Apple Podcasts  ·  Spotify  ·  Buzzsprout

Paper A

"With great power comes great responsibility": A meta-narrative review of ethical considerations and implications in the crossroads of AI and Marketing

Jens Christian Bekkestad, Christian Solvang — 2023

dissertation  ·   ·  read now

Key findings

• AI tools in marketing collect huge amounts of personal data — often without people truly knowing or agreeing to it. The review found that companies need to be upfront about what data they collect, get real consent from users, and give people a way to control their own information.
• AI-powered ad targeting and recommendation systems can be biased — meaning they can treat different groups of people unfairly, for example by showing certain job ads or loan offers only to some demographics. Fixing this requires actively checking algorithms for bias and correcting them.
• AI can subtly manipulate how consumers think and make decisions — for example by using personalized nudges or dark patterns that push people toward purchases they might not otherwise make. This threatens people's ability to make free choices.
• Companies using AI in marketing are under increasing pressure to behave responsibly (corporate social responsibility). The review finds that having a clear ethical framework built into how a company uses AI — not just following the law — is what separates responsible from irresponsible AI marketing.

Marketing implications

• If your team uses any AI tool that targets ads or personalizes content, audit it once a year for bias — check whether it's showing different messages to different demographic groups in ways that aren't fair or legal.
• Add a plain-language data consent notice to your sign-up forms or cookie banners that actually explains what your AI tools do with customer data. Generic legalese doesn't count as informed consent.
• Before launching an AI-driven campaign (chatbots, personalized offers, dynamic pricing), ask yourself: 'Is this nudging people in a way they'd be uncomfortable with if they knew?' If yes, redesign it.

Paper B

Implications of generative AI on small to medium sized e-commerce businesses

Aydin, Justina — 2025

dissertation  ·   ·  use cautiously

Key findings

• The most common use of generative AI among these small e-commerce businesses was writing and creating marketing content — things like product descriptions, ads, and social media posts. Personalization (tailoring messages to individual customers) was only a moderate benefit.
• In product development, AI helped speed up early-stage tasks like mockups and prototypes, getting ideas to market faster. But it did not meaningfully help companies come up with truly new or breakthrough products.
• Generative AI made it cheaper and easier for new small businesses to enter the market — lowering the barriers of getting started. However, it did not dramatically help small businesses compete against much larger companies.
• Companies that integrated AI more deeply into their marketing saw stronger results across content creation, cost savings, and overall marketing performance — suggesting that going all-in matters more than dabbling.

Marketing implications

• If you run a small online store and haven't tried AI yet, start with content creation — writing product descriptions, social posts, and ad copy. That's where small businesses in this study got the most obvious payoff.
• Don't expect AI to replace your product innovation process. Use it to speed up early prototypes and mockups, but the creative breakthroughs still need human thinking.
• If you're going to use AI, actually integrate it into your workflow rather than using it occasionally. The businesses that committed more deeply saw better results across the board.

Paper C

A Structured Large Language Model Approach to Market Intelligence and Creative Content Generation

Rajitha V, Kavyashree K L — 2026 — International Journal For Multidisciplinary Research

peer reviewed journal article  ·   ·  test this week

Key findings

• The system cut the time analysts spent manually reviewing and summarizing marketing data by 70%, according to the authors' own testing — though this was not measured against an independent control group.
• By feeding structured app store data and D2C metrics into Google Gemini 2.5 Flash, the system automatically produced confidence-scored marketing insights, ranked by statistical relevance, without a human analyst having to interpret the raw numbers.
• The same pipeline that analyzed performance data could also write marketing copy — ad headlines, SEO meta descriptions, and product content — creating a single workflow that goes from raw data to ready-to-use creative assets.
• The system could calculate key campaign health metrics like Cost per Acquisition (CAC) and Return on Ad Spend (ROAS) automatically, giving marketers a real-time funnel view alongside the AI-generated copy.

Marketing implications

• If your team spends hours pulling data from multiple app stores and campaign dashboards before writing a brief or report, this paper shows that an off-the-shelf LLM (like Gemini) can be set up to do that aggregation and first-draft analysis automatically — worth piloting if you have a developer on the team.
• Next time you need to produce ad headlines or SEO meta descriptions at scale (say, for 50 app categories or product lines), try feeding your structured performance data directly into Gemini with a schema-constrained prompt — the paper suggests this can produce consistent, ranked outputs without a copywriter starting from scratch.
• For D2C brands tracking CAC and ROAS across platforms, this paper demonstrates a blueprint for building a single dashboard that auto-calculates those metrics and generates a plain-language summary — reducing the analyst bottleneck between data and decