of content repurposing

AI will play a big role here, especially with the release of new tech like OpenAI's "Whisper," which can do human-quality speech-to-text. New SaaS products will continue to emerge that will help to create a variety of automated pipelines that will allow content creators to create a single piece of content, and then have that content automatically repurposed into a large variety of new forms/mediums.

Source: Growth in searches for AI Image Generators from Google Trends

Specifically:

• Automatically turning written content (blog posts, aggregated tweets/threads, newsletters, etc.) into video and audio content. AI text-to-image and text-to-video technology will be the primary drivers of this kind of tech. Meta's "Make-a-Video" model and Stable Diffusion are early AI models leading this technology. Future versions of this tech, and the SaaS products that enable them, will make it trivial engaging visual content from text alone.
• Automatically turning video/audio content into written content (blog posts, tweets/other social media posts, newsletters, email, etc.)
• Automatic creation of previously unpublishable content
• Automated human-level transcription will make it easy to capture and publish content that previously would never have seen the light of day. One good example is the publishing of recurring voice meeting transcripts. AI can now accurately transcribe, organize, summarize, and synthesize human speech into useful and readable text output. Value can now be extracted from nearly any form of media, and used to create more media, in different mediums, automatically.

b. Automation and optimized distribution of content to many channels.

There have long been platforms for helping practitioners distribute content to the ever-growing list of social media platforms, video-sharing sites, news aggregators, etc.

Until recently, most of these tools have been "Dumb." In other words, they simply facilitated the technical process of posting content to different platforms. They did almost nothing to optimize the content being distributed.

Each platform has its own structure that leads to success or failure. The titles or descriptions used in posts, the icons/images used in posts, the length of the post, the social norms and content trends of the given distribution platform, audience differences between platforms, and much more must be considered when posting content.? New tools will vastly improve this process, automating many of these aspects of successful distribution.

Within 5 years, I think we will see a SaaS tool that can:

Take a written blog post and:

• Automatically create a set of tweets from the blog post, either to be scheduled out over time or as a threaded version of the article to live on Twitter
• Automatically create a long-form Youtube video from the blog post.
• Create multiple short-form videos from the blog post for Youtube Shorts, Tiktok, etc.
• Create audio and video versions of a podcast from the blog post

4. Bots are the secret weapon of the most successful marketers.

Until recently, bots had limited abilities and were mostly unable to impersonate real humans convincingly. Large language models like GPT-3 have fundamentally changed the equation. Bots regularly pass for humans on Twitter, Reddit, and everywhere else, especially short-form platforms.

Source: https://xkcd.com/810/

Bots represent a force multiplier, enabling those controlling them to create dozens, hundreds, or thousands of bots that can represent their interests, write content, and automatically distribute that content to the right place at exactly the right time.

It is now trivial to write a Reddit or Twitter comment bot, that can write human-looking comments within real-time conversations. A good example would be a Reddit or Twitter bot that makes product recommendations to any Twitter user who mentions a competitor brand name, or a bot that constantly watches for negative sentiment posts about X topic or Y brand, and writes positive remarks.

5. Text to Video/Audio and Audio/Video to Text become ubiquitous – Search indexes grow exponentially as a result.

Generative AI will have a dual impact on content and search. First, AI-generated content will massively expand content creation volume, improving both speed of content creation, and accessibility of content creation at scale. Models like OpenAI’s Whisper demonstrate we have passed a critical threshold in the accuracy of human voice transcription.?

For both of these reasons, the pace of growth for search indexes is poised to explode. In addition to massively more content being created, huge swaths of previously difficult or impossible-to-parse and index content will become easily indexable. All audio/video/image content will be able to be understood by search engines in the same way as text is understood, allowing search result pages to include answers to queries that exist inside videos, inside podcasts, and inside images. This is the beginning of where ALL Content becomes parseable, indexable, and searchable to end users.

6. Blog/Specific site readership declines as recommendation engines and aggregators continue to grow, subscription readership grows on Substack, Patreon, Etc. (but only for those providing massive value).

Content recommendation and aggregation technology have passed a critical barrier. As any TikTok user will tell you, we’ve reached a point where content consumers expect nearly every piece of content delivered to them to be highly valuable, relevant, and interesting to them in particular.

I believe this means attracting followers, regular blog readers, Patreon supporters, and email list subscribers is getting exponentially harder. Why even follow someone when you can be relatively certain the recommendation engine or crowd voting/aggregator will make sure you see anything important to you in the future? “I trust TikTok to bring me back here” is a common comment found in TikTok videos, an offloading of the work of self-curation of content. Content consumers are increasingly trusting recommendation algos and their advanced AI to comprehensively deliver any/all content actually worth seeing to that individual, making the bar for following/subscribing to a content creator that much higher.

This raises the overall bar required to capture ongoing attention as a content creator. The value provided needs to exceed what content consumers can get from recommendation engines and content aggregators like Reddit in order to attract regular readership/viewership.

I think this also means that private subscription content will become more important. While the bar to acquire a subscriber will be higher, consumers are realizing that there are many places where important hidden/gated exists outside the purview of content aggregators and recommendation engines. Substack, Patreon, Onlyfans, and other content-walled gardens should see increased popularity but also increased difficulty in attracting subscribers.

7. Alternative SEO grows massively (Tiktok, Amazon, Etsy, Shopify, Reddit, App Store, Etc).

Google’s supremacy as the canonical place to search is eroding, and savvy searchers are increasingly realizing they can get better search results from other places. Even Google is acknowledging this shift . Appending site:reddit.com (or simply searching on Reddit) is also extremely common, as are site searches on Github, Stackoverflow, Amazon, and many other sites. This is likely the result of two factors:

• Massive improvements to onsite search technologies, including AI-driven search tech like that from https://www.algolia.com/
• Ongoing and increasing issues with search spam clogging up Google’s index.

Why search google for a product recommendation, knowing the top 10 results are from affiliate aggregators with hidden motives, when you can search Reddit for real(maybe) evaluations and recommendations from real(maybe) people. Why search Google for programming-related questions when you know all your answers will most likely be found on Stackoverflow or GitHub? Just search those sites directly. Especially as the quality of onsite searches continues to improve drastically with the implementation of new AI-based search tech.

8. Content agencies and internal content teams become more like newsrooms. Presenting truly new information becomes the standard bar.

Increasing competition for attention will continue to raise the bar for breaking through the noise. This is true for individual content creators, agencies, and brands alike. The rehashing of known/existing information will become increasingly ineffective, and there will be growing pressure for all content creators to be contributing something truly NEW.

This will lead to the increasing sophistication of content across the board, where agencies and brand content teams will become better and better at generating truly novel work.

For many, this will look like a transition to a “newsroom” style content environment. The expected value of the content created will be understood as how much new information is being presented, and how relevant that information is to current news and information-seeking trends.

Exploration and analysis of data will become a primary method for content creators looking to “make news.” Similarly, investigative reporting style content will also see significant growth.

The key overarching trend here is the growth in demand for “new information,” and those that best create new information will attract the greatest volumes of new viewers.

9. New AI and no/low code data processing tools, alongside massive increases in open data availability will democratize the ability to do data journalism for individuals and brands.

Source: Growth in Figshare, Dryad, Dataverse, Mendeley Data, Open Science Framework and Vivli DOIs in the published literature (2011-2021)

The volume and accessibility of open data have exploded in recent years and will continue logarithmic growth into the foreseeable future. At the same time, we are also seeing massive growth in the companies, tools, and infrastructure to support the collation, exploration, and distribution of open data.

As improvements in data availability and accessibility compound, a significant shift is happening. Data and analysis of that data are being democratized, and I believe it is pulling us toward a golden age of individual citizens and corporate data journalism.

As I mentioned previously, the currency of the future attention economy will be based on bringing “NEW” information to the table. Finding new and compelling stories in data will be one if the best, and easiest ways for content creators and brands to bring something entirely new to the table, and to do it consistently.

10. Brands increasingly becoming the drivers of investigative stories/news.

As more brands realize the increasing need to present truly new information in order to stand above the fray, they will increasingly begin leveraging open data and low/no code data analysis tools to do hard-hitting investigative journalism. What has long been under the purview of news organizations, will see a gradual shift, where brands start to become the originators of these types of stories, with PR folks acting as the intermediaries, helping brands connect their investigative work to the news organizations that can distribute the stories effectively.

11. Outreach/PR will become harder, requiring good PR to be hyper-targeted, and provide real value (read: data-driven stories, actual news).

For PRs, things are going to get crowded. The total volume of pitched stories is going to continue to increase, and it will become harder to get noticed in journalist inboxes. Frustration will continue to grow for journalists who are on the receiving end of poorly targeted pitches of stories that lack newsworthiness. At the same time, the volume of great stories being pitched by individuals and brands will also increase.

