Amazon, Baidu, BMW, Cerence, ecobee, Microsoft, Orange, Salesforce, SFR, Sonos, Spotify, Sound United, Tencent, Verizon and more to promote customer choice by supporting multiple, interoperable voice services on a single device
With multiple, simultaneous wake words, customers can access multiple voice services by simply saying the corresponding wake word – from Alexa and Cortana to Orange’s Djingo, Salesforce’s Einstein, and more
Solutions providers like Intel, MediaTek, NXP and Qualcomm to develop hardware and reference solutions that support multiple wake word engines
Today, Amazon (NASDAQ: AMZN) and leading technology companies announced the Voice Interoperability Initiative, a new program to ensure voice-enabled products provide customers with choice and flexibility through multiple, interoperable voice services. The initiative is built around a shared belief that voice services should work seamlessly alongside one another on a single device, and that voice-enabled products should be designed to support multiple simultaneous wake words. More than 30 companies are supporting the effort, including global brands like Amazon, Baidu, BMW, Bose, Cerence, ecobee, Harman, Logitech, Microsoft, Salesforce, Sonos, Sound United, Sony Audio Group, Spotify and Tencent; telecommunications operators like Free, Orange, SFR and Verizon; hardware solutions providers like Amlogic, InnoMedia, Intel, MediaTek, NXP Semiconductors, Qualcomm Technologies, Inc., SGW Global and Tonly; and systems integrators like CommScope, DiscVision, Libre, Linkplay, MyBox, Sagemcom, StreamUnlimited and Sugr.
“Multiple simultaneous wake words provide the best option for customers,” said Jeff Bezos, Amazon founder and CEO. “Utterance by utterance, customers can choose which voice service will best support a particular interaction. It’s exciting to see these companies come together in pursuit of that vision.”
The Voice Interoperability Initiative is built around four priorities:
Developing voice services that can work seamlessly with others, while protecting the privacy and security of customers
Building voice-enabled devices that promote choice and flexibility through multiple, simultaneous wake words
Releasing technologies and solutions that make it easier to integrate multiple voice services on a single product
Accelerating machine learning and conversational AI research to improve the breadth, quality and interoperability of voice services
Multiple, interoperable voice services deliver choice and flexibility for customers
Companies participating in the Voice Interoperability Initiative will work with one another to ensure customers have the freedom to interact with multiple voice services on a single device. On products that support multiple voice services, the best way to promote customer choice is through multiple simultaneous wake words, so customers can access each service simply by saying the corresponding wake word. Customers get to enjoy the unique skill and capabilities of each service, from Alexa and Cortana to Djingo, Einstein, and any number of emerging voice services.
Companies participating in the initiative – including Amazon, Baidu, BMW, Bose, Cerence, ecobee, Free, Harman, Microsoft, Orange, Salesforce, SFR, Sonos, Sound United, Sony Audio Group, Spotify and Tencent – are committed to adopting a similar technological approach, whether building voice-enabled products or developing voice services and assistants of their own.
“We’re in the midst of an incredible technological shift, in which voice and AI are completely transforming the customer experience,” said Marc Benioff, Chairman and co-CEO at Salesforce. “We look forward to working with Amazon and other industry leaders to make Einstein Voice, the world's leading CRM assistant, accessible on any device."
“We value freedom of choice, empowering listeners to choose what they want to listen to and how they want to control it,” said Patrick Spence, Sonos CEO. “We were the first company to have two voice assistants working concurrently on the same system, a major milestone for the industry. We are committed to a day where we’ll have multiple voice assistants operating simultaneously on the same device, and are working to make that happen as soon as possible.”
“Access to the music and podcasts you love should be simple, regardless of the device you’re on, or the voice assistant you use,” said Gustav Söderström, Chief R&D Officer, Spotify. “We are excited to join the Voice Interoperability Initiative, which will give our listeners a more seamless experience across whichever voice assistant they choose, including the ability to ask for Spotify directly.”
Developers and device makers have a shared commitment to customer trust, and will work together to protect the security and privacy of customers interacting with multiple voice services. Companies participating in the initiative will work to ensure this commitment extends to products that support multiple, simultaneous wake words.
Making multiple, simultaneous wake words more accessible for developers and device makers
Alexa machine learning and speech science technology is designed to support multiple, simultaneous wake words. As a result, any device maker building with the Alexa Voice Service (AVS) can build powerful, differentiated products that feature Alexa alongside other voice services.
