IDC Revision

Week 9 · Ofir Richman

Lecture 9 — Emerging Platforms & Trends

Lecture 9 — Emerging Platforms & Trends

Part of: Digital Marketing Topic 9 — Key Shifts (Part 2): where attention is moving next · In-class slide deck (Ofir Richman, 2026A) Key concepts: FAST, Linear Viewing, SEO, SERP, Personalization, CLV

About this note

This is a reconstruction of the in-class deck, which is slide-heavy — many slides are a title plus a logo, screenshot or chart, so the connective explanation came from Ofir's narration in the room. Where a slide carried body text (the CTV monetization and AI-marketing slides especially), it's reproduced faithfully. Where a slide was only an image, chart or product photo, the surrounding explanation is reconstructed from standard theory and the slide's stated data — those passages are flagged with a "reconstructed" note. Charts (Statista, Gartner, PwC, Activate) are described at the level printed on the slide — no figures are invented beyond what's shown. Use it as a study scaffold, not a verbatim transcript.

TL;DR

This is the second "key shifts" lecture, and it's a tour of where attention (and ad money) is migrating. (1) Connected TV — the move from linear broadcast to connected, addressable streaming, its value proposition ("Netflix: no ads"), the five streaming monetization models (SVOD, AVOD, TVOD, HVOD, FAST), CTV metrics (VCR, CPCV) and ad formats. (2) Three targeting frontiers — location-based (IP, GPS/proximity, geofencing), voice-based (voice-search optimization, question-format content), and visual search (Google Lens, Gen Z). (3) Spatial computing — VR/AR/XR headsets and AI glasses, with a healthy dose of "not yet." (4) AI-generated marketing — the split between AI enthusiasts and skeptics, AI assistants as a fast-growing traffic source, Gartner's 2025–2028 predictions, and how to structure content so AI search engines cite you.

Part 1 — Connected TV (CTV)

Linear vs Connected TV

The core distinction the deck hammers home:

Linear TV vs Connected TV
  • Linear TV — commercials aired on cable or broadcast during scheduled programming. Different users see the same ad during the broadcast. Broad reach, but it lacks precise targeting, gives limited insight and frequency control, and is expensive.
  • Connected TV (CTV) — ads delivered to viewers streaming content over connected devices (smart TVs, streaming sticks, consoles). Different users see different ads during streaming. It leverages deterministic data like IP addresses to target specific households, and offers real-time, granular measurement.

The one-line takeaway: CTV is addressable TV — the lean-back experience of the big screen combined with the targeting and analytics of digital. See Linear Viewing for the term itself; the broader linear-vs-on-demand platform landscape is set up in Digital Marketing Lecture 2.

The value proposition is changing

From the slide — the Netflix billboard

A blank billboard reads "YOUR AD HERE". Next to it, a Netflix billboard shows a TV and the line "NOT HERE. No ads. No distractions. Just Netflix."

The juxtaposition makes the point: the value proposition of premium streaming was originally the absence of advertising. That's exactly why the shift back toward ad-supported tiers (below) is such a big deal — the platforms that sold "no ads" are now building ad businesses. The TV OS slide (a smart-TV home screen studded with app tiles and a banner ad) reinforces that the TV itself has become an ad-serving platform. Reconstructed connective tissue; the slides were images.

Benefits of CTV advertising (from the deck)
  • Precision targeting — target specific households via IP addresses and deterministic data, not broad demographics; better efficiency and ROI.
  • Enhanced measurement — track viewership to a one-to-one level using streaming-specific metrics; optimise in real time.
  • Multi-channel retargeting — device graphs link multiple devices to one household, so CTV plugs into a multi-channel strategy (build awareness on the big screen, retarget elsewhere).
  • Improved / incremental reach — reach cord-cutters and younger audiences (Gen Z, millennials) who watch less linear TV.
  • Easier access to premium content — small brands can bid on premium inventory (live sports, awards shows) via Netflix, Disney+, Prime Video at a fraction of linear cost.
  • Increased viewability — ads are often unskippable, on a large screen, alongside fewer ads.
  • It's cookieless — relies on 1st-party / probabilistic signals (IP, device IDs) rather than 3rd-party cookies, which are being phased out.
  • Flexible pricing — no long-term contracts; scale spend up or down like other programmatic channels.

Streaming monetization models

This is the most testable list in Part 1. Five models, each defined by who pays and how:

The five CTV / streaming monetization models
Model Full name How it makes money Examples
SVOD Subscription VOD Users pay a monthly/annual subscription for unlimited access. The most widely used model; loyal viewer base. Netflix, Amazon Prime, Disney+, HBO Max, Hulu
AVOD Advertising VOD Content is free; the publisher earns from advertising (banners, sponsorships, paid placements — advertisers pay when the ad is viewed). YouTube, TikTok, Instagram, Facebook
TVOD Transactional VOD Viewers are charged per single piece of content. Pay-per-view (PPV), download-to-rent (24/48h), electronic sell-through (EST = pay once, keep forever). Prime Video, Sky Sports Box Office, iTunes, LinkedIn Learning
HVOD Hybrid VOD Combines models (e.g. a cheaper ad-supported subscription tier). Addresses competition and stretched household budgets; widens the audience and reduces churn. Netflix ad tier ($6.99/mo), yes+
FAST Free, Ad-Supported (Streaming) TV Free and ad-supported, with a linear-like experience — you watch over the internet but content is scheduled with ad breaks and you can't choose what to watch. Available across devices. Peacock, Roku TV, Pluto-style channels, N12
FAST is the one to nail

FAST = Free, Ad-Supported Streaming TV. The trap is confusing it with AVOD: both are ad-funded and free, but FAST is linear (scheduled channels, no on-demand choice) while AVOD is on-demand. The deck's data slide notes that on FAST services consumers gravitate heavily toward news programming (≈42% of viewing hours), with entertainment, nature and movies trailing. FAST is a direct past-paper MCQ — see Exam takeaways.

Where the money is going (charts, described as shown)
  • Activate forecast: advertising revenue across select ad-supported SVOD services is projected to grow ~25% annually through 2027 (from ~$6.2B in 2023E to ~$15.3B in 2027E), with Netflix's ad revenue growing fastest (~59% CAGR off a small base). Figures are the ones printed on the slide (Activate / WSJ).
  • PwC Global Entertainment & Media Outlook 2024–2028: Subscription VOD revenue is the largest and still rising, Advertising VOD is climbing steadily, and Transactional VOD is flat — the growth is in subscriptions and ads, not per-transaction rentals. Described at the slide's stated level; no exact endpoints invented.

CTV metrics

Metrics to measure a CTV campaign (from the deck)
  • Impressions — total number of times an ad was displayed / served (scale and reach of the campaign).
  • Reach — number of unique viewers who saw the campaign over a period.
  • Incremental reach — the additional unique viewers a CTV campaign reaches beyond linear TV.
  • VCR — Video Completion Rate — the % of viewers who watch the ad from start to finish (attention / effectiveness).
  • CPCV — Cost Per Completed View — how much budget is spent per ad watched to completion (efficiency of paying only for full views).

The teaching point: CTV gives "granular insights advertisers could only dream of with linear." Incremental reach is the argument for adding CTV on top of linear (net-new audience), and VCR / CPCV are the completion-quality metrics you can't really get on broadcast.

CTV ad formats

Four common CTV ad formats
  • In-stream (linear) ads — play before, during or after content (pre-roll, mid-roll, post-roll), like a TV ad between scenes.
  • Non-linear (overlay) ads — static images, text or animation on top of the content without interrupting playback.
  • Companion ads — appear alongside / surrounding the video (text, images, rich media, skins) without disrupting the viewing experience.
  • Interactive & shoppable ads — engage the viewer directly: choose a storyline, play a mini-game, or buy by scanning a QR code from the ad.

Part 2 — Location, voice & visual

Location-based marketing

Three key methods (from the deck)
  • 1. IP-address marketing — every server, computer or mobile device has an IP address; you use that location information to target. The simplest and broadest location method.
  • 2. GPS / proximity marketing — set a campaign around a specific GPS point or radius so you market to people within a certain distance of a spot. Can target residents, commuters or tourists; combine with interests/demographics; base it on location history.
  • 3. Geofencing — use GPS to draw a virtual boundary; when a device enters, leaves or lingers inside it, a software response triggers (e.g. an app push notification). You can geofence a campus, a neighbourhood, even a lecture hall. Example on the slide: weather targeting — use live weather to trigger ads/promotions.
The IKEA example

The deck plays an IKEA case as the worked example of proximity/geofencing done well — a location-triggered campaign that reaches people when they're physically near (or inside) the relevant zone. The slide was a video link, so the specifics are reconstructed; the point is that geofencing turns physical presence into a marketing trigger.

