2017: The year in martech

After so many years of teases, we might as well call it.

In 2017, marketing and ad tech turned the corner toward becoming pseudo-human.

Arguably, we turned the corner long ago, but let’s just say that we’ve been able to make out the actual curve in the last 12 months. It’s nowhere near marketing tools that are artificially human yet, but we don’t have to squint that hard any more to see the pieces falling into place.

Let’s start with the realm we humans have long claimed as exclusively ours: a high level of intelligence.

Built-in intelligence

Last year, it became common for virtually every tool to boast some form of intelligence, ranging from large marketing platforms like Salesforce and Adobe to product lines like Microsoft’s and to specialized tools like influencer platform Linqia. One way to tell that a corner has been turned is that the new tech — in this case AI — doesn’t quite feel like news any more.

Artificial intelligence has already begun to spawn major new marketing/ad product lines, like self-driving cars that will provide new opportunities to deliver marketing messages since none of the occupants will be driving. Or IBM Watson-based cognitive ads that are more like specialized concierges, not just delivery systems for messages. Or “propensity marketplaces” — like an initial auto-based one from identity management provider Infutor — that employ purchase records to predict which car-based products the customer will next want to buy.

Our intelligence wouldn’t be all that useful, of course, if we couldn’t see what was in front of us. In 2017, computer vision — once a rare and limited capability — became much more common.

User-generated posts of visual material can now be visually understood to various degrees by such tools as Salesforce’s Social Studio, social management tool Crimson Hexagon or image platform Pinterest. Gumgum is one of the companies that can display ads over appropriate images by understanding what is in the image, like an ad for cat food over a photo of kittens.

Not long ago, visual logo recognition was the primary capability available to marketers, but nowadays, computers can discern a wide array of objects, scenes and sentiments. New York City-based Uru’s technology, for instance, can scan video imagery to find more than 10,000 specific objects or themes or to decipher text inside images. A research project in the UK has even trained neural nets to determine which rural or urban scenes are most beautiful, possibly forecasting a time when AI performs sophisticated creative direction.

For humans, recognizing another face is a core capability of our vision, making biometrics another example of tech’s encroaching pseudo-humanness. AI firm Sensory, for instance, recently announced a partnership with Fujitsu that could make facial recognition a more common way for customers to log into their banks, possibly the toughest customers for this kind of identity. And marketing/commerce platform SAP Hybris has added facial recognition that can target product recommendations based on a visual determination of a shopper’s gender and age.

Bots and agents, oh my

The ability to discern intent and meaning from free-form text and speech has led to the current wave of bots and intelligent agents.

They haven’t yet passed the Turing Test of becoming indistinguishable from human conversation, but they are in the process of disrupting customer service, loyalty programs and advertising, particularly since their creation is becoming easier. The year 2017 saw a variety of new text bot and voice agent building platforms for non-technical users from such vendors as Sprout Social and PullString, as well as bots with high-end intelligence, such as the IBM Watson-powered bots in LivePerson’s platform.

Intelligent bots and agents are also gaining a foothold as business interfaces, including the bot interface for data provider Knoema, the Alexa interface from marketing/ad data management platform Datorama and Amazon’s new Alexa for Business.

Voice is rapidly becoming another channel for customer journeys, which is how it is now being handled by the Resulticks marketing platform. Voysis Commerce has recently launched a “voice as a service” business so that any commerce provider could take verbal orders.

In fact, public relations firm Walker Sands reported in July that 19 percent of all consumers have used a voice-controlled device to make a purchase in the last 12 months. Among millennials, that figure was 43 percent. Research firm Gartner has predicted that voice interaction will handle about a third of all web browsing within three years.

One instinctive measure of how far around the voice curve we’ve come is that it’s no longer ridiculous to imagine that we might expect to talk to any computing device we encounter, the way Scotty expected in Star Trek IV.

If you want to build a human-like entity, such capabilities as intelligence, vision, reading and understanding of language are certainly central. But that only gets you communication. Another sign that marketing and ads are approaching human qualities is the widespread use of emotional recognition.

Persado announced this year that it can now generate emotionally targeted messages to individual users in addition to segments. Tracking of emotional responses and attention for ads have become standard tools for some ad agencies, thanks to vendors like Tobii and Sticky (who are now the same company).

GDPR, expanded platforms

Beyond the humanization of marketers’ tools and consumer/business interfaces, 2017 was a busy and possibly pivotal year.

For one thing, it’s difficult to envision an internet where the biggest entities rule the US landscape because net neutrality doesn’t exist, but there is extensive protection of individual data by many brands because of the upcoming General Data Protection Regulation (GDPR). But, at year’s end, that scenario is imminent.

Over the past year, we’ve seen the release of a number of GDPR compliance solutions for different parts of the user data flow, from Segment, Evidon and others. Several solutions are particularly inventive and could offer some possibilities for marketing and advertising if cookies and third-party data continue their downward spiral.

UK-based Smartpipe, for instance, has developed a way to target ads using session-based markers that “evaporate” like snowflakes after each use. But each marker can be targeted according to an accompanying customer profile, which exists only within that mobile network and is available for many mobile customers because they have given consent to the network.

Media Trust, which offers compliance and anti-malware monitoring, has set up what it says is the first network of third-party vendors who comply with GDPR. Segment has released a third-party cookieless solution for tracking users between providers in its network, propelled as much by Apple’s initiative in iOS 11 and Safari to battle third-party cookies as by GDPR.

New directions, environments, channels

This past year also saw several brands move in dramatic new directions. For instance, following a series of acquisitions, social media management platform Sprinklr has now recast itself as a customer experience cloud.

MailChimp, which may soon need a new name, is expanding beyond email to become a marketing platform, while content recommendation platform Taboola is busily adding new acquisitions, integrations and partnerships as it evolves into a service encompassing personalization, analytics and advertising.

And Reddit, long a risky environment for marketers, took steps in 2017 to make itself a bit friendlier to ads and to outside-managed social posts.

Speaking of new environments for ads, look outdoors. This past year saw digital-out-of-home (DOOH) becoming more of a member of the programmatic digital ads family, as well as a kind of interactive medium, with campaigns featuring ads that respond to traffic flows, weather data or user prompts.

A digital billboard for an Audi campaign, displaying images based on data feeds

The original King of Media Channels — that is, television — is inching closer to full integration with programmatic online ads. Its over-the-top (OTT) incarnation continues to acquire new targeting and analytical capabilities, such as new services from Tru Optik, Taplytics, ColorTV, Swrve and mParticle.

Speaking of channels, 2017 will be notable for the coming-of-age of a major new one. Augmented Reality (AR) got marketers’ attention after the Pokémon Go phenomenon in 2016, but the launch this year of Apple’s ARKit and Google’s ARCore for Android means that every smartphone now becomes a viewer of a hidden layer of reality. While the jury is still out on whether virtual reality will ever become a mainstream channel that marketers can employ, it seems apparent that AR will.

Finally, no summary of marketing tech in 2017 would be complete without a mention of blockchain. The technology infrastructure behind bitcoin, blockchain is emerging as a way to permanently record transactions and immediately share them among participants, a kind of built-in transparency that digital advertising should welcome.

Marketers may also adopt two other blockchain-supported capabilities: smart contracts, which are self-implementing agreements, and crypto-currency or tokens, used to incentivize loyal customers or vendors. The components of a surrounding ecosystem for blockchain began to emerge this past year, so don’t be surprised if our 2018 annual summary, a year from now, declares that blockchain has — you know —  turned a corner.