As IBM Ramps Up Its AI-Powered Advertising, Can Watson Crack the Code of Digital Marketing? | Ad Week

As IBM Ramps Up Its AI-Powered Advertising, Can Watson Crack the Code of Digital Marketing?; ; In Ad Week (Advertising Week); 2017-09-24.
Teaser: Acquisition of The Weather Company fuels a new division

tl;dr → Watson (a service bureau, AI-as-a-Service) is open for business.

Mentions

The Weather Company

  • lines of business
    • location-based targeted audiences, delivered to the trade.
    • weather indica, delivered to consumers.
  • 2.2 billion locations/15 minutes
  • Dates
    • WHEN?, Acquisition by IBM
    • 2016-01, new business strategy,
      “AI” as a service (AIaaS)
  • Artificial Intelligence (AI)
  • Cloud Computing
  • Products
    • WeatherFx
    • JourneyFx
  • The Weather Company is a <quote>legacy business<quote> (deprecated).
  • AIaaS is a <quote>cutting-edge advertising powerhouse</quote> (house of power).

Watson Advertising

  • Cognitive Advertising
    • contra Computational Advertising, circa the ‘oughties (2005)
    • something about
      • <buzzzz>transform every aspect of marketing from </buzzz>
      • something about image and voice recognition to big data analysis and custom content.
  • What is it? (What is Watson-as-a-Service?)
    • Count: <quote>dozens</quote>
    • Interfaces
      • API
      • Projects <quote>studio-like</quote>
    • Pricing: <quote>millions of dollars</quote>
    • Structure: four (4) sub-units
  • “<snip/>It’s not been designed to target consumers the same way that Alexa or Siri have been,” attributed to Cameron Clayton.

Units

The 4 pillars of Watson Advertising.
  1. Targeting, Audience construction & activation
  2. Optimization, Bidding & buying
  3. Advertising, Synthesis of copy and creative
  4. Planning, media planning, the buy plan, the execution plan

Audience Targeting

  • the flagship service
  • neural networks
  • scoring users, propensity scoring <quote>based on how likely they are to take an action</quote>
  • towards CPA or CCPV or CPVisit or <more!>
  • Performable on the Weather Company O&O
    • <quote>but on TV, print, radio and other platforms. <quote>
    • Partnerships
      • Cognitiv
      • Equals 3

Optimization

Bidding Optimization
  • Is too boring for details early in the article.
  • Optimize against brand-specific KPIs.
  • Uses <buzzz>deep learning and neural networks</buzzz>
  • Optimize Cost Per Action (CPA).

Advertising

  • Badged as Watson Ads and Watson Advertising
  • Services
    • content creation
    • content copywriting
  • Launched: 2016-06.
  • Is merely: nterest-Based Advertising (IBA)
    which in turn is a but regulatory term of art, that covers a wide range of in-trade practices.
  • Sectors, aspirational
    • <fancy>aviation</fancy> (airline ticket booking?)
    • insurance
    • energy
    • finance
  • Cognitive Media Council,
    • a focus group.
    • a user group, “friends & family” of the business.
    • a group of important customers representatives
      <quote>senior-level executives from agencies and brands</quote>
Reference Customers
Toyota
  • Mirai
  • Prius Prime
  • Benefits
    Attributed to Eunice Kim, Toyota (TMNA), something about…

    • <buzzz>create a one-to-one conversational engagement</buzzz>
    • <buzzz>garner insights about the consumer thought process that could potentially inform our communication strategies elsewhere”</buzzz>
Campbell’s
  • the Soup people
  • Something about creative synthesis
    themed as: recipe generation with flu symptoms with location
H&R Block
  • Something about creative synthesis
    themed as: automated robot tax expert, suggest tax deductions.
UM [You and Em]
  • An agency. Off shore? They have a “U.S. CEO” Maybe one of those English Invasion thingies.
  • Refused to name their client.
  • Something about auto dealerships.
  • <quote>meshing Watson data with client stats to analyze metrics across a large number of car dealerships in a way that optimizes ad spend while also checking local inventory to see whether or not it should personalize an ad to someone in that market.<quote>
  • <quote>combination of weather data, Google searches and pollen counts to trigger when media should be bought in various markets.</quote>

Planning

  • <quote>AI-powered planning</quote>

Partners

Cognitiv
Something about a partnership for understanding marketing texts.
Jeremy Fain, CEO and co-founder
Equals 3
Lucy, a product-service-platform.
Something about <quote>to uncover extra insights and research.<quote>

Fairness & Balance

Promotions

Ogilvy & Mather
  • Honorific <quote>longtime agency<quote> [fof record for IBM].
Stunts
2011
Jeopardy
2015
[Television] campaign, with Bob Dylan.
2016
Synthesis of the trailier for Morgan (a move; genre: science fiction)
2017-02
Performance, an “analysis” of the stylings of Antoni Gaudi, <quote>inspire an art installation </quote> (what does that mean?)
The “art installation” was exhibited at the Mobile World Congress in Barcelona.
Statista

…is quoted
the future is boosted.

Sectors
  • “AI services”
  • “Big Data services”

Themes

Problem
  • The people are “afraid” of AI.
  • The people need to be groomed to accept AI.
Remediation

Ensmoothen & enpitchen the Artificial Intelligence (AI) as…

  • humble
  • friendly
  • ”I’m here to help’ type personality”

Attributed to Lou Aversano, Ogilvy.

Detractors

James Kisner, Jeffries

Via: James Kisner, A Report, Jeffries, 2017-07.
Jeffries is an opinion vendor in support of an M&A banking operation.
tl;dr → Watson is a failing product-service. <quote>IBM is being “outgunned” in the race…</quote> (yup, he mixed the metaphor).

  • as evidenced in measured job listings at Monster.com
    Apple had more listings booked thereon than IBM.
  • Customers were interviewed.
    Watson’s performance/price ratio was low (the rate card is very high).
    2016-10, IBM reduced the rate card for API access <quote>by 70 percent</quote>
  • Lots of press
  • Not a lot of monetary results, as evidenced in the quarterly & annual reports.
Joe Stanhope, Gartner

Via: an interview, perhaps;
Gartner Group is an opinion vendor.

  • Too much hype, can be forgiven
  • Gartner runs the Hype Cycle brand
  • Claims: <quote>IBM does seem to be all-in with Watson.<quote> (be nice to hear that from IBM, not as a “hot-take” from a newshour pundit
DemandBase, Wakefield Research

A Report; attributed to “staff”; DemandBase and Wakefield Research

  • A survey,
    • “how do you feel?”
    • Do you “have plans-to …” in the next N months.
  • There are a lot of uncertainties

Uncertainties

Training Data
  • Just isn’t there.
  • And … computers can only give answers, it can’t give [pose] questions.
Does it [even] Work?
  • No one knows.
  • Many are nervous.
  • No one wants to be first to fail
    (& be fired for outsourcing their job function to The AI).

Competitors

  • Einstein, of Salesforce(.com)
  • Sensei, of Adobe
In-House
  • Buying operations, Xaxis of WPP
    the “AI” is a “co-pilot” to the trading desk operator; optimization recommendations towards CPM and viewability; North American operations only.
  • others?
    Surely everyone nowadays has some initiative that does “co-pilot”-level decision support to adops.
Research Efforts
  • Amazon
  • Facebook
  • Google
Venture Capital
  • Albert
  • Amenity Analytics
  • LiftIgniter
  • Persado
Amenity Analytics

An exemplar of the smaller-nimbler-smarter clones of the Watson genre.

  • A Watson-type experience, but cheaper
  • Does text mining of press releases
  • Reference customers:
    Pepsi
  • A spin-out from some hedge fund, <quote>origins in the hedge fund world</quote>
  • Nathaniel Storch, CEO, Amenity Analytics.
  • <zing!>“Think of it as ‘moneyball’ for media companies,”<zing!>, attributed to Nathaniel Storch.

