Is automated copywriting the next big innovation in email?

Perhaps — with some caveats.

Considering the rapid pace of innovation in communications broadly, the email sector has remained surprisingly little-altered over the past 25 years.  But maybe that’s about to change.

We’re now seeing developers building tools that can create email copy using text-generation technology.  This past June, artificial intelligence research lab OpenAI unveiled a language model known as GPT-3, which has quickly led to several automated writing tools being developed.

Just what is GPT-3?  Here’s a definition according to The Great Book of Wikipedia:

“Generative Pre-trained Transformer 3 is an autoregressive language model that uses deep learning to produce human-like text. It is the third-generation language prediction model in the GPT-n series created by OpenAI, a for-profit San Francisco-based artificial intelligence research laboratory.”

In a nutshell, automated writing tools built on GPT-3 send bits of keyword text provided by an author – otherwise known as “prompts – to OpenAI’s cloud service, which instantaneously sends back full-flowing text that’s deemed appropriate and accurate based on the statistical patterns it recognizes in the online text.

Even though GPT-3 technology accesses a vast information bank of training data comprising nearly 500 billion tokens in cyberspace to “derive” the copy, there’s always the possibility that the results could end up like the early attempts at automated language translation at the start of the 21st century – garbled and awkward.  However, with more AI “practice” and crowdsourced feedback, we’ve seen an established service like Google Translate deliver excellent translations for most commonly used languages like German, French, Spanish, Chinese and Japanese. 

Languages such as Hungarian, Turkish and Lithuanian are another matter – presumably with more seasoning needed to get those more esoteric tongues in ship-shape for AI translation.

Facebook, which has developed its own “walled garden” automated translation app, appears to be lagging Google considerably in the quality of its output – even when working in the most common languages like translating from French to English.

For now, the most practical applications of the GPT-3 language model look to be in the realm of business email writing, rather than for long-form business thought-pieces or most forms of creative writing.  In email communications, the author can jot down three or four key points and let the writing application do the rest. In this manner, instead of having to craft a memo completely from scratch, authors can provide key snippets — then take a moment or two to edit the proffered text before sending the email on to its intended recipients.

For those of us who write for a living, such a procedure might not seem particularly attractive. But for the many people who dislike the task of writing business communications — or find it laborious and too time-consuming — the new AI-powered writing may well be a welcome tool.

OpenAI’s automated writing service is on the pricey side today, but we can expect that it won’t take long before costs borne by end-users start to drop precipitously — no doubt due to the proliferation of free services subsidized by the same monetization model that now supports Google Maps and Google Translate.   

And this brings up a question that people should start to think about sooner rather than later:  Who will own the copyrights to the automated texts generated in this manner? 

For the many people who will undoubtedly choose to use freeware, the freeware’s terms of use may explicitly override the provisions of most copyright laws that vest ownership with the party who hires the ghostwriter.  In other words, if someone wishes to keep the copyright, then he or she has to pay for the writing service; otherwise, the service retains the copyright.

It’s only a matter of time before the leading purveyors seek to leverage their ownership of the freeware and the licenses they grant to use it – thereby giving them the ability to promote or censor whatever information they please.

The social acceptability of this medium could also be eroded when the volume of ghostwritten email masquerading as personalized communications begins to overwhelm people’s inboxes. At some point, email recipients will come to realize that any message that doesn’t include a disclaimer such as “I am the author; please disregard all spelling and/or grammatical errors,” can be marked as spam and routed automatically to the recipient’s junk email folder.  In such an environment where we’ll have a perceived quality demarcation between “real” and “manufactured” writing, we may find ourselves in the same place as we are today with tweets — that is, weighing if they are the work of humans or bots and judging their worth accordingly.

In other words, the new “next thing” in email communications won’t be happening without its share of issues and controversies – along with more than a little disruption.  It will be quite interesting to see how it all unfolds in the coming years.

What are your thoughts on the role of AI in writing?  Is the technology poised to become mainstream quickly, or will it remain more of a curiosity for a good while longer?  Please share your thoughts with other readers.

Let the AP Stylebook explain it all to you …

For many people – not just journalists but also business and tech writers – the Associated Press’ AP Stylebook is something of a Bible when it comes to adhering to proper presentation of the written English language.

There are other style guides out there – FranklinCovey is another popular resource – but the AP Stylebook has been the “go-to source” for so many decades, it’s hard not to think of it as the ultimate arbiter of what’s considered “proper” in written communications.

This vaunted reputation is why so many people take notice whenever new revisions to the AP Stylebook are released.  The most recent ones, published within the past few months – all 991 of them – are in some cases eyebrow-raising.

Reading through them, it appears that the Associated Press has gone all-in on “keeping up with changing times” by tackling a wide range of sometimes-provocative topics.  Here are some examples:

  • AP is weighing in on environmental terminology, contending that “climate change” is a more accurate scientific term than “global warming.”
  • References to people with disabilities should now exclude descriptions that connote pity, such as “afflicted with,” “battling” or “suffers from.” Moreover, referring to a disability as a “handicap” is no longer appropriate.
  • The word “mistress” should no longer be used to describe a woman involved in a relationship with a married man (although rendering judgments about “paramour” or “kept man,” common references to the male version of the same, are noticeably absent from the guidelines).
  • On ethnic/racial topics, the term Black is now preferred over “African American.” What’s more, the term should always be capitalized whenever used.  (No similar pronouncement is made about capitalizing the word “white” in the same context.)
  • When it comes to age demographics, “senior citizen” and “elderly” are no longer appropriate terminology. Instead, the reference should be to “older adult” or “older person.”

But the most extensive new guidelines in the updated AP Stylebook are the 11 paragraphs and 22 specific examples presented under the heading “gender-neutral language.”

Banished are terms like “businessman,” “manpower,” “man-made,” “salesman” and “mankind.”  In their place are “businessperson,” “crews,” “human-made,” “salesperson” and “humanity.”

“Freshman” is now also frowned upon – but at least the replacement term isn’t the awkward-sounding “freshperson,” but rather “first-year student.”

While AP is to be commended for attempting to keep current on cultural changes, let’s hope that its efforts don’t devolve into the level of parody; some may think that it already has.

But I do have one question:  When will AP finally acknowledge that the entire world is using U.S. Postal Service abbreviations for state names – and has been doing so for well-nigh decades now?

These days, it seems that nobody other than AP is writing “Ore.” for “OR,” to cite just one example among 50.  Tenaciously holding on to outmoded state abbreviations — when no one else is doing so — seems almost like a nervous tic on AP’s part.  (Or is “nervous tic” yet another descriptor we can no longer use?)

What are your thoughts about the newest AP Stylebook guidelines?  Right on the money … or blunt overkill?  Please share your views with other readers here.

The New World of PR

Companies work to find their place in the changing ecosystem — some more effectively than others.

For those of us who have been active  in the marketing communications industry over the past few decades, there’s been a sea change in how the industry operates — not least in the realm of PR and media relations.

One of the underlying reasons for this change is the dramatic shift that’s happened in the field of journalism. Traditional media companies which have long relied on professional reporters and editorial contributors have been dealing with a range of existential threats.  Print circulation has sagged while audiences have fragmented over a plethora of digital content publishers — most of which offer news and information free of charge.

At the same time, publishers’ revenues from advertising have plummeted as the media inventory has expanded to encompass the new digital content publishers.  The bottom-line impact of these twin developments is that it has become much more difficult for traditional media companies to employ the same number of staff reporters; indeed, many publishers have shrunk their newsrooms while relying increasingly on independent contributors and freelancers to fill in the gaps.

But the situation gets even more complicated thanks to the evolution of digital media and the explosive growth of self-publishing platforms. The reality is that there’s a new class of authors who are increasingly publishing from their own platforms, without being involved with any of the major media outlets.

