AI for Marketing Exam #1
What is a Customer Data Platform (CDP)?
A platform that gathers and unifies customer data from many sources to create one consistent, accessible profile. Provides real-time, persistent, and actionable customer profiles to any external system.
What is a CRM?
A system that centralizes customer interactions and data across sales, marketing, and support. Leads to better customer relationships, smoother communication, and improved sales efficiency.
What was the "AI Winter" (1970s-1990s)?
A time when people lost trust in AI because it was overhyped, computers were weak, and progress was slow.
A win-back email flow underperforms. Audit shows duplicate records and missing purchase events due to unmerged sources, so segments are wrong. Which specific risk from the slides does this most directly illustrate? A. Over-Reliance on Automation B. Data Quality & Silos C. Privacy & Compliance (GDPR/CCPA) D. Change Management
B. Data Quality & Silos
"Digital twins" are presented as a routine, uncontroversial substitute for real respondents in research. A. True B. False
B. False
CRMs are primarily used by marketing analytics teams and deliver real-time personalization by default. A. True B. False
B. False
NotebookLM compiles founder interviews, customer emails, and product photos; Gemini storyboards a launch film. Which direction best reflects the lectures' view of authentic brand storytelling? A. A catchy jingle emphasizing discounts to maximize recall B. Founder's struggle to solve a specific user pain, told through real customer moments C. Feature list narrated by an engineer for accuracy D. Stock footage + generic "innovation" tagline from a template
B. Founder's struggle to solve a specific user pain, told through real customer moments
You must stand up a quick concept test and have no time to hand-craft a full survey. Which AI use case best accelerates survey authoring? A. Digital twins of customers B. Generative AI features inside survey tools to draft/iterate questions C. Computer vision tagging of UGC D. Predictive pricing models
B. Generative AI features inside survey tools to draft/iterate questions
What are the key parts of a positioning statement?
Benefit + promise + proof. A strong statement is short, clear, and compelling.
What helped AI rise again in the 2000s?
Big data, better algorithms (like decision trees), and faster computers (like GPUs).
How does brand identity stay strong?
By using consistent visual and verbal cues.
n the Rapid Brand Positioning Challenge, which workflow best follows the assignment and leverages the tools correctly? A. Ask Gemini "What's the best brand strategy?" and accept the first output B. Generate a logo first; the rest can align later C. Use Deep Research to gather industry trend sources and top competitors; import into NotebookLM; synthesize each competitor's USP/voice/mission with citations; then make human decisions about your unique angle D. Ask NotebookLM to invent competitors if sources are thin
C. Use Deep Research to gather industry trend sources and top competitors; import into NotebookLM; synthesize each competitor's USP/voice/mission with citations; then make human decisions about your unique angle
What is Analytical CRM?
CRM that focuses on analyzing data to understand customer behavior and insights.
What is Collaborative CRM?
CRM that improves teamwork by sharing data across different departments.
What is Operational CRM?
CRM that streamlines business processes like salesforce automation and marketing automation.
Using Gemini, a team scrapes competitor claims and pricing to auto-plot a perceptual map: X = "Niche/Community → Mass Appeal," Y = "Budget → Premium." Existing brands occupy Mass/Premium, Mass/Budget, and Niche/Premium. What AI-informed positioning best fills white space? A. "Everyone's cheapest option with ads" B. "Prestige platform for cinephiles" C. "All live sports, highest price" D. "Community-first platform for fandoms at mid-tier pricing"
D. "Community-first platform for fandoms at mid-tier pricing"
A fitness startup feeds top competitor content into NotebookLM to summarize tone, then asks Gemini to draft its own voice guide. Which outcome best aligns with a coherent, humanized voice while avoiding mimicry? A. "Reserved, compliance-first" because it's safest legally B. "Whimsical, home-spun" to avoid any overlap with competitors C. "Minimalist, legalistic, documentarian" to sound serious D. "Energetic, empowering, competitive" expressed in original stories and consistent across channels
D. "Energetic, empowering, competitive" expressed in original stories and consistent across channels
You must "box" the output of a message-testing analysis so slides drop in cleanly. Which constraint set best reflects the guidance? A. "Be thorough." B. "Return insights however you like." C. "Write a narrative with examples." D. "Output a 5-row table: Theme | %Mentions (est.) | 2 Verbatims | Sentiment | 1-line Message. Max 120 words/row; no prose outside the table."
D. "Output a 5-row table: Theme | %Mentions (est.) | 2 Verbatims | Sentiment | 1-line Message. Max 120 words/row; no prose outside the table."
A retailer wants to unify web browsing, in-app events, and purchase history into a single, always-updating profile that can trigger real-time personalized journeys across email, push, and web. Which platform is the best fit? A. Tag manager B. CRM C. Data warehouse D. CDP
D. CDP
A subscription apparel brand sees a spike in cancellations after month 2. The team wants to identify at-risk customers and trigger targeted save offers before they churn. Which capability from the slides best fits? A. Next-best offer engine B. Generative SMS concierge for FAQs C. Dynamic content blocks in email D. Customer churn prediction models triggering retention campaigns
D. Customer churn prediction models triggering retention campaigns
Your CMO says, "Design CRM programs that keep customers and grow CLTV." Which initiative from the slides most directly aligns? A. Lower payment processing fees B. Reduce product returns through stricter policies C. Cut warehouse pick/pack time by 20% D. Encourage one more purchase or a larger basket
D. Encourage one more purchase or a larger basket
Your stakeholder calls an in-feed article "display." Based on the deck, what makes it native instead? A. It uses keyword targeting B. It runs only on social networks C. It is purchased via an ad network D. It seamlessly blends with surrounding content
D. It seamlessly blends with surrounding content
For free-text survey responses, which combo best delivers fast, reliable brand-equity insights per the deck's tech map? A. Computer Vision + Edge AI B. CV + Predictive Analytics only C. GANs + AGI D. NLP for parsing + Sentiment Analysis for polarity + LLM for summarization/QA
D. NLP for parsing + Sentiment Analysis for polarity + LLM for summarization/QA
A team fully automates lifecycle messaging and stops reviewing performance or creative quality. Engagement declines. Which slide-named risk most precisely applies? A. Broken last-click attribution B. Data residency constraints C. Vendor lock-in D. Over-Reliance on Automation = Weak marketing
D. Over-Reliance on Automation = Weak marketing
A fintech suspects "hidden fees" are a core pain. What AI-enabled approach best validates the USP? A. Ask AI to fabricate likely complaints to accelerate research B. Run a logo A/B test to see if people like the vibe C. Only check app-store star ratings D. Scrape Reddit/YouTube, have AI cluster pain points (hidden fees, sync issues), then review representative threads for context before deciding
D. Scrape Reddit/YouTube, have AI cluster pain points (hidden fees, sync issues), then review representative threads for context before deciding
What is the Lower/Bottom Funnel (Conversion) stage?
Its goal is to drive actions that lead to conversions and sales. Metrics: conversion rate, cost per acquisition (CPA), revenue, and ROAS.
What is the Mid Funnel (Consideration) stage?
Its goal is to nurture interest and engage potential customers. Metrics: click-through rates, engagement, time spent on site.
What is included in brand identity?
Logo, colors, typography, name, tagline, and messaging.
What was the "Deep Learning Revolution" in the 2010s?
Neural networks made big progress in speech, images, and language, and tech companies used AI in their products.
What is brand voice?
The personality of the brand, with clear traits and a consistent tone. It makes the brand sound human and keeps the message the same across all channels.
What are digital twins in marketing research?
They are AI-made avatars of customers that replace real responses, making research faster and cheaper.
How are brands using AI for emails?
They use Analytical AI to find the best time of day to send emails for better open rates and engagement.
How are retailers using AI in SMS chats?
They use Generative AI to answer questions, upsell, and drive sales with two-way SMS.
Why do marketers do market research?