Consumer

  • Siri, of Apple
  • Cortana, of Microsoft
  • Now, of Google

Who

  • Lou Aversano, U.S. CEO, Ogilvy & Mather (Ogilvy, O&M).
  • Jordan Bitterman, CMO, Watson (Business Unit), IBM.
    attributed in quoted material aso “earlier this year” (2017?); c.f. Michael Mendenhall
  • Kasha Cacy, U.S. CEO, UM
    UM is an agency.
  • Cameron Clayton,
    • General Manager, Content and IoT Platform, Watson (Business Unit), IBM..
    • ex-CEO, The Weather Company
  • Jacob Colker, “entrepreneur in residence,” The Allen Institute
    …quoted for color, background & verisimilitude.The Allen Institute is a tank for thinkers.
  • Jeremy Fain, CEO and co-founder, Cognitiv.
  • Chris Jacob, director of product marketing, Marketing Cloud, Salesforce(.com).
  • Eunice Kim, media planner, Toyota Motor North America (TMNA).
    …quoted for color, background & verisimilitude.
  • James Kisner, staff, Jeffries.
    …quoted for color, background & verisimilitude.
    Jeffries is an advice shop, like Gartner, but different.
  • Francesco Marconi,
    …quoted for color, background & verisimilitude.

    • strategy manager and AI co-lead, Associated Press
    • visitor, MIT Media Lab
  • Michael Mendenhall, CMO, Watson (BU), IBM.
    announced as CMO in prior press [Ad Week, Marty Swant, 2017-07-07].
  • Sara Robertson, VP of Product Engineering, Xaxis of WPP.
  • Joe Stanhope, staff, Forrester
    …quoted for color, background & verisimilitude.
  • Nathaniel Storch, CEO, Amenity Analytics.
  • Marty Wetherall, director of innovation, FallonFallon is the agency that certain campaign booked on Watson for H&R Block

Pantheon

  • Antoni Gaudi, architect (per civil engineering), citizen of Spain.

Previously

In archaeological order, within Advertising Week

Previously filled.

Conversational Architecture Visual Expression (CAVE), an experience design language

CAVE Language

Who

Mentions

  • Conversational Architecture Visual Expression (CAVE).
  • Persona, Affinity, Goal, Environment (PAGE).
  • Content, Functionality, Rules, Style (CFRS?)
  • Switch (a state; why the fancy misdirecting name?)
  • Adaptive Context Engine (ACE)
  • Experience Metaphors
    • Desktop
    • Page
    • etc.
  • Forces
    1. Mobile Devices
    2. Social Media
    3. Data
    4. Sensors
    5. Location
  • Layers
    • Modes (Application)
    • Context (User, Environment)
    • Data
  • Conversation Concepts
    1. User Personas
    2. User Affinity
    3. User Goals
    4. User Environments
    5. Aggregating and Analyzing Context
    6. Your Agenda
    7. Model Response
    8. Continuity
    9. Ubiquity
    10. Authenticity

Examples

Bike App Modes Rain Imminent Switch
Bike App Modes Beat the Weather Switch

Background

  • Axilent
    • Loren Davie, CEO and founder
  • Adaptive Context Engine (ACE)
    • A SaaS content publishing platform.
    • <quote>ACE makes it easy to target content to users based on their context.</quote>
    • Has plugins.
    • User Profiles
  • Something about
    • Content Channels
    • Content Everywhere
    • Targeted Content
  • Content Management System (CMS)
    • needs one
    • supplies one
    • integrates with one
  • Axilent At a Glance

Promotions

  • Loren Davie; How We Built A New Design Language For “Aware” Apps; In Fast Co.LABS; 2013-11-15.
    Loren Davie is the CEO and founder of Axilent, which aids developers in building contextual applications.
  • Loren Davie; Why The Next Big Thing In Computing Is Conversation; In Fast Co.LABS; 2013-09-03.
    Teaser: For 30 years, the dominant metaphor in computing has been the desktop: files, folders, and documents. But with apps like Apple’s Siri and Google Now, developers should be looking to a far more futuristic paradigm: the natural-language, contextually aware human-computer “conversation.”

Similar

  • StateCharts
  • UML

Via: backfill

How Mobile Is Eating The World | Benedict Evans, Enders Analysis at Business Insider

Benedict Evans (Enders Analysis); How Mobile Is Eating The World; In Business Insider; 2013-11-17.