In such a world, the notion of PR departments simply keeping in close touch with a limited number of key journalists as the most effective way of gaining earned media coverage seems almost quaint.

And it gets even more problematic when considering how much easier it is for businesses to prepare and disseminate PR news. At their best, PR pitches rely on the same tools as marketing in general: profiling the audience; personalizing the news pitch, and so forth.

The problem is, according to the U.S. Bureau of Census, there are now more than six PR pros for every journalist. This means that more PR news releases than ever are hitting the inboxes of far fewer journalists and reporters.

Is it any wonder that PR news released by companies is so often being ignored?

According to a recent survey of ~1,000 journalists by PR Newswire, the following aspects of PR pitches are the most annoying to reporters and journalists:

  • Too much overt “marketing” in the pitch
  • Lack of relevant or useful content
  • Unclear or misleading subject lines on e-emails
  • Insufficient news detail

On the other hand, some aspects help in a PR pitch, including:

  • High-resolution photography
  • Video clips
  • Infographics

In today’s PR landscape, obtaining earned media is more difficult than ever. These days, not only do you need a great story to tell, you need to craft the perfect narrative. And even then, you might never get the news covered by a so-called “Tier 1” publication.

But missing out on Tier 1 coverage isn’t necessarily the kiss of death. Sometimes the lower tier represents the best targeted audience to receive news from companies. Moreover, by employing low-cost self-publishing tools, a decent social media strategy plus some basic search engine optimization, it’s actually possible to build an audience and garner as many well-targeted readers as those elusive Tier 1 pubs might be able to deliver.

In the new world of PR, the “tried and true” avenues to earned media coverage aren’t getting the job done.  But there are more routes than ever to get the news out instead of having to channel your efforts to go through the gate-keepers of yore.

Yet another challenge for publishers: Subscription fatigue.

As if it isn’t enough that newspaper and magazine publishers have to compete with an ever-widening array of information providers, in the effort to migrate subscription revenues from print to digital these publishers are squaring off against an additional challenge: subscription fatigue.

Not every consumer is opposed to paying for media, but with so many fee-based streaming services now being offered — including the ones by powerhouses like Netflix and Spotify — trying to get people to focus on “yet another” resource is proving difficult.

Moreover, when it comes to news information sources it’s even more challenging. According to a recent Digital News Report prepared by Reuters Institute, when given a choice between paying for news or paying for a video streaming service, only ~12% of respondents in the Reuters survey stated that they would pick the news resource.

It seems that with so many time demands on people’s online activities, fewer are willing to pay for access to information that they don’t wish to commit to consuming on a regular basis. Unlike most of the entertainment streaming services, news stories are often available from free sources whenever a consumer might choose to access such news stories. Those alternative news sources may not be as comprehensive, but i’s a tradeoff many consumers appear willing to make.

This is hurting everyone in the news segment, including local newspapers in smaller markets which have faced a major falloff in print advertising revenues.

Underscoring this dynamic, more than 1,800 newspapers in the United States have closed their doors in the past 15 years. Today, fewer than half of the country’s counties have even one newspaper within their borders. This fallout is affecting the availability of some news information, as local media have a history of covering stories that aren’t covered elsewhere.

But it’s yet more collateral damage in the sea change that’s upended the world of newspapers and periodicals in recent years.

More findings from the Reuters Institute report can be accessed here.

Like synthetic fabrics, synthetic media has its good and bad attributes.

Decades ago, people had a choice of cloth fibers like cotton, wool and silk. Each of these natural cloths had positive attributes … as well as negative ones, too.

Cotton is comfortable to wear, but wrinkles when washed. Wool is great for the cold weather months, but needs to be dry-cleaned.  Too, moths and other insects love to burrow their way through woolen clothing, making many an item made from wool ready for the trash far too soon.

Silk? It has all the detriments of cotton and wool without any of the positives — except that it looks rich and expensive if one wishes to put on airs or otherwise “make a statement.”