Still, device makers interested in supporting multiple, simultaneous wake words often face higher development costs and increased memory load on their devices. To address this, the Voice Interoperability Initiative will also include support from hardware providers like Amlogic, Intel, MediaTek, NXP Semiconductors and Qualcomm Technologies, Inc.; original design manufacturers (ODMs) like InnoMedia, Tonly and SGW Global; and systems integrators like CommScope, DiscVision, Libre, Linkplay, MyBox, Sagemcom, StreamUnlimited and Sugr. As part of the initiative, these companies will develop products and services that make it easier and more affordable for OEMs to support multiple wake words on their devices.
“Giving people flexibility in how they interact with their PCs is foundational to a great user experience, and the mission of this initiative aligns with Intel’s Project Athena innovation program,” said Ran Senderovitz, vice president and general manager of Mobile Product Marketing, Client Computing Group at Intel Corporation. “We are excited to collaborate to drive the industry to scale voice experiences beyond the many 10th Gen Intel Core based systems expected to launch with multiple voice assistants this year.”
“Qualcomm chipsets allow multiple wake word engines to run simultaneously on a single device already, and we believe joining the initiative will help make these solutions accessible to more device makers and on more form factors,” said Rahul Patel, senior vice president and general manager, connectivity, Qualcomm Technologies, Inc. “We are excited to work closely with OEMs and developers to understand their needs in this fast growing area of innovation and to develop powerful and scalable solutions to support multiple services on voice-enabled products.”
Advancing the state of the art in machine learning and wake word technology
The academic community has played a vital role in advancing the core machine learning and conversational AI behind voice technology. Companies involved in the initiative will work with researchers and universities to further accelerate the state of the art in machine learning and wake word technology, from developing algorithms that allow wake words to run on portable, low-power devices to improving the encryption and APIs that ensure voice recording are routed securely to the right destination. This continued innovation will provide an important building block for long-term advancements that improve the quality, breadth and interoperability of voice services in the future.
“Customers want flexibility, in addition to greater value and functionality. They don’t want to be locked into using a specific voice service, and that means we’re going to see more households become multi-assistant environments,” said Mariana Zamoszczyk, senior analyst for Smart Living at Ovum. “This trend means that device makers and AI developers need to prioritize interoperability with other services, and work to deliver differentiated, personalized experiences through their own products or assistants.”
Bobble AI launches Emoji and Sign usage trends on ‘International Day of Sign Language’
Bobble AI has launched a trend report on the usage of emojis and signs in our day to day conversations making the chats more fun, expressive, personalized and quick.
When it comes to a conversation on a WhatsApp or Facebook Messenger, there is no difference between deaf and non-deaf users. Their conversation is dependent on words that will be typed into the keyboard, or emojis that will be picked out. The gestures remain the same, regardless of them being done physically or virtually. Emojis make it simpler and much more efficient for people to express themselves. Emojipedia 12.0 has already released a special Deaf Person emoji which will soon be a part of devices in 2019.
With the popularity of personal messaging platforms increasing rapidly, there is a need for representation for all sections of the society. This is not limited to only physical aspects, but special needs too. International Day of Sign Language is celebrated annually on 23rd September, along with the International Week of the Deaf.
Bobble has demonstrated capabilities of and by making smartphones ‘Smarter’. Solutions like Bobble Indic Keyboard allow users to choose themes, fonts, & a variety of features as per their liking, allows real-time content creation and personalization with its proprietary technology. Bobble AI-based emoji recommendation engine suggests and predicts a combination of the most used and relevant emojis by interpreting the chat context, sentiments and intent. The more you use this feature, the more it adapts to your emoji usage and personalizes conversations accordingly.
The trend report is based on the data collected on the Bobble AI platforms only. This information is made available by Bobble for educational purposes only as well as to give you general information on the usage of emojis.
Bobble AI took into consideration a sample set of emojis depicting gestures using hands.
● To understand a trend, the usage of the most popular emojis are mapped. Over a course of 12 months, the trend suggests a gradual increase in usage.
● With the advent of conversational tech like WhatsApp, Instagram, Facebook, etc., users are opting more and more for quick responses using emojis that convey their motive, as well as a text message, could.
● It can also be pointed out that these emojis are not limited to specially-abled people, but are empowering all other linguistically abled users to interact easily as well.
● Millennials are the biggest example of this. With fast-paced lifestyles and an intense need to keep up with everything, emojis let these users take part in conversations with the least amount of effort.
● 63% of our users are on Facebook every day, and 97% on Whatsapp, most of them millennials.
● The Thumbs Up emoji is among the top ten most popular emojis used on online dating apps.
● The most recognized symbol in American Sign Language (ASL), the sign for “I love you'' was added to the Unicode after repeated efforts and a petition by the Western Oregon University professor C.M. Hall and Oregon Association of the Deaf interim president Chad A. Ludwig.