Mentions

  • Lots of up-and-to-the-right.
  • Irrelevance of Microsoft
  • Scale at Samsung
  • Scale at Apple
  • Tablets
  • Post-PC Vision
  • Four Horsemen
    • Google
    • Apple
    • Facebook
    • Amazon
  • Chinese Android is not Android
  • WebKit Everywhere
  • Smartphones are inherently social
    • address book
    • photo library
    • push notifications
    • home screen, task switcher
    • switching apps is easy
  • Unbundling Functions vs Unbundling Friends
  • Cards as Content Packets
  • Social as Discovery
  • Tablet Trends
  • Open Questions
    • What is the identity platform
    • Watch: protocols-and-services

Outline

  • The state of PCs
  • Smartphones are exploding
  • More mobile growth coming
  • The future is mobile
  • The world is 2017
  • Growth in emerging markets
  • Fundamental change
  • Fundamental change in scale
  • Fundamental change in use
  • What does mass mobile internet use really mean? From this…
  • … to this
  • Industry scale
  • Polarisation of manufacturers
  • The irrelevance of Microsoft
  • Scale at Samsung…
  • and scale at Apple
  • Very different products
  • Apple sticking to the high end?
  • Glass is eating the world
  • Tablets overtaking PCs
  • Tablet market splitting
  • iPad dominates use everywhere
  • Two distinct ‘tablet’ markets
  • Tablet [is] dynamic quite different to smartphones
  • Tablet [is] dynamic quite different to smartphones, continued
  • Blurring definitions
  • Tablets in 2013
  • Still lots of unknowns
  • ‘Four horsemen’ driving the agenda
  • Ecosystem sizes
  • Reach != value
  • (Chinese Android isn’t Google)
  • Geographic variation
  • Ecosystem is the key leverage point
  • People like apps
  • Mobile platform wars over?
  • Speed of innovation?
  • Different focus for innovation
  • App engagement
  • Self-selection
  • Ecosystem cohorts
  • Ecosystem cohorts?
  • Future of Android
  • Mobile social scale
  • Mobile social scale, continued
  • Children’s use of messaging
  • Smartphones are inherently social, unlike the desktop web
  • People happily abandon history, 3x slides
  • Facebook is one of many
  • Facebook is doing well on mobile
  • Half of DAUs are mobile-only
  • Is the mobile opportunity so big that it doesn’ matter to Facebook if it isn’t dominant?
  • Unbundling
  • The Aggregation Cycle
  • Unbundling Facebook
  • Unbundling functions or unbundling friends?
  • Mobile social is still in flux
  • There’s money in stickers
  • The next opportunity is creating the next platform
  • Cards as content packets – social as discovery
  • Two trends for mobile content
  • Again, all this is in flux
  • Broader uncertainty and opportunity
  • Blurring boundaries

Via: backfill

Jeannie (like Siri) is fantastically dangerous

Jeannie (Like Siri)

This means that any elements on any calendar that is visible to you (shared with you) is totally public.  It can be, and you consent to have this indexed and displayed to the world irrevocably.  For this, the app receives the coveted Poisonous App appelation.  Avoid.

Who are these people and why would they even want to hold that sort of data?  Temporarily?  Ever?

From the permissions page on Play Store.

Your personal information

read your contacts
Allows the app to read data about your contacts stored on your tablet, including the frequency with which you’ve called, emailed, or communicated in other ways with specific individuals. This permission allows apps to save your contact data, and malicious apps may share contact data without your knowledge. Allows the app to read data about your contacts stored on your phone, including the frequency with which you’ve called, emailed, or communicated in other ways with specific individuals. This permission allows apps to save your contact data, and malicious apps may share contact data without your knowledge.
read calendar events plus confidential information
Allows the app to read all calendar events stored on your tablet, including those of friends or co-workers. This may allow the app to share or save your calendar data, regardless of confidentiality or sensitivity. Allows the app to read all calendar events stored on your phone, including those of friends or co-workers. This may allow the app to share or save your calendar data, regardless of confidentiality or sensitivity.
add or modify calendar events and send email to guests without owners’ knowledge
Allows the app to add, remove, change events that you can modify on your tablet, including those of friends or co-workers. This may allow the app to send messages that appear to come from calendar owners, or modify events without the owners’ knowledge. Allows the app to add, remove, change events that you can modify on your phone, including those of friends or co-workers. This may allow the app to send messages that appear to come from calendar owners, or modify events without the owners’ knowledge.

Actuality