Beginning in the 1940s, polyesters and other synthetic fibers were introduced, giving rise to all sorts of new clothing items that touted a variety of positive attributes: They washed up fine, didn’t need ironing, and kept their shape over time.

Never mind the fact that the clothing didn’t breathe, and made more than a few people stink to the heavens after wearing a synthetic cloth shirt for barely an hour on a hot summer day.

Along these same lines, today we have synthetic media. It’s essentially how people and machines are collaborating to create media that is algorithmically created (or modified).

In its earliest incarnations, synthetic media was a blend of “real” and “faux” components. Think of a newscast with your favorite, very real anchor person … but the background, screens and graphics are computer-generated.

But things have gone much further than that in recent times. Text, photography and videos are being created by software with such precision and seeming authenticity that it’s nearly impossible to determine what content is “real” versus what has been “synthesized.”

On the plus side, content can be automatically translated and delivered in multiple languages to different audiences spanning the world, bringing more news and information to more people simultaneously. But what if the avatar (host) could be customized to be more “familiar” to different audiences — and therefore more engaging and believable to them?

There’s a flipside to all of this innovation. So-called “deepfakes” (a recent term that took no time at all to be added to the major dictionary databases) harness digital technology to superimpose faces onto video clips in ways that are so realistic, they appear to be totally authentic.

Considering the advances in the technology, one can only imagine the plethora of “news” items that will be unleashed into cyberspace and on social media platforms in the coming months and years. Most likely, they’ll have the effect of making more than a few people suspicious of all news and information — regardless of the source.

Which brings us back to synthetic fabrics. They’re with us and always will be; there’s no turning back from them.  But people have learned how to use them for what makes sense, and eschew the rest.  We need to figure out how to do the same with synthetic media.

Company e-newsletters: Much ado about … what? (Part 2)

This post is a continuation of a topic I wrote about several days ago. That column focused on the (lack of) reader engagement with customer e-newsletters and what may be the root causes of it.

This follow-up post focuses on what marketers can do to improve their newsletters’ worth to readers. It boils down to addressing four main issues:

Too much e-newsletter content is “full of it” – People don’t want to read about how great the company is or other navel gazing-type content that’s completely company-focused.  Instead, offer soft-sell (or no-sell) content that’s truly of value.  Simply ask yourself, “If I weren’t an employee of this company, would I care at all about this topic?”  This exercise applies equally to B2B and B2C newsletters.

Tired writing – There’s nothing more tiresome than a newsletter article that’s filled with corporate-speak or comes across as a patchwork of language from multiple sources.  But this happens all too often.  Sometimes it’s because the writer is overworked and hasn’t had sufficient time to source the article and create a compelling narrative.  Perhaps the author is a non-writer.  Often, it’s simply that the people inside the company love how the copy reads – tin ear or not.  Regardless of the topic of your story, newsletter copy should have personality, and it needs to move.  Otherwise, it’s your reader who’s going to move on.

Gaining an audience – Too many newsletters are playing to an empty house, whether it’s because of an opt-in audience that doesn’t care about you anymore, or from a total lack of visibility in search results or on social media.  Build circulation through in-house databases, optimizing copy to draw in new readers via SEO, and promoting article content through social posts.  Again, these prescriptions work for both consumer and business marketing, although the individual tactics may differ somewhat.

Neglect – It happens way too often:  An e-newsletter initiative begins with great fanfare, but it doesn’t take long for the novelty to wear off.  What starts out as a bi-weekly turns into a monthly or a quarterly, with gaps in between.  Eventually the only thing “regular” about it is its irregularity.  It’s surprising how many corporate websites show links to archived newsletters all the way up to 2016 or 2017 — but then nothing more recent than that.  And we all know what that means …

Wrapping it all up, it’s worth asking this basic question every once in a while: “Is our newsletter any good?” The answer should be unmistakable — if you read your content with a completely open mind.