● This emoji - stands nowhere on the top 10 list of related emojis. This could be because of lack of knowledge around the subject.Love-You Gesture on Google Android 10.0
India’s sole TV viewership measurement company, BARC India, which also operates the largest TV panel in the world, recently collaborated with India’s premier Institute for Management Education– The Indian Institute of Management, Ahmedabad (IIM-A) to undertake a review of the current sample design and methodology.
The statistical review team at IIM-A led by professor Tathagata Bandyopadhyay suggested the implementation of a new design which takes into account the cost of data collection and heterogeneity across markets. However, no significant deviation was found in the sample size arrived at through the new design and the one being implemented by BARC India presently, thereby validating the representativeness of the panel. The team also expressed its pleasure with the overall sample allocation mechanisms put in place by BARC India.
It may be recalled that in 2018, the Indian Statistical Institute (ISI) Kolkata had certified the representativeness of BARC India’s Panel design and methodology. More recently, in July 2019, IIM-C had also presented a report on BARC India’s TV sample and appreciated the existing practices.
IIM-A in its report stated that no significant increase in the current sample size is required with the proviso that the current cost levels are being maintained. The team noted that a panel size of 50K TV homes is reasonable to move forward, with an additional 5K -7K homes to help capture detailed insights from TV viewership data.
The review teams of both IIM-C and IIM-A were of the opinion that an ideal sample size is subjective and based on factors such as sample design, cost of sampling, overall budget, and population characteristics. There is therefore, ‘no unique or ideal solution’. They also maintain that sample sizes should be reviewed periodically.
Said Tathagata Bandyopadhyay, Professor of Statistics at IIM Ahmedabad, “The entire review process has been an enriching experience and full of learnings for both teams. BARC India utilizes an efficient sample design and sampling procedure for the management of their television viewing panel in a large and diverse market like India. The interaction with the BARC India team was extremely seamless and they extended their full support to us during the project.”
“IIM-A’s endorsement of our sample design and methodology is yet another affirmation of our ability to accurately measure and report ‘What India Watches’ 365 days a year. We work round the clock to deliver accurate and credible data to a dynamic TV industry and will take into consideration the recommendations made by the review teams to make our systems more robust” added Dr. Derrick Gray, Chief of Measurement Science, BARC India.
Dentsu India, the brand solutions agency from the house of Dentsu Aegis Network, has won the Marketing Communications mandate for the iconic stationary brand - Kokuyo Camlin. As part of the mandate, the agency will handle creative and brand strategy for Kokuyo Camlin. The account will be serviced from the agency’s Mumbai office.
Commenting on the new business win, Simi Sabhaney, CEO, Dentsu India said, “There’s nothing more special than to work on a brand that’s been an intrinsic part of one’s growing up years. We are absolutely delighted with this win and look forward to creating some memorable work on Kokuyo Camlin.”
Gaku Shinoda, COO, Dentsu India added, “We are very pleased that Kokuyo Camlin has chosen us to be their marketing communications partner. Kokuyo in Japan and Camlin in India are iconic brands and have played a huge role in setting benchmarks in the stationary category. Winning the mandate for such a notable brand is an endorsement of Dentsu India’s ability and we will do our best to further the legacy.”
Asheesh Malhotra, Branch Head- Mumbai, Dentsu India said, “It’s a brand that is probably entrenched in every living Indian’s memory. With its 80 year legacy, it is one of India’s most loved brands. We’re excited having been given the opportunity to partner Camlin and will look to continuously create exciting and talk worthy work on the brand.”
Saumitra Prasad, Chief Marketing Officer, Kokuyo Camlin said, “Kokuyo Camlin is happy to have on board Dentsu India to manage the creative strategy and execution for its brand Camlin. The brand will look to accelerate its growth story in the evolving stationery market in India by rejuvenating our iconic brand and strengthening the bond with consumers.”
As marketing automation has become more mainstream, the barriers that businesses experience are also clearer.
Expectations from marketing automation platforms have not matched reality. Prior to adoption, companies imagine marketing automation software to be out-of-box solutions that can completely overhaul their marketing with very little effort
The reality is that marketing automation tools are not simple plug and play tools.
You need to invest time and resources to generate returns just like any other operations system that your business adopts.
To see high returns from marketing automation systems you need:
- a clear strategy of how the automation platform fits into your overall marketing
- knowledge of marketing automation and data management practices
- resources who can manage and scale the automation system
Companies continue to invest in automation platforms without realizing the investment required in resources to make it successful.
A 2019 report by Smart Insights identifies Data, resources, talent and strategy as top barriers to success of marketing automation implementations.