Voice-based marketing

Voice assistants (Amazon Echo/Alexa, Siri, Google Home) shift search from typing to speaking, and the deck frames why that changes content:

Optimising for voice search — "people speak differently than they type"
When typing When speaking / AI-ing
Convenience Ask questions
Short, keyword-rich phrases Full statements / natural sentences
→ keyword content → Create content as answers to questions

The strategic move: because voice and AI queries are conversational and question-shaped, structure your content as questions and answers so it matches how people actually ask. The deck points to AnswerThePublic ("discover what people are asking about…") as the research tool for finding the real questions to answer. (This question-format idea returns, amplified, in the AI-marketing section below.)

Direct MCQ — voice search

The "people speak differently than they type → write content as answers to questions" idea is a past-paper MCQ. Nail the logic: voice/AI queries are full, natural questions, so question-and-answer content wins. See Exam takeaways and 2024 Exam B1.

Google Lens / visual commerce (from the deck)
  • There are now ~25bn Google Lens searches every month — point your camera at an object and search it.
  • ~80% of Gen Z relies on Google Search for shopping discovery, research and decisions.

The shift: search is no longer only text or voice — it's visual. You photograph the shoes, the lamp, the outfit, and the engine finds visual matches and shopping results. For visual commerce this means product imagery, structured product data and being findable by image matter, especially for the Gen Z audience the slide calls out. Reconstructed strategic implication; the slide stated the two stats.

Part 3 — Spatial computing

VR / AR / XR

Spatial computing is the "what's next" frontier: immersive VR/AR/XR experiences via headsets and glasses. The devices on the deck:

  • Apple Vision Pro — the flagship spatial-computing headset ($3,499); enthusiasts on X call it "UNREAL… I don't regret ordering a Vision Pro for even a second."
  • Meta — cheaper VR headsets plus holographic AR glasses prototypes; the Ray-Ban Meta and Oakley Meta AI glasses line (from ~$239) and the September 2025 Meta Ray-Ban Display.
  • Google — its "comeback" with Android XR AI glasses (frames from Gentle Monster and Warby Parker, Gemini built in) announced for release.

Adoption scepticism — "not yet"

The deck's honest verdict: NOT YET

For all the hype, the slides deliberately puncture it. Headlines shown: "Report: Apple may stop producing Vision Pro" (production cut), "Please Don't Make Me Wear Mark Zuckerberg's Ugly Glasses — no one wants to wear a computer on their face." The framing ("It happens to the best…", "NOT YET!") is that spatial computing is a real but not-yet-mainstream channel — worth watching, not yet worth betting the marketing budget on. The human-premium / AI-fatigue counter-current — why people push back on the always-on tech future — is developed in Digital Marketing Lecture 8.

Part 4 — AI-generated marketing

Consumers in the AI age: enthusiasts vs skeptics

The deck uses Statista's "4 types of consumers in the AI age" to show the market is split:

AI enthusiasts vs AI skeptics (Statista, described as shown)
AI Enthusiasts AI Skeptics
"Now is the time to purchase" (~66% of US enthusiasts) Low trust in news/media (~11%) and government
Social media is a shopping channel (~42% use social commerce) Suspicious of personalization (only ~22% feel AI offers match them)
Open to AI-generated influencers (~52% of UK fans would buy on their recommendation) "Exhaustion prevails" (~26% of German skeptics feel "exhausted")
Correlate with wellness / GLP-1 use, "omni wellness" Prefer sports/wellness offline, in nature, not virtual

The marketing implication: the same message won't land on both groups. Enthusiasts reward convenience, social commerce and AI-generated creative; skeptics distrust personalization and want the human/offline. (This maps onto the Personalization tension — powerful for enthusiasts, off-putting to skeptics.)

AI assistants as a growing traffic source

E-commerce AI referral traffic (Statista chart, as shown)

AI-assisted shopping "has officially arrived." Monthly e-commerce traffic from AI sources rose from just over 100,000 visits (Jan 2024) to more than 4 million (June 2025) — described on the slide as a ~35-fold increase in under 18 months, driven by ChatGPT, Perplexity and new e-commerce integrations. The discovery phase of shopping is migrating from search engines and social to AI assistants.

The strategic consequence runs straight into AI search: if buyers now start their journey inside an AI assistant, you need your brand to be the thing the assistant surfaces and cites.

Gartner's marketing predictions (2025–2028)

Gartner "Top marketing predictions" (from the deck)
  • By 2026 — over a third of web content will be developed exclusively for AI / search-engine consumption.
  • By 2027 — mobile app usage will decrease by 25% as audiences shift to using AI assistants.
  • By 2027 — 85% of customer data will be collected from automated interactions or ones led by AI agents.
  • By 2028 — half of B2C companies using dynamic pricing algorithms will abandon them (to protect trust/brand differentiation).
  • By 2028 — 30% of digital marketers' paid social budgets will go to subscription-based channels.
  • By 2028 — mass "digital detoxing" will push CMOs to spend 70% of marketing budgets on offline channels to better engage consumers.

These are Gartner's stated predictions, reproduced at the slide's level — not forecasts of my own. The app-usage and offline-channel predictions tie directly back to the AI-assistant traffic trend above.

Areas of AI application in marketing

Where AI plugs into marketing (Statista AI Compass, 2025)

Content creation (generate quality content fast) · Personalized emails (tailor to the individual) · CLV forecasting (predict future customer value) · SEO optimization (optimise content for search) · Data analysis (enhanced analytics) · Sentiment analysis (gauge public sentiment toward brands).

AI-generated / AI-adjusted content: how to get cited by AI search

As AI Overviews reshape the SERP, the deck's biggest practical block is a checklist for making content that AI search engines can read, trust and cite. The framing: bots aren't human — they can't skim; they need structure, clarity and authority.

Optimising content for AI search (condensed from the deck)

Structure it for machines

  • Old SEO principles still apply: a clean site tree, well-ordered headings (H1 → H2 → H3 → H4), and detailed image alt-text.
  • Use FAQs / Q&A — the single most effective move. AI strongly prefers question-and-answer content because it mirrors how people use AI (they ask; it answers). Put questions as subheadings.
  • Open the article with a legend/summary of topics; use open tables and charts (built in the editor, not uploaded as images) so AI can actually read the data.
  • Clear categories and a structured "About" page — otherwise AI may fabricate facts about you.

Differentiate & answer directly

  • Stand out — say something with a unique angle; answer questions competitors under-serve.
  • AI prefers clarity — give direct answers without jargon; cover follow-up questions so AI links you to the whole query chain.

Build authority (E-E-A-T-style)

  • Earn brand mentions on authoritative sites (Wikipedia, gov, academic, news); model-training sources (Wikipedia, Reddit, Quora) carry extra weight.
  • Support claims with research and explicit citations (name the source, link it).
  • Prefer in-depth articles (600–700+ words) over thin 250-word ones; update frequently (both Google and AI favour freshness).
  • An authentic, first-person tone ("I checked", "I researched") reads as more trustworthy.
  • Include rich media with transcripts (captions/SRT so crawlers know what's in a video) and statistics in every article — data makes you a source AI wants to cite.
The through-line of this lecture

Voice search, visual search and AI search all point the same way: content should be structured as clear, question-shaped, well-cited answers. The "write content as answers to questions" idea from the voice-search slide is the same idea AI search rewards — that's why it shows up twice.

Exam takeaways

Direct MCQ themes from this lecture
  • FAST — Free, Ad-Supported Streaming TV: free + ad-supported + linear (scheduled, no on-demand choice). Don't confuse with AVOD (free + ads but on-demand). Direct MCQ — see 2024 Exam B1.
  • Linear Viewing — linear TV = same ad to everyone during a scheduled broadcast; CTV = different ads to different households while streaming. The linear-vs-on-demand contrast is a recurring MCQ theme, set up in Digital Marketing Lecture 2 and tested in 2024 Exam B1.
  • Voice-based search — people speak in full questions, so optimise with question-and-answer content ("create content as answers to questions"). Direct MCQ in 2024 Exam B1.
  • Monetization models — know SVOD / AVOD / TVOD / HVOD / FAST by who pays and how.
  • CTV metrics — VCR = video completion rate; CPCV = cost per completed view; incremental reach = net-new audience beyond linear.
  • AI search — structure content as Q&A, keep it clear and cited, build authority (brand mentions, freshness) so AI surfaces you.
Cross-links to the rest of the deck series

Digital Marketing Lecture 3 (POEM / media metrics / personas) · Digital Marketing Lecture 4 (campaign strategy / funnel) · Digital Marketing Lecture 5 (brand) · Digital Marketing Lecture 7 (content marketing / journey) · Digital Marketing Lecture 10 (data & measurement / OMTM / app metrics).