The linchpin for successful brands and creators pitching stories will be in how well the PR is matching the story being pitched, to the needs of the individual journalist. Those engaging in low-touch outreach (email blasts to large numbers of journos, or old-school press releases) will continue to see declining results.

Conversely, a new crop of extremely useful AI-powered PR tools will begin making effective PR easier to scale. Journalist intelligence tools will start to see uptake, as it becomes increasingly obvious that success in PR is a result of successful targeting (matching the story being pitched to the beat of the journalist), and hyper-personalization of the email. Both of which can be addressed by a new crop of big data and AI-driven SaaS tools.??

12. Deepfake representatives and fine-tuned AI "Oracles" will start to become brand/person face/advocates, taking on customer-facing roles in marketing and sales.

This one is a bit more speculative, but I believe we will start to see a new class of AI-driven “Oracles” and “Deepfake Representatives” that will start to take on customer-facing roles for brands and individuals.

These Oracles will be AI models that are “trained” on an individual or brand’s public persona and will be used to generate responses to customer inquiries, social media comments, sales questions, etc.

Similarly, we will also see the rise of Deepfake Representatives, which are AI-generated faces and voices, that are used to represent an individual or brand in public-facing roles.

These could take the form of chatbots, or fully autonomous AI-driven social media avatars, that post and interact on behalf of the individual or brand.

I believe these technologies will be used in two ways:

• To replace customer service/salespeople for high-volume, low-touch interactions.
• To supplement/augment existing customer service/salespeople for higher value interactions.

For example, a Deepfake representative could be used to automatically answer simple questions from customers on social media or to automatically post updates/announcements on behalf of a brand.

These Oracles and Deepfake representatives will not be perfect, but they will be good enough to be useful, and they will be a fraction of the cost of human employees. I believe we will start to see widespread adoption of these technologies in the next 5-10 years.

13. Longtail shrinks for high-intent terms (Google will capture with AI)

The longtail of search has been a boon for content creators and marketers, providing a nearly endless supply of low-competition, high-intent keywords to go after.

This has led to a gold rush of sorts, where content creators and marketers have raced to create and rank content for these terms.

However, I believe this trend is coming to an end.

Google has gotten very good at understanding user intent and is increasingly able to provide results for high-intent queries without the need for longtail keywords. In addition, Google is also getting better and better at understanding the relationship between entities (people, places, things, etc.), and can provide results for queries that are not explicitly stated in the query itself.

This trend will lead to a shrinking of the longtail, as Google increasingly captures more and more of the search market with its AI-driven search results.

This will make it harder for content creators and marketers to find low-hanging fruit in the form of longtail keywords and will require a shift to a more holistic approach to SEO.

There will be a renewed focus on understanding user intent, and on creating comprehensive content that covers all aspects of a given topic.

14. Google will fail to keep up with the AI glut. Dead Internet Theory proving right. Community signals become more important because they are harder to fake.

The total volume of content on the internet is growing at an exponential rate, and Google is struggling to keep up. This is leading to several problems, including:

• An increase in the number of low-quality, spammy, and/or fake results in Google’s search results.
• An increase in the number of “broken” search results, where the page that is returned by Google does not match the query.
• An increase in the number of “zero results” pages, where Google is unable to find any results that match the query. I believe this trend is only going to get worse, as the volume of AI-generated content continues to increase.

This is what's been prophesized as the "Dead Internet Theory" where the internet becomes so flooded with low-quality content and bot activity, that it becomes unusable.

I believe Google will eventually be forced to make a major change to its search algorithm, where it will begin to give more weight to “community signals” (likes, shares, comments, etc.) in its search results.

This is because these signals are much harder to fake and because they are a good indicator of the quality of the content.

This change will have a major impact on content creators and marketers, who will need to focus on creating content that is more likely to be liked, shared, and commented on, in order to rank highly in Google’s search results.

15. Video/Audio Results will become MUCH more common as Google improves the indexing of words said in video and audio content.

Google has long had a problem with indexing the words spoken in video and audio content. This has led to several problems, including:

• The majority of the words spoken in video and audio content are not indexed by Google.
• The words that are indexed by Google are often inaccurate, leading to “broken” search results.
• Video and audio content are often not included in Google’s search results, because the search engine is not able to understand the content.

However, I believe this is changing, as Google is getting better and better at indexing the words spoken in video and audio content. In addition, Google is also starting to include video and audio content in its search results, as it can understand the content better.

