By Jenna Tiffany on Wednesday, 20 March 2019
Category: Email Strategy

Round Table Discussion: How Will Voice Tech Change Email Marketing?

The Following Email Marketing Round Table Discussion is based on Jenna Tiffany's article: Verbalising Email – How Will Voice Tech Change Email Marketing? 

Jenna Tiffany: So let's start with, do we think it will change email marketing? And if yes, how? 

Maybe you use voice tech already to read your emails - what's the experience like? Have we seen any brands doing this well or badly? 

The exciting part for me is that this should bring accessibility back to the forefront. Something that is regularly forgotten about or left until the last thing to do on the checklist before sending an email campaign. 

I'm looking forward to what I hope will be an interesting discussion this week and finding out what you think. 

Adrian Nabarro: This is a great topic and I’m glad we’re going to discuss it here.

As a fairly early adopter of voice tech (I’ve had most versions of Echo, HomePods in every room and a big Siri user), I’m quite aware of the tech's many strengths, but I’m also acutely aware of its limitations. Frankly, I love the tech, but really question its use for email, especially marketing. The big caveat here is Google. I’m not too familiar with their offering, so perhaps they may surprise us...

So far, neither Siri or Alexa handle email particularly well, so I’d be surprised if very many consumers would be willing to suffer through their limitations to regularly consume their emails.

If we ignore the low market share, from a marketers perspective, apart from the focus on accessibility which you’ve already mentioned, I think there’s limited value that they’ll be able to realise via that medium; Apple's ecosystem will not allow actions outside of the message and while Alexa may do in the future, via Skills, this will require a lot of development and even so, is still fairly limited (unless you’re selling products directly on Amazon!).

I personally feel this is just the latest topic the email community is getting excited about before realising it’s not that practical and moving on to the next big thing. That said, I can’t wait to hear other perspectives! I’m sure I’m missing something.

David Baker: I think it's too early to look at the impact of voice tech on email marketing, which I typically bundle with AI Assistants, AI Bots.    Email is largely a broadcast medium vs. a marketing dialog medium, IMHO. 

What I do see as a derivative of this voice market shift is, its impact on customer service, site experiences and even future POS kiosks that are interactive.  Wil the future website simply be a voice prompt?   I think this will change in the next 10 years dramatically from a UX and CRM perspective when AI will prove to have a higher quality of interaction than live call center staff.  The natural voice algorithms are progressing so fast now, its almost humanistic.      But what is interesting to me is the amount of data that will explode search phrases, terms and can be mined for patternistic trends by cohort.. that will be valuable in designing outbound experiences, how you think about how you communicate in written form.  

I think this uproots traditional CRM and customer/product support to some extent.  I wrote about this a while ago  (Future consequences of voice/AI)   https://salient.blog/2018/09/14/ai-voice-future-and-consequences/ , might be an interesting read.

Tim Watson: I tend to agree.

I don’t expect a large number of people to get their emails read aloud. The current user experience is terrible, even if you did want it. The exception being the segment Jenna highlighted for accessibility.

Websites will move to having a personal shopping helper that you can chat with via voice or message. Turn up to a website on your mobile and either tell the chatbot or the voice search assistant you want ‘red shirt’ and let it take you through the choices, show you options, ask questions to better define your wish.

That could flip seamlessly between message chat and voice for the duration of the shopping visit -  and all that insight used to improve email content.

Back to email. Perhaps using voice to retrieve information from your email archive. “Find me a recent brand X offer in my email inbox”, “Unsubscribe from Y”, “Check the early bird sale end date for Z”.

With Gmail bundles Google are paving the way to get more structured shopping information in email content, making voice search of the inbox easier. Read my blog as to why this Gmail innovation is self-serving for Google.

Jordie van Rijn: 

Related to voice tech for email I think two of the elements in the recent research Tim and me
can add some perspective

We asked which email innovations people (consumers) are most looking forward to.

https://www.emailmonday.com/email-addiction-email-marketing-research/

31% percent says they would be interested in having long emails automatically summarized.
only 13% said have a computer voice assistant read out the emails.

So we might have an “OK Google, but make it short” on our hands.  

It could also be a “faster horses” type of consumer opinion, we will have to see.

But clearly not everyone is ready to talk-shop™.