Email marketing campaigns remain the primary objective
London Research reports that even though 86% of companies are using email campaigns as a primary activity, less than one third are using features that make automation platforms truly valuable – behavioral scoring, lead nurturing and predictive personalization.
By not using these core features, the marketing automation system is reduced to merely a campaign launcher. And there are cheaper, simpler tools out there that let you blast email campaigns.
Why spend so much on marketing automation? The purpose of automation platform is to automate processes.
Our real-world experience of working with businesses from all over the world also validates these challenges. Throughout our journey we’ve asked ourselves - what's the difference between companies that see great results from marketing automation and those for whom the impact is underwhelming?
We continue to see a massive opportunity for companies to better utilize their marketing automation capabilities. This can supplement or even replace decisions based on legacy knowledge or intuition.
From our experience, we recognized 4 things that successful companies do best to optimize their marketing automation ROI.
Define a clear data management strategy
Success with marketing automation relies on accurate, complete and valid data. Good data practices ensure customer information stays relevant.
An ad-hoc approach to data is detrimental as it creates pools of low quality, junk or irrelevant data. These pools become indistinguishable from good quality data. When you try and run campaigns on these mixed data sets, you are unable to cleanly separate bad data from the good. This leads to poor ROI as you don’t even know which users deserve your attention.
Lack of data strategy also severely limits automation capabilities as it makes it harder to identify actions taken by customers. Thus, it becomes impossible to segment users into relevant buckets for creating personalized nurturing campaigns.
You can start building a data management strategy by taking 3 key steps:
Setup data hygiene standards
It’s important to prioritize quality of data at every stage. Letting data hygiene slide even on a few occasions can snowball into bigger challenges. Over time your growing marketing database becomes harder to sanitize – simply because of the sheer volume.
Hygiene process begins when you first use the automation platform. At the early stage you can take steps to stop junk and irregular data from making its way into the automation system. Any lists you upload must be cleaned prior.
As you execute marketing campaigns, and more data gets generated, hygiene management must continue. To encourage compliance, you can setup guidelines and share them with the team. Some basic ones to start with are:
- Exclude junk or test emails – if a few test emails need to be added, they must be appropriately flagged using a ‘lead status’ or ‘source field’ set to ‘Test’.
- Reject irregular characters – create forms that reject invalid phone numbers, public emails or randomly typed strings letters. You can automate this detection.
- Pre-validate contact details – upload emails that have opted-in, have not hard bounced and are valid (they can receive emails). You should implement OTP verification for phone numbers and link verification for email addresses.
- Upload external opt-out lists (if any) – if you are switching from another platform you can import the historical opt-out list to the new system. You can also choose to purge such data permanently prior to uploading it.
Once a month or quarter, review your data and see if there are any loopholes which are adding bad data in the system and include new guidelines to keep it out.
Add and customize fields to segment users
Users interact uniquely with your business. Each user takes different actions at different times. Some may take weeks to make their first purchase, while some may buy on the first day. You must be able to distinguish them based on the actions they take so you can setup different ways to convince them.
This is where one of the most underused features of marketing automation can help – custom fields. Most platforms have a few pre-defined fields for segmenting customers, but you can also add new fields to reflect your business process.
Once you add or customize the fields you need, setup workflows to change their values as your users interact with your business. These fields become identifiers of user behavior. They indicate which users are at what stage of their journey with your brand.
Here are some fields you can define to start with:
Source – which channel or program was the origin when you acquired the user
Source history – which channels were touched on their journey before sign-up
Lead status – identify whether the contact is new, in a stage of nurturing, is a returning customer, is a high value customer. is invalid, and so on.
Engagement level – define a scale or score to measure user’s overall engagement – a sum of all his/her interactions
Let’s take a quick example. Assume you create a field that tracks the first purchase of a customer and sets its status from “Signed up” to “Made first purchase”. Then you set up a nurturing flow that automatically detects when a user is set to this status. The system then sends a series of emails and notifications over a week to convince the user for the next purchase. Similarly, you can set up flows for inactivity, abandoned carts, category stickiness – the possibilities are endless.
Fine-tune the strategy as your business grows
Revisit your data strategy at regular intervals to ensure everything is working as desired and make changes if necessary. Reviews scheduled once a month or once a quarter are enough for most businesses.
During these reviews your team can set aside time to clean data that escaped the agreed standards.
As you discuss gaps in the policy and make changes you can also make enhancements such as adding new fields or value, proposing new standards and rethinking some of the automation processes.
Build a deep understanding of your customers
Collecting data on your customers is important but it’s rarely the complete picture. It's just as important to pin down their behavior. 70% of marketers struggle to create a unified view of their customers’ behavior.