If you’re involved in your company’s e-newsletter initiatives, do you have any additional insights about what makes for a successful program? Please share them with other readers here.

 

Company e-newsletters: Much ado about … what?

One of my clients is a multinational manufacturing firm that has published its own “glossy” company magazine for years now. The multi-page periodical is published several times a year, in several regional editions including one for the North American market.

It’s a magazine that’s full of interesting customer “case histories” accompanied by large, eye-catching photos. The stories are well-written and sufficiently “breezy” in character to read quickly and without strenuous effort.  The North American edition is direct-mailed to a sizable target audience of mid-five figures.

And I wonder how many people actually read it.

The reason for my suspicion stems from the time we were asked to produce a survey asking about readers’ topic preferences for the magazine. The questionnaire was bound into one of the North American issues, including a postage-paid return envelope.  The survey was simple and brief (tick-boxes with no open-ended questions).  And there was an incentive offered to participate.

In short, it was the kind of survey that anyone who engaged with the publication even marginally would find worthwhile and easy to complete.

… Except that (practically) no one did so.

The unavoidable conclusion: people were so unengaged with the publication that they weren’t even opening the magazine to discover that there was a survey to fill out.

In the world of company e-mail newsletters, is the same dynamic is at work? One might think not.  After all, readers must opt-in to receive them – suggesting that their engagement level would tend to be higher.

Well … no.

A just-published study titled How Audiences View Content Marketing, finds that company e-newsletters are just as “disengaging” as the printed pieces of yesteryear.

The study’s results are based on a survey conducted by digital web design firm Blue Fountain Media. Among the findings outlined in the report are these interesting nuggets:

  • One in five respondents completely ignore the e-newsletters they receive, while more than half scan headlines before deciding to read anything.
  • Two-thirds of respondents admitted that the main reason for opting in to receive e-newsletters is to take advantage of special offers or discounts, while only around 20% expressed any interest at all in receiving information about the company.
  • More than half of respondents (~52%) feel that newsletter content is too “commercial” (as in “too sales-y”). Other complaints are that the e-newsletters are “too long” (~21%) or “boring” (~19%).

Even more alarming is this finding: Approximately one-third of the respondents felt that e-newsletter content is so lame, it actually leads them to question using the product or service.

That seems like marketing going in reverse!

What Blue Fountain has uncovered may be indicative of another challenge as well:  the diminishing allure of content marketing. Over time, readers have become cautious about accepting online content as the gospel truth; this research pegs it at two-thirds of respondents feeling this way.

At the same time, only about one-third of the respondents think that they can distinguish well between fact-based content versus content with an “agenda” behind it. And therein lies the basis for suspicion or distrust.

On the plus side, the research found that readers are more apt to engage with video content, so that may be a way for e-newsletters to fight back in the battle for relevance.  But it still seems a pretty tall order.

I address the topic of company e-newsletters in a second blog post to follow.  Stay tuned …

Are 5-star online reviews really the best ones?

It would seem that the more top ratings a company or product can receive in online reviews, the better it would be for their business.

As it turns out, this isn’t exactly the case. A recent national study has concluded that businesses earning star-ratings averaging between 3.5 and 4.5 on a five-point scale earn more revenues annually than those with other ratings – higher or lower.

And even more surprising, top-rated businesses with five stars actually earn less in revenues than those whose customer ratings are two stars or lower.

What’s going on here?

It would seem that five-star ratings are considered “too good to be true.”  Seeing them, people tend to think something’s fishy about how the ratings can be so high. And if there’s something worse than getting low ratings, it’s the feeling that the ratings a company has earned aren’t “genuine.”

The analysis, conducted recently by small business SaaS supplier Womply, sought to study the correlation between online customer reviews and company revenues, and in doing so it looked at data from a large number of U.S. small businesses.

The more than 200,000 businesses studied had an average annual revenues of around $300,000. The Womply research spanned diverse industries and markets including restaurants, auto shops, retailers, medical and dental offices, hair and nail salons, etc.

