Rand's 10 Predictions for 2016

#1: Data will reveal Google organic results to have <70% CTR

For many years, Google gave public numbers about the ratio of ads to organic click-through rate. Historically, these were between 80–85% for organic results and 15–20% for ads. But for years now, Google's been getting incredibly aggressive with non-organic results (knowledge graph, various AdWords formats, instant answers, etc). I think 2016 is the year we get better data and real results despite the search giant's silence on the subject.
That's in part going to come from the search marketing industry, where folks like AWR have builttools to track CTR reported in Google Search Console, and from outsiders like SimilarWeb and/orJumpshot, who have access to clickstream-level data.
My prediction is that, on average, across Google's billions of daily searches, even accounting for searches that generate multiple clicks, less than 70% of searches result in at least one click on an organic result.
That's not particularly gloomy though, given Google's growth. Even if we found the number to be 50%, there's still more opportunity in Google's organic results today than there was 4 years ago thanks to the incredible growth rate of searches (desktop's flattened, but hasn't shrunk in that time, and mobile's skyrocketed to more than double total search volume).

#2: Mobile will barely cut in to desktop's usage and its growth rate in developed countries will slow

I'm not sure if 2015 was the year of mobile, but I think 2016 will be the year that growth rates for mobile begin tapering off. Note — I'm not saying we're at "peak mobile," but I am saying the growth rate has reached its apex and I think in 2016, the percentage growth numbers will fall.
Simultaneously, I think we may have reached the peak of desktop/laptop decline, and I expect that in 2016, desktop usage rates will stay largely unchanged. Most everything that mobile was going to replace or take away from desktop use has been taken, and I'm skeptical that things like creative work, programming, long-form writing, computer gaming, and other tasks that big monitors and full keyboard+mouse inputs were made for can be successfully cannibalized by the screen that fits in our pockets.

#3: Twitter will figure out how to grow again

It's been a rough couple years for Twitter, but I'm bullish on the company in 2016 and long term. It takes new CEOs some time to right the ship and it takes time for changes in a company's leadership to impact their growth. It might be toward the end of 2016, but I believe we're going to see numbers from Twitter that improve their market position and their stock price (note: I hold no public market stock).
For marketers, this means investments in Twitter will pay bigger long-term dividends. I think folks who learn how to effectively advertise with Twitter, how to earn an audience there, and how to leverage the conversations — not just to promote, but to learn and participate — will have remarkable opportunity. There's flexibility in the features afforded by Twitter, combined with the way their stream and amplification function makes for much greater creative and content potential than the limits imposed by Facebook and Instagram in their organic streams (which is not to say that Facebook and Instagram aren't incredible opportunities themselves — they both are).

#4: Social content engines will become a force

In 2015, I started using 5 new content recommendation engines:
  • Pocket's New, Beta Recommendations which recommends content to me based on how other Pocket users are employing it, and lets me follow other Pocketers who publicly "recommend" content they read. BTW — if you want to follow my recommendations, you can see those here.
  • Nuzzel – A great app and website that's very nearly a replacement for my favorite content engine of all time, the long-defunct Trunk.ly. It aggregates links shared on Twitter and Facebook by folks you follow, and shows content from "friends of friends," too.
  • ProductHunt – A popular website in the startup & tech world that works like Reddit for software and apps. Creating an account and following others can help personalize the site's suggestions and what appears in their daily emails.
  • Snapzu – a new(ish) service that aggregates content that's been shared in other places. They call themselves a "community of communities" and have an invite system that keeps signal:noise ratio high.
My prediction is that one or more of these will rise to the level of some of the second-tier social networks in popularity, referral traffic, and active users. That's probably Pocket, but it wouldn't surprise me to see Feedly or even a brand new player like Refind make a big splash. Social networks make content delivery and signal:noise secondary priorities behind engagement and conversation. I believe some of the players focused on this space will have an outstanding 2016 — content consumers need them.
33K followers only a month after the beta launch?? That's a powerful platform...
Of course, as these platforms take off, they also become of deep interest and importance to marketers, especially content marketers, as social amplifiers, influencers, and probably, as a result, some level of search rankings will get a boost if content does well on them.

#5: Yext will IPO, prompting even more interest in the world of local listings

In 2014, there were numerous articles predicting Yext's IPO. It didn't happen in 2015, but I think this is the year. They've executed well, built a moat with their contracts and their powerful sales team, and they play in a space where businesses desperately need help — local listings.
I think Yext's rise will create a great deal more interest in the local listings world from tech, financial, and business folks who've previously mostly ignored it. I'm also bullish on Moz Local's growth (a semi-direct competitor to Yext), and I actually think Yext's making of this market will help with that, just as Hubspot's IPO helped so many companies in the inbound and content marketing worlds.

