Rise of Voice Marketing — Brand Success Metrics / Alexa, Google Assistant
We had an incredible night of 3 talks and 4 panelists at Bots and AI.
Deep industry insiders shared data and trends on voice plus what’s next for monetization, form factors, and a glimpse at a possible future from what’s going on in China’s voice and messaging ecosystem already.
Whether you’re a solo developer or a Fortune 100 product/marketing leader learn what works in voicefirst.
Matt Hartman, Betaworks Ventures : The Third Wave of Audio
Matt calls this era of voice Voice 3.0 — the next soundwave. As a partner on the betaworks venture capital fund Matt invests in brand new companies. Matt’s fund covers companies mostly in social but as of 4 years ago started looking more deeply into voice. Betaworks is a prolific investor in voice. Betaworks was a first investor in: Gimlet Media — HBO for podcasting, Anchor — the easiest way to create a podcast (both were acquired by Spotify earlier this year), Shine — a subscription service for daily positive affirmation. Matt’s VC fund also conducted Voice Camp — a pre-seeding program only focused on voice — that invested in 8 companies building the future of voice. Some examples are: Spoken layer — converts written content into podcasts and Jovo — a technology platform for building voice interfaces and multimodal experiences.
Matt produces a newsletter called Hearing Voices (hearingvoices.xyz) and also built an Alexa skill called Wiffy that tells you the password for wifi in the room you’re in — it has between 15–20,000 users on Alexa.
Matt spoke about the waves of innovation that we have seen in voice over the past few years. Wave 1.0 was all about podcasts and RSS feeds.
Wave 2 was moving to streaming — it was a renaissance era for podcasts producing content like Serial.
The VC’s thesis in 2014 was that:
- Cell phone coverage was getting better
- Battery life was getting longer
- Smart cars from just under 50% of cars in 2014 taking 7 years to go to about 95–99%
Then smart speakers came out. So what’s coming in the next 5 years?
Matt groups innovations in the next wave under 4 different categories:
- Discovery — how do people find out about the Alexa skill you’ve built? Or discover the podcast you have?
- Monetization — not many paid podcasts in the US — how can monetization work for podcasts? Apps like Headspace are interesting because we don’t think of them as podcasts, but people pay $15 a month for that.
- Airpod first technology — what do utilities look like for podcasts? Air pods have the potential to be an Alexa-like interface. What happens when you can walk around with Alexa like devices and have robots in your ear?
- Synthetic voice — there’s a gap between listening to smart speakers read you a whole article versus listening to a human being. Being able to customize voice for you — as synthetic voice gets better you can have voices that have personality
Brett Kinsella, voicebot.ai: Voice Assistants, Smart Speakers and the Rise of a New Digital Channel
Brett has been working in the Voice space since before Alexa and Mobile. While conducting research he realized that good data was not available out there and so ended up creating Voicebot.ai and shared it publicly. Voicebot team attends events and interviews folks in the space and shares that via their podcast and their newsletter for people who don’t get a chance to participate.
Brett’s talk covered 3 areas:
- Evolution of voice
- Research and market data
- What are brands doing with voice
Since 2018, more than a billion devices now access to various voice assistants. Phase 1 of the voice evolution is effectively over. Voice is no longer contained in a device and is now traversing platforms. From a supply standpoint, in phase 1, the people who were building the technology so far were focused on introduction and reach. They are not changing focus to habituation and specialization. This is very well aligned with the consumer lifecycle. Phase 2 will see a proliferation of devices with integrated voice assistants. As the founder of Voicebot.ai, Brett is generally at the front end of these technology life cycles. Some are more successful than others however. Consumer data tells the real story. While it’s important for suppliers to push technology forward, it is important to take a step back and look at what consumers are doing to understand where it might head. AR/VR by example has been the next big thing for a long time, however it is unlike to take off until the consumers start to spark and move the adoption curve up quickly.
While Siri came in and created the first ecosystem for voice, smart speakers came in and took the in a different direction that was more consumer focused. A lot of people don’t realize that the far field microphone is one of the bigger inventions that has happened compared to the voice systems themselves, because it now gives you the room to think of all these new use cases that weren’t possible before. People are buying smart speakers to stream music, ask questions check the weather, set alarms and timers from across the room. People are taking features that existed elsewhere and putting them in a new environment and consumers are loving it. The smart speaker installed base grew 40% year over year from 2018 to 2019. Loup ventures forecasts the global smart speaker sales (except China) to be nearly 300 million units by 2025. Smart speakers are just one area to look at, but they are the leading indicator for what’s happening in voice. Smart speakers are only a third of the interfaces for voice assistants. About 60% more people are using voice assistants on a monthly basis in the car. And nearly twice the number are using them on a smartphone. The ability to speak and get some information back or get a task completed is starting to become something people are habituating around.
Brett then explored the idea of voice first but not voice only. Smart displays (or smart voice first tablets) are growing significantly. People like the idea of having a voice addressable screen.
Half the Google assistant sessions are multi-modal
Over a third of the smart speaker owners say that the moment they buy their first smart speaker they start using the voice assistant more on the phone.
Daily voice users are twice as likely to be owners of smart speakers and also use it on the phone.