John Thies

I personally think marketers need to start thinking about this sooner than later. Let’s look at some statistics:

Voice tech is changing how consumers interact with the world around them. We are now immersed in the technology. It’s in our cars, homes, cell phones. Millennials are the King of multitasking and this plays right into that skillset. We can send text messages, start a phone call, create timers/reminders and now read email when involved other mundane activities like doing the dishes.

Google will soon release their version of reading and replying to emails. If you ask your Google Assistant to read your Gmail it will respond: “I don’t know how to do that yet”. The key word there is yet. Given the Gmail has over 1 billion mailboxes and the number of devices that provide access to Google Assistant, marketers have to start thinking about how they are going to engage with subscribers through this medium. If not, they are going to miss out on a huge opportunity.

I have written a couple of articles about this topic and how this will eventually lead to Auditory-Call-to-Actions (ACTA’s)

https://www.emailonacid.com/blog/article/industry-news/the-evolution-of-email-predictions-for-the-future/

https://www.emailonacid.com/blog/article/industry-news/alexa-and-email-how-to-develop-email-for-amazons-virtual-assistant/

Loren McDonald: 

I think the value of AI with email will be on its reading, processing and recommending actions to you the user and that the value of voice interaction is as the interface to respond more than reading our emails to us.

What if at the end of the day you asked the Google Assistant to review your emails and it responded something like, "Loren you received 97 emails today ..

- 12 from Only Influencers and you opened 5 and replied to 3 of them

- You received bills from Verizon and PG&E, but you did not take an action

- Today is the last day to take advantage of Delta's flash sale for European fights

- You booked a car service with Wingz, would you like to forward that email to your TripIt account?

- Your wife's birthday is next week and we noticed a Tiffany email with a 20% off sale on necklaces.

- You received an email from your daughter Erin but didn't open it, would you like me to read it or summarize it?

- You received an email from Patz & Hall winery that your club shipment would arrive via UPS tomorrow.

You get the idea. To me the real value in the AI is it learning over time what I interact with, don't read, what is important, what requires an action, what it things I might have missed but is of interest to me, etc. And then voice interaction just becomes how I interface with the AI and how it communicates what is important to me.

Secondly, putting aside how it would be done technically, might something like pre-header text evolve into some type of standard in the future so that AI assistants would read that text and stop - we could in essence design preheader or similar text primarily to be read via a voice assistant. The value increasingly is not while sitting on the commute train, but rather while driving in the car, walking the dog, working out, etc.

Brian Gruidl: I REALLY hope that people don’t start searching their inbox/archive folders using voice search in the office. 

Where I think voice technology can be really beneficial for the email channel is the additional data that can be collected via the new medium.  This data would likely have a different weight than say, site search or browse data, and could be used to increase targeting capabilities.

I think this also brings up the business opportunity to expand search classifiers to include voice data…so those looking for a new startup, just make sure you add me to the board ;)

Elliot RossThe tech will get better, it’s already good. “Hey Siri, read my email” is serviceable already.

Driving, is an obvious use case.

It’s yet another reason why all-image emails suck, (PSA, so don’t do them)

Jacob FanningI can picture people checking their inboxes via search and making decisions to read them later/archive them or delete them based on the sender and subject line (and potentially preheader), depending on what information Alexa, etc. can retrieve from them. 

Dela QuistElliot, 

I completely get why you don't like all image emails, but there are one or 2 sound business reasons for sending them if not every time occasionally. I would be interested if anyone on the list has found situations or segments where they perform better. 

John Theis: +1

This is also another use case to NOT use a “no-reply” email address. Given that the auditory clients do not tell you the senders email address, it would be a bad experience for the subscriber to reply to your email and either get no response or receive a bounced return.

I see auditory clients in a similar vein as text-only versions of an email. We know some subscribers read emails in this format, we can’t track it, but it’s still a best practice to include when sending an email campaign. The same goes for auditory clients. It’s our job as email marketers to put our best foot forward in whatever medium our subscribers want to engage with our emails. It’s not all about conversion, but also retention and branding.

Matthew Dunn

My frank take - voice tech without substantial AI is just a novelty edge case as far as email marketing,

and there are too many negatives for widespread use.

Then there's biology. In a nutshell, eyes are smart and ears are comparatively dumb.*

If marketing emails get demoted to the audio plane, they're even less likely to get actual attention.

Fun discussion, but I don't see factors to drive take-up.

David Baker: Lets not throw all the fish into the boiled ocean on this matter..