#6: The death of normal distributions will hit both publishing and search results hard

I found Alex Danco's recent post about the death of normal distributions to be prescient (despite some criticism of the lack of data underlying the graphs). I agree that we're moving away from a world of normal distributions in demand curves and towards a winner-take-all model. The "long tail" theory of digitization has been mostly proven false (with a few notable exception areas like keyword search demand and social network connections).
In 2016, I think we're going to see more of what happened to GigaOm and what was described inThe Sad Economics of Internet Fame. That is, content becoming so ubiquitous and so hard to monetize that only the big winners will be able to keep up the game. Adblocking's growth may play a part in that, as will Google's algorithms that reward faster-loading, higher-engagement pages that don't drive away visitors with interruptions and annoyances.
Likewise, I think we're going to see more consolidation overall in search results, with fewer unique domains in the top 10 rankings of search results (we should be able to measure this via MozCast). I'm not happy about this prediction — my hope is that it's wrong. I believe in and love the diversity of content that the web and its organic marketing channels have enabled.

#7: The rise of adblocking is going to trigger attempts at legislation and incite more sites to restrict adblocking users

Adblocking was part of a huge conversation in 2015, and my guess is that the reaction to this growing technology is going to mimic how entrenched players have reacted to technology leaps in the past — by trying to legislate it away. I anticipate that in either the US or the EU, some form of government action will arise (in the US, most likely due to lobbying AKA our legalized system of bribery) to "protect the interests of publishers and journalists who serve the public good."
If that exact quote appears from a source, I think I should get double the points, don't you? :-)

#8: DuckDuckGo will be the fastest-growing search engine of 2016

Google had a good 2015. So did Bing, actually (given they now power AOL and had relative success with Windows 10 and the Edge browser). But in 2016, I think our fastest-growing broad, web search engine will be DuckDuckGo, whose growth in 2015 was pretty remarkable, too:
Via DuckDuckGo's wonderfully transparent Traffic Stats page
Part of this is the continuing storylines of privacy, part of it is the early adopters in the tech world moving to DDG, and part of it is DDG's accomplishments as an engine — for a lot of queries, they're now truly as good as Google (although they miss some, too). If I weren't in SEO, with a professional need to see as many Google results as possible, I'd honestly consider switching.
Given DDG has many fewer queries to being with, I'll make a slightly bolder claim — their growth rate in 2016 will exceed their 2015 growth rate, and no currently-existing other major web search engine will grow as quickly. If Apple or Facebook decide to get directly into search, well... A) they should buy DuckDuckGo and B) all bets are off.

#9: Content marketing software for the non-enterprise will finally emerge

For a long while now, large enterprises and big content shops have been the sole beneficiaries of content marketing software that tracks performance, suggests improvements, helps with publishing, ideas, outreach, etc. But I think 2016 will be the year that some of the SMB and mid-market players emerge. There are numerous players poised to take this market, including:
  • Buzzsumo – To date, they've been the strongest performer of the bunch, but I'd categorize them primarily as a research tool (albeit an amazing one). They've broken into content alerts, but I think they might do more with ongoing campaigns, suggestions, and analytics in 2016.
  • Kapost – Built for content creators to pitch and share ideas, manage the content publication process, and distribute effectively across channels, Kapost has been around for a few years, but continues to improve and gain market traction among their target B2B customers.
  • Priceonomics – I think these folks will make a big splash with this product once it's public. They've almost exactly followed Moz's model, first building an impressive blog and a passionate following, and now launching what looks like very interesting software that they themselves have been using internally.
  • Quicksprout – Like Priceonomics, Quicksprout was first a popular blog in the marketing space and according to a blog post from cofounder Hiten Shah, they'll be launching a SaaS product to help people produce better content in the near future.
  • Moz Content – One of the quietest launches we've done at Moz was in November of last year when we announced Moz Content, which provides audits, ongoing tracking, analytics, and competitive research tools for content marketers. It's hard for me to be objective, but I think Moz Content has a good chance of growing substantially this year.
My prediction is that one or more of these will gain significant market share of 2,000 or more new, paying users in 2016.

#10: The "big" trends for 2016: Wearables, VR, smart home, and Internet of Things will have almost no impact on the world of web marketing (yet)

I've been known to be a skeptic about a lot of supposed "game-changers" in tech and marketing, and this year, I'm sticking to that archetype. I know that all the tech publications are basically saying thesame thing — it's the year of wearables, virtual reality, smart home, Internet of Things, and personal transport.
I won't argue against any of those (though I'm not sure VR will be as mainstream by next Christmas as the tech blogosphere is prognosticating), but I will argue that none of them will have a big impact on how we do web marketing. There may be a few niche opportunities, but my prediction is that none of these new trends will create marketing/advertising platforms or potential in the $50mm+ range. If ads on refrigerators or in VR helmets or if new social media platforms on wearables attract that kind of business, I'll lose my 2 points.
BTW — I'm obsessed with my Fitbit; I just really hope no one can market to me through it.

There we have it! My ten predictions for 2016 are set in stone (errr... an easily editable blog post, but I'd never abuse your trust or Google's cached snapshot like that). I look forward to your comments, critiques, and any predictions you'd like to share here. We'll see how it plays out next January.

Source :- https://moz.com/