To make better decisions, spot trends early, understand customers, talk like them, build what they want, and find competitor weaknesses.
Why do marketers use paid media?
To reach the right audience at the right time (precise targeting), track results with data (measurable ROI), quickly adjust to trends (flexibility), and boost visibility + sales (brand impact).
What is AI Democratization?
Tools and platforms are more accessible, enabling non-technical marketers, small businesses, and startups to leverage AI capabilities
According to the lecture, paid media complements owned & earned and can drive both immediate results and brand awareness. True False
True
Programmatic is defined in the deck as automated, data-driven ad buying in real time. True False
True
What are dynamic content & product recommendations?
Website or email content that updates based on user behavior.
What is Analytical AI?
AI that looks at data and finds patterns.
What is Generative AI?
AI that makes new things like text, pictures, or music (example: ChatGPT).
What is Predictive Analytics & Machine Learning?
AI that predicts consumer trends using past data. It improves targeting, budgeting, and return on investment (ROI).
What is Conversational AI?
AI that simulates human-like conversation (using NLP/LLMs). It provides 24/7 engagement, collects feedback, and qualifies leads.
What is Natural Language Processing (NLP)?
AI that understands and interprets text (like reviews, posts, or surveys).
How does AI help in social listening?
AI tools study billions of posts, images, and videos to find customer trends and opinions.
How does AI help in trend analysis?
AI uses web scraping and analytics to find new trends before they get big.
What is Computer Vision?
AI that identifies objects, logos, or faces in pictures and videos. It tracks logo use, brand visibility, and consumer reactions to visuals.
You have 30 minutes to brief the CMO on air-compressor reviews. Which prompt is most likely to produce decision-ready insights on time? A. Act as a research analyst. Task: extract 5 prioritized themes from attached reviews; for each, provide: (a) 2 verbatims, (b) est. share of mentions, (c) sentiment, (d) suggested marketing message. Use a professional tone; output a table only. B. List pros/cons. C. Write me a great report fast. D. Summarize reviews.
A. Act as a research analyst. Task: extract 5 prioritized themes from attached reviews; for each, provide: (a) 2 verbatims, (b) est. share of mentions, (c) sentiment, (d) suggested marketing message. Use a professional tone; output a table only.
An awareness campaign is judged a failure because 48-hour ROAS was low. What's the most accurate diagnosis using the deck's funnel? A. Misaligned metrics for the funnel stage B. Poor attribution model C. Creative fatigue reduced CTR D. Inadequate bids limited delivery
A. Misaligned metrics for the funnel stage
A strong prompt in this course framework should explicitly set Persona, Task, Context, Reference, and Tone to "put the model in a box." A. True B. False
A. True
Exporting high-value segments to Meta/Google to build lookalike audiences is an example of using customer data with Analytical AI. A. True B. False
A. True
Predictive analytics/ML can be used to forecast consumer trends from historical data—supporting proactive marketing decisions. A. True B. False
A. True
You trial multiple logo generators. Results are inconsistent. What is the best AI-aware path forward per the slides? A. Use AI outputs as mood-board drafts; then codify a brand book (logo, colors, type, usage) with human oversight B. Keep prompting until "perfect" appears C. Ship the first logo; consistency can come later D. Abandon identity work and focus only on features
A. Use AI outputs as mood-board drafts; then codify a brand book (logo, colors, type, usage) with human oversight
What is automated journey orchestration?
AI chooses the best channel (email, SMS, push) and timing to reach customers.
What is the future outlook of AI (2030 and beyond)?
AI everywhere in devices, humans working with AI, and possible Artificial General Intelligence (AGI).
What is customer churn prediction?
AI finds customers likely to leave and triggers retention campaigns to keep them.
Why use AI for CRM and retention?
AI helps predict customer churn, recommend next best offers, personalize content, and automate customer journeys.
How does AI support paid media?
AI helps with audience targeting (predictive segmentation), creative testing (dynamic optimization), campaign automation, performance prediction, cross-channel management, ad creation, and measuring results (attribution).
Ubiquitous AI
AI is expected to be embedded in most devices and applications (from autonomous vehicles to health
What is Artificial Intelligence (AI)?
AI is when computers or software act smart like humans, doing things such as learning, problem-solving, and decision-making.
What is Edge AI?
AI models deployed on devices (like smartphones or IoT devices) to reduce latency and enhance privacy
How does AI help in qualitative research (interviews/focus groups)?
AI platforms automate the work and use Analytical + Generative AI to dig into topics faster.
What is "Next Best Action/Offer"?
AI suggests personalized product offers or incentives in real time.
What is Agentic AI?
AI that can take actions on its own, like doing tasks for people.
What are native ads?
Ads that blend in with the content so they feel natural.
Who created the Turing Test and what does it measure?
Alan Turing in 1950; it tests if a machine can act like a human in conversation.
How does AI help in data analysis?
Analytical + Generative AI clean and analyze data quickly so marketers can focus on insights.
How does AI help with targeted ads?
Analytical AI builds high-value segments, then sends them to platforms like Meta or Google to target lookalike customers.
What is programmatic advertising?
Automated, data-driven ad buying in real time.
What are the 3 stages of the marketing funnel?
Awareness (Upper Funnel), Consideration (Mid Funnel), and Conversion (Bottom Funnel).
Which statement best captures the CRM vs. CDP distinction emphasized in the slides? A. CRM and CDP are interchangeable if you add automation B. CRM centers on direct interactions for sales/support; CDP unifies behavioral/transactional/demographic data across channels and often powers real-time personalization C. CRM is always real-time; CDP is batch only D. CRM centralizes multi-channel behavioral data; CDP focuses on sales tickets
B. CRM centers on direct interactions for sales/support; CDP unifies behavioral/transactional/demographic data across channels and often powers real-time personalization
Your first pass interacting with an LLM produced generic themes. What's the best next step per the prompt-iteration guidance? A. Switch to image analysis. B. Break the task into smaller prompts: one to cluster topics, one to quantify, one to draft messaging; add constraints (schema, word limits, evidence). C. Add humor to the prompt tone. D. Ask the model to "try harder."
B. Break the task into smaller prompts: one to cluster topics, one to quantify, one to draft messaging; add constraints (schema, word limits, evidence).
What is a brand?
It's more than a product. A brand is the emotions and perceptions people have, forming the identity of a company. It helps a company stand out, creates emotional impact, and builds loyalty and pricing power. Emotion + perception = brand.
After refreshing its messaging, a water brand uses Gemini to analyze thousands of social posts and reviews. "Confidence" and "safety" emerge as dominant themes; sales rise after a 7% price increase. What most plausibly explains the lift? A. Product-market fit improved because the filter got better (no evidence provided) B. Channel expansion drove higher conversion, not brand effects C. Brand-driven emotional impact increased willingness to pay, revealed by AI analysis D. AI inflated sentiment, creating a false positive
C. Brand-driven emotional impact increased willingness to pay, revealed by AI analysis
What is a brand story?
It's more than products — it creates emotional connections through authentic storytelling.
UGC shows customers posting shop-floor photos; leadership wants brand/logo visibility quantified. Which AI tech fits best? A. Conversational AI B. NLP C. Computer Vision D. Predictive ML only
C. Computer Vision
A team wants Gemini to produce positioning statements that follow the lecture structure. Which prompt outcome shows the structure is actually met? A. "Everyone deserves audio." B. "Our AI audio is future-ready." C. "For focus-driven remote workers who want fewer distractions during calls, we provide noise-adaptive earbuds that learn your environment—so meetings stay clear—backed by third-party lab tests." D. "We sell premium earbuds. The best."
C. "For focus-driven remote workers who want fewer distractions during calls, we provide noise-adaptive earbuds that learn your environment—so meetings stay clear—backed by third-party lab tests.
Gemini proposes four mission statements. Which aligns with the slide criteria (why + aligned with business goals + resonates with consumer need)? A. "Deliver KPI-positive boxes at scale. "B. "Food experiences for everyone, everywhere." C. "We exist to remove weeknight stress by making healthy cooking effortless." D. "Be the #1 brand by 2027."
C. "We exist to remove weeknight stress by making healthy cooking effortless."
Your brand's open rates are flat. You want a system that picks the best send time per person based on behavior. Which AI use case directly matches? A. Generative SMS assistant for Q&A B. Dynamic product recommendations C. Best-time-of-day email optimization D. Churn prediction with save offers
C. Best-time-of-day email optimization
NotebookLM clusters support tickets and Reddit threads for a skincare brand. Demographics are noisy; clusters reveal routines (e.g., "acne after workouts," "fragrance sensitivity"). What should the team do next? A. Freeze ICP until the logo is finalized B. Double down on age/location splits because they're easiest to target C. Rebuild ICP around behaviors/needs/pain points surfaced by AI clustering D. Ask Gemini to invent psychographics if data are sparse
C. Rebuild ICP around behaviors/needs/pain points surfaced by AI clustering
Which of the following is not listed as a key CRM feature in the slides? A. Analytics B. Data centralization C. Resource planning D. Automation
C. Resource planning
You're asked: "Which two problems of classic research does AI most directly improve when analyzing existing reviews within 30 minutes? A. Expensive & Sampling bias B. Expensive & Generalizability C. Slow & Manual D. Manual & Ethical risk
C. Slow & Manual
What is a USP (Unique Selling Proposition)?
It's what makes a brand different and valuable compared to competitors. AI isn't great at building a USP from scratch.
What are the 4 core elements of paid digital media?
Creative (what you show), Targeting (who you show it to), Bidding (how much you pay), Measurement (how you check success).
What is the Upper Funnel (Awareness) stage?
Its goal is to build brand recognition and reach a wide audience. Metrics: impressions, reach, brand lift.
What is Sentiment Analysis?
It's a type of AI that reads text (like a tweet, review, or survey answer) and figures out the emotion behind it: Positive → happy, excited, satisfied (ex: "I love this product!") Negative → angry, upset, unhappy (ex: "This product is terrible.") Neutral → no strong emotion (ex: "The product arrived yesterday.")
What is competitive positioning?
It's how a brand shows it is different from others, using audience, offering, and context. Perceptual maps (like the US watch brands map). are used to visualize this
What was Symbolic AI and Expert Systems?
Early AI that used rules to copy human knowledge into software.
What does defining a unique brand position mean?
Explaining the brand's "why," aligning with business goals, and meeting consumer needs.
What is a brand mission?
Explains the brand's "why," aligns with business goals, and connects with customer needs. It guides the brand's purpose and helps it stay meaningful to customers.
What is AGI or ASI?
Future AI that could think like or beyond humans — it does not exist yet.
What are the key AI trends in the 2020s?
Generative AI (like ChatGPT), Responsible AI (ethical use), Edge AI (on devices), and AI tools becoming easier for everyone.
How does AI help in survey research?
Generative AI makes writing surveys easier, faster, and less stressful.
What are the outcomes of using a CDP?
It removes data silos, ensures accurate personalization, and enables advanced segmentation and targeting.
What is Responsible AI?
Increasing focus on ethical considerations, bias mitigation, and explainability
Why does brand storytelling matter?
It builds trust, emotions, and relationships with customers.
What are the 3 main problems with research and insights?
It can be slow, expensive, and manual. AI makes it faster, cheaper, and automated.
What is the role of paid digital media in marketing?
It complements owned & earned media, drives quick results, and builds brand awareness.
What makes a good prompt, good?
Persona - Tell the AI who it is. Defined task - Make its goal(s) clear. Context - Add important context & detail. Reference - Share examples. Tone - Tell it how to respond.
What are the main risks of using CRM/CDP?
Poor data quality/silos, privacy and compliance issues (like GDPR, CCPA), over-reliance on automation, and challenges with change management.