As the graphic below from a research by McKinsey & Company shows, returns from deep personalization are growing rapidly regardless of the industry you operate in – retail, travel, CPG, banking or any other.
Investing in understanding your users, with a mix of data and behavior, is essential to offering deep personalization that people are becoming accustomed to.
Marketing automation platforms can take you closer to fully understanding your customers. They combine static demographic data with behavioral signals to clearly identify unique purchase patterns that require personalized content.
Lead scoring is a classic mechanism for this. It’s one of the core features of any automation platform.
To setup lead scoring, you assign scores to every foreseeable customer activity. Then you set up rules to tell the system how to change the total score of a customer based on different actions. Once the rules are set the process is automatic.
For example, you can add 5 points if they click on an email campaign. Add another 30 points if they make a purchase. This way a customer’s total score is a cumulative measure of their engagement with your brand.
As the score changes based on user actions, you can setup milestones that trigger branches of nurturing flows to engage them further. This is an ideal way to understand behavior as buyers with similar preferences will gravitate towards the same nurturing flows. You can target these groups with even more specialized content.
Track this process regularly to adjust the score and add/remove actions as your users keep maturing.
Progressive profiling is another great tool to create frictionless experiences.
For users with strong engagement scores you can progressively request their information in phases rather than all at once. This makes the experience less overwhelming for the user.
During acquisition, for example, you only ask for their name and email. As they take more steps that require additional info, you append those pieces to build their complete profile.
Retailers like Amazon do this well. They ask users for shipping, payment and credit information only when they reach those steps. Progressive profiling improves acquisition rate as users are not as overwhelmed. It also promotes a more seamless experience and gives them greater control over their personal information.
Think multi-channel, multi-touch campaigns
Online purchases are becoming more complex with an endless stream of information available to users. Buyers comprehensively research and compare products before finalizing a purchase, but they also transact more often with brands that maintain an ongoing interaction.
Companies who understand their users better can maintain these interactions with automated content sent over a mix of channels – email campaigns, SMS marketing, push notifications, in-app messages, etc.
As an example, for buyers who just made their first purchase, you can setup an automated flow of 3 email campaigns, at 1-day, 3-day and 5-day intervals, with recommendations for the next purchase. You can include products from the categories that the user already browsed or additional recommendations based on what other buyers purchased. This is a great way to enhance your interactions with the buyer using personalized content.
There are endless ways of designing nurturing flows over email, SMS, notifications and other available channels. Some basic ones are:
Welcome and on-boarding series – welcome users and offer ideas for their first purchase
Cart Abandonment flow – remind users that they have items in the cart and include suggestions for other products they can combine with their purchase
Event flows – designed around birthdays and anniversaries to create brand loyalty
New category flow – sent to buyers that don’t respond to existing flows. You can use experimental content to measure their response.
The benefits of automated multi-touch campaign are clear. One, it keeps your brand name in active memory of buyers. They are more likely to buy from you if they remember you.
Second, it builds trust and loyalty. Users enjoy personalized recommendations and start building an affinity towards brands that they feel truly understand their preferences and choices.
Finally, it makes your marketing more efficient. As you can automate a series of touches with a large percentage of your users your team has more time to create content and improve underperforming campaigns.
With multi-channel, multi-touch campaigns you will start seeing results that are closer to your expectations from the automation platform.
Manage and grow your marketing operations
Ultimately, you must realize that none of this is a one-time fix. There is a reason marketing automation companies charge thousands of dollars for ongoing consulting.
As a platform, marketing automation gives you all the tools, but you must put in the time and effort to customize and extract value from it.
Companies that generate maximum returns from their marketing automation platforms strive to continuously improve it. Treating it as an ongoing project is necessary.
You can assign a dedicated marketing operations team to manage and grow the automation platform. This team can work with your wider marketing teams to ensure that data, automation, tracking and reporting processes support marketing programs smoothly.
Additionally, they can identify gaps and improve those in short sprints.
If it’s not possible to assign dedicated resources to marketing operations, you can allocate 10-15% of time of your marketing team at regular intervals for improvement. This can be done for example, once a month, once a week or a few times a quarter depending on your scale.
Marketing automation maturity is a journey that takes time and resources, but the results are worth the effort!
Companies that are successfully using marketing automation platforms have grown exponentially. Remember to start small, do the right things, and keep improving.
Written by Avadhoot Revankar,Chief Growth Hacker & Product Evangelist at Netcore.Avadhoot carries extensive knowledge in the MARTECH & Mobile Marketing space.
This post was originally published on Netcore Smartech. Click HERE for the original article