Source: https://weareroast.com/resource/the-roast-serp-landscape-report/

I believe we will see a significant increase in the number of video and audio results in Google’s search results over the next 5 years, as the search engine continues to improve its indexing of this type of content.

As a result, search results pages will become increasingly “rich”, with a mix of text, image, video, and audio results. This will have a major impact on content creators and marketers, who will need to create content in a variety of mediums, in order to rank highly in Google’s search results.

16. New search engines gain some of Google's share.

Source: Backlinko DuckDuckGo stats

Google has long had a monopoly on search, and this has led to several problems, including:

• Google’s dominance of the search market has led to a stagnation of innovation in the search space.
• Google has used its monopoly power to stifle competition and innovation in the search space.
• Google has been able to charge advertisers exorbitant rates, due to its monopoly power.

However, I believe this monopoly is coming to an end. Several factors are leading to this:

• The rise of alternative search engines, like DuckDuckGo, which are starting to gain market share. - The rise of new AI-driven search technologies, which are making it easier for non-Google search engines to provide better results.
• The European Union’s antitrust investigations into Google, could lead to a break-up of the company.

I believe we will see a significant increase in competition in the search space over the next 5-10 years, which will be good for consumers, and bad for Google.

17. Reactive/Newsjacking campaigns from brands grows substantially.

To be successful, PRs and content creators will need to have a deep understanding of the news cycle, and of the trends that are driving the news. They will also need to have a network of contacts that they can rely on to help them quickly bring their ideas to life.

Multiple trends are leading to this:

• The increasing volume of content and the shrinking attention span of audiences.
• The increasing number of channels and platforms that content must be distributed to.
• The increasing number of content creators and marketers competing for attention.

I believe the successful PRs and content creators of the future will be those that can quickly generate and execute on ideas that are relevant to the current news cycle. This will require a deep understanding of the news cycle, and of the trends that are driving the news.

Tools that provide real-time insights into the news cycle, and into the trends that are driving the news, will be essential for success. As will be tools that help to quickly generate and execute on ideas.

The expectations of journalists, when being pitched stories, will continue to increase. They will be looking for stories that are relevant to the current news cycle, and that can provide new and insightful perspectives on the news.

Again, data analysis and investigative journalism will be two of the best ways to create truly novel content that can break through the noise.

18. Brand based social media becomes much more reactive and voice shifts to meet the audience (less branded, more human/authentic).

The social media landscape is constantly changing, and brands are struggling to keep up. To be successful, brands will need to be much more reactive to the changes in the social media landscape and will need to adjust their voice to meet the needs of their audience.

Some of the specific changes that brands will need to make include:?

• Shifting from a “branded” voice to a more “human” voice.
• Customizing brand voice to the given channel/platform. Each social media platform has its own norms and its own audience.
• Authenticity will be key.
• Being more responsive to current events, and the needs of their audience.
• Posting less “salesy” content, and more content that is truly helpful, informative, and presents entirely new information.

Brands that can make these changes will be the ones that can break through the noise, and connect with their audience.

19. Maintaining brand voice across all channels becomes very difficult. Brand voice becomes more determined by channel than overall.

As the number of channels and platforms that brands need to be present on continues to grow, it becomes increasingly difficult to maintain a consistent brand voice across all of these channels.

This is because each channel/platform has its own norms and its own audience. What works on one platform, may not work on another. As a result, brands will need to start customizing their voice to the given channel/platform.

This will require a deep understanding of the norms and the audience of each platform, and a willingness to experiment with different voices on each platform.

This will be scary for many brands, who have long been focused on maintaining a consistent brand voice across all channels. However, I believe it is the only way for brands to be successful in the future.

20. Facebook groups lose influence, linkedin groups gain influence, walled garden chat groups share the best content.

The social media landscape is constantly changing, and this is having a major impact on the way that people consume content. In particular, I believe we will see a shift in the way that people consume content in groups.

Specifically, I believe that Facebook groups will lose influence, LinkedIn groups will gain influence, and walled garden chat groups will become the primary way that people consume content in groups.

Here’s why:

• Facebook groups are becoming increasingly noisy, and it is difficult to find truly high-quality content.
• LinkedIn groups are becoming more and more niche, and the content is becoming more and more targeted.
• Walled garden chat groups are becoming a primary way that people consume curated content in groups. This is because they can provide a more intimate and personalized experience and because the content is more likely to be truly high-quality.
• More tightly controlled information can be shared in closed communities with less risk, and higher certainty that the community will provide value in exchange for access.