While the ratings dynamics may be surprising, another Womply finding reinforces the intuitive view that attracting more reviews online is better than attracting fewer ones.

The businesses studied by Womply averaged ~82 total reviews across multiple online review sites. But for those businesses attracting more than the average number of reviews, they earned ~54% more in annual revenues than the average.  And for those with 200 reviews or more, the average annual revenues were nearly double the average revenue figure.

The propensity for companies to respond to reviews appears to boost revenue performance as well. The Womply study found that businesses that fail to interact with their customers’ reviews earn lower revenue on balance – as much as 10% less than their counterparts.

The key takeaway points from the Womply research appear to be:

  • Too many top-rating reviews risk making a company’s reputation appear less genuine, actually repelling business rather than attracting it.
  • To improve revenues, businesses should encourage their customers to post reviews online.
  • To improve revenues, businesses should engage with reviewers by responding to their comments, addressing concerns, and expressing gratitude for praise.
  • People feel more affinity with companies that acknowledge their customers and treat them like they care. It’s basically the Golden Rule in practice.

What are your thoughts? Do the findings surprise you?  Please share your perspectives with other readers.

Do consumers really understand “native advertising” labeling?

There’s no question that “native advertising” – paid editorial content – has become a popular “go-to” marketing tactic. After all, it’s based on the time-tested notion that people don’t like advertising, and they’re more likely to pay attention to information that looks more like a news article than an ad.

Back in the days of print-only media, paid editorial placements were often labeled as “advertorials.” But these days we’re seeing a plethora of ways to label them – whether identified as “sponsored content,” “paid posts,” or using some kind of lead-in descriptor such as “presented by …”

Behind all of the verbal gymnastics is the notion that people may not easily distinguish native advertising from true editorial if the identification can be kept somewhat euphemistic. At the same time, the verbal “sleight of hand” raises concerns about the obfuscation that seems to be going on.

These dynamics have been tested. One such test, conducted several years ago by ad tech company TripleLift, used biometric eye-tracking to see how people would view the same piece of native advertising, that carries different disclosure labeling.

The results were revealing. Here are the percentages of participants who saw each ad, based on how the content was labeled:

  • Presented by” labeling: ~39% saw the content
  • “Sponsored by” labeling: ~29%
  • “Promoted by” labeling: ~26%
  • “Brought to you by” labeling: ~24%
  • “Advertisement” labeling: ~23%

Notice that the content that was labeled “advertisement” was noticed the least often. This provides yet more confirmation that people ignore ads.  When advertisers used softer/fuzzier terms like “presented by” and “sponsored by,” they achieved a bigger lift in the content being noticed.

It comes as little surprise that those same “presented by” and “sponsored by” labels are also the most potentially confusing to people regarding whether the item is paid content. And when people find out the truth, they tend to feel deceived.

Members of the Association of National Advertisers look at it the same way. In an ANA survey of its members conducted several years ago, two-thirds of the respondents agreed that there should be “clear disclosure” of native ads – even if there’s a lack of consensus regarding who should be responsible for the labeling or what constitutes “clear” disclosure.

Asked which labeling describes native ad disclosure “very well,” here’s what the ANA survey found:

  • “Advertisement”: 62% say this labeling describes native ad placements “very well”
  • “Paid content”: 37%
  • “Paid posts”: 34%
  • “Sponsored by”: 31%
  • “Native advertising”: 12%
  • “Presented by”: 11%
  • “Promoted by”: 11%
  • “Branded content”: 8%
  • “Featured partner”: 8%

Considering that the findings are all over the map, it would be nice if a universal method of disclosure could be devised. But the language that’s agreed upon shouldn’t scare away readers, since in so many cases native advertising isn’t directly pitching a product or service.  Labeling such content “advertising” would be as much of a misnomer as failing to divulge the company paying for the placement.

My personal preference for adopting consistent labeling language among the options above would be “Sponsored by …”  What’s yours?