Out of home digital communications/connections is on the rise, so while I don't see users using voice tech to read out your promotional tab, I do see use cases today that is very useful for those on the run who need to process work emails, that are far more functional in nature and directive than marketing language offers. 

Voice tech is here and its not going away.  Yes its been around since dragon speech back in the early 90's, but its so much different.  Same reasons siri, alexa are so popular, voice will continue to be mainstream in most of what you'll do online and again, when convenience is the priority, people will adapt to voice controls (even reading emails).

David  (if you haven't tried voicera's new meetings AI took, its fun..   "Eva"   there's so many sides to the voice story.  Will be amazing to seee what happens in the next 5 years on this front.

Justin Khoo:  Speaking about voice data, once a significant portion of users start using voice assistants, I can foresee Google or Apple adding capabilities to embed audio files or body parts (like the Apple Watch watch/html) in email that can be read in-lieu of text. This way senders can customize the message specifically for the medium.

Also… audible logos ala BIMI for audio.

John Thies: Matthew,

Interesting points. However, I think consumers are smart enough to not have their email inadvertently shared.

From a smart speaker perspective, yes, it “could” be shared. However, when you look at that stats I mentioned previously, SIRI/Google Assistant/Alexa, they all are apps (or provide apps) on a device – meaning headphones/mic could be used to interact with an email from an auditory perspective. Pretty much everyone in a 1st world country has a mobile device (phone, tablet, etc.) and when you think about it in that perspective and looking at the use-cases (walk, workout, driving, etc.) a vast majority of the interaction would be private.

I also think that from a privacy/etiquette perspective a consumer is not going to be listening to an email for everyone to hear – just like most people don’t talk on the phone with the speaker in a public setting.

Dela Quist: Elliot, 

Here are 2 pretty good reasons in my mind. 1 may surprise you the other perhaps not.

Reducing your “Inactive” file

Emails that are not image only will generate clicks from “non-openers”. We typically find that around 3% (I have heard of as much as 10%) of clicks do not have a corresponding open – call them false inactives. By comparison image only emails generate no clicks without a corresponding open because subscribers have to download the open pixel with the image. Forcing your subscribers to download images can reduce your inactive file by 3% or more! Bearing in mind the average “reactivation” campaign gets a 3% open rate this is not an insignificant number. In our view at Alchemy Worx is worth doing at least once a quarter.

Delivering Visual Excellence - High value “sexy” or aspirational brands or products - eye candy if you like.

For some brands and markets, think high end jewellery, cars, fashion or beautiful holiday destinations, imagery is integral to the sell, I have noticed those folks would rather you saw NOTHING than see an email optimized to convey information without you “having” to see the image. Alt tag saying Red Ferrari or Blue Sweater or a mandatory 20% text for readability or deliverability just doesn’t cut it for those brands.

Add to this the fact many of those brands have a LOT of in-house design resource, who for the most part email is a tiny part of their work. So, the ROI on email best practice just doesn’t justify the additional effort required.

The other thing to bear in mind is brands that send image only email and the marketers that work for them are not all foolish and they must at some level be ignoring this particular best practice, because the data shows that image only email is what their customers either prefer or respond to.

For all these reasons - I don't think voice tech will stop the practice

Marc Lantrok:  From personal experience I  have already adopted voice (as per Elliot) in my commute to work and find it works great on text heavy emails (internal correspondence), less so from emails from brands due to the way they are configured. This fairly new habit can also be linked to lifestyle, being an adopter of smart home accessories, it’s only natural that you can use voice instructions for many things, ie lights, home temperature, TV, car, Alexa integrated earbuds etc. If you look at our youngsters today, they also have a reliance to talk into apps such as Youtube via the search bar as opposed to typing which is what you expect in their early years. At Hilton, we’re looking at rolling out the connected room, a future state could see your servicing requirements done by the in room Alexa type of device for guests that would want that interaction. Technology will need to get better, including issues with heavy accents (think Scottish lads in the elevator video).

So back to the question in hand, I do see voice becoming a serious player in email / push, with everything moving to a contextual relationship with digital. I would however be mindful on the intended customer journey, ie if there was a voice CTA included, can you respond via voice – possibly better placed in a message or push environment. Aspirational / informative or statement content with no applied CTA could be an interesting test especially if you can personalise it. Maybe, using a similar dynamic such as HTML, Plain Text and Rich Text formats, one could piggy back of that or create a 4th format that caters for voice. Tracking / reporting would be essential.

Leave Comments