The idea of voice is independent of device at this point and consumers are beginning to grasp this. And it is expanding expectations around use cases on the smartphone. Given how different markets evolve, Brett recommends that even if you start with the biggest lead today the story will change in the next few years. So when coming out with products today, they must be supported by at least both the platforms (Amazon Alexa and Google) right out of the gate.
Now that we know voice is everywhere, how can marketers think about voice and what should they be doing to take advantage of this reach?
Voicebot.ai developed a 7-step model to help answer this question:
- Generate awareness — When consumers ask a question about insurance, health or general awareness, the voice assistants shuttle them off the voice apps to help answer it. e.g: Boston Children’s Hospital, Progressive, HQ
- Create engagement — Voice is a way for brands to create a greater experience for superfans after they are done watching a show or using a product. e.g: Westworld, Tide, Mayo Clinic
- Facilitate transactions — e.g. Bank of America’s Erica for banking transactions, Dominoes for ordering pizza, Swedish (hospital) for booking appointments
- Enable distribution — Voice is a better channel for raising availability and importance of audio in our lives and a great way to distribute product and content. e.g: The Washington Post, Cumulus, Entertainment Weekly
- Integrate into product — Embedding voice into their product is another way to leverage voice. Cars are doing this a lot. e.g.: Mercedes Benz, Sonos, Livongo
- Improve operations — Voice assistants can execute processes and operations within consumers daily work life. e.g: Jill (real estate), Voicea, Atomicorp
- Service customers — Consumers can contact customer service through their smart speakers or using voice assistants. eg: Audible, Capital One, Express scripts
Takeaways
1. Don’t be “silent” — brands must have a voice app
2. Monitor consumer expectations
3. Voice first but not voice only
4. Be wary of Voice App bloat
5. Think continuous improvement
Will Hall, RAIN: The Great AI Awakening
Every third article written these days has to do with the Rise of the Machines. The content of these articles might seem laughable and ridiculous to a tech forward group, but within them lie the very real seeds of innovation and transformation disruption. Rain works with Fortune 100 companies around voice and conversation and the collective sentiment is that we don’t know where voice is going and we don’t know where to begin.
There is a principle that Will likes to call Steeper and Deeper that can give us a reliable roadmap for where these technologies are going. While here in the US we rarely talk about a mobile first internet, in China it is a mobile only internet. China’s mobile adoption curve is steeper and deeper (more fully integrated and at a greater scale) than what we might see in the US. Steeper and deeper has radical and cascading implications. When you have a mobile first internet, the internet itself has to change to account for it. Companies like Baidu, Alibaba and Tencent have risen to prominence in China. They are called Super Apps.
To win at voice you have to think in systems.
A super app is a complete ecosystem. Rain thinks of these companies as a thin layer of technology that connects everything to everything else and everybody to everybody else. Even in the US big tech companies like Google, Amazon , Facebook etc are evolving into the US version of super apps. We are seeing a consolidation of users around these nodes. And those nodes are enabled by a mobile first internet.
If you want to know where big tech is going look no further than China. China is bleeding data, everything is digitized. Using Wang Xing as an example, Will explained how the Chinese Super Apps are transforming the internet using system thinking. Wang Xing built Dianping by replicating Silicon Valley in China (Yelp, Reddit, YouTube, Facebook). Yelp tried to retaliate by going to China and beating them at their own game. However, they did not account for Dianping’s ability to “go heavy”. Dianping kept asking the question “what else?” until they built their own version of PayPal, HubSpot, GrubHub all in the same app. When they looked at the data Dianping realized that if you didn’t deliver food to people in 20 minutes or less they wouldn’t use your product. So Dianping built a fleet of smart scooters to deliver food. They went further by adding data sensors into the scooters and identifying the most robust traffic pattern data. They even went as far as to petition the government to build traffic lanes for their food.
In China there is a saying that when you win you don’t declare victory, you declare war (go heavy) and it makes them intensely resilient.
In the end, they won. Yelp exited the market and Dianping’s pre-IPO valuation today is around $55.5 billion. Netflix and Amazon are examples of how going heavy has helped big tech become successful in the US. Every company regardless of vertical is essentially a tech company and tech companies build systems.
Tech companies don’t sell cat food — they build systems!
Will thinks that smart speakers are a rounding error on a larger world that is voice. To win at voice you have to think in this ecosystem sort of approach because voice directly and profoundly affects your mobile, e-commerce and your web presence. An Alexa skill without a system is like a keyboard without a computer.
Will talked about how they used this approach in Rain’s work with Starbucks and Nike. In the case of Starbucks, they developed an NLP and a reconciliation layer to place an order within the app. When they realized that the data in the physical locations didn’t speak to each other, they worked on reconciling datasets to build out a thin layer so anywhere in the world you knew exactly where you were and how much your order costs. It allowed Starbucks to innovate in ways that are meaningful to customers for example being able to order coffee in the car. In the case of Nike, Will spoke about how Rain worked with Nike and R/GA to launch the BB (self lacing) shoe on TV where anyone could shop the game just by asking Google Assistant (Hey Google, Ask Nike).
Will closed with an important lesson from his experiences.
When voice does its job well it becomes the preferred input mechanism
Many thanks to Rucha Gokhale from the BotsAndAI team for putting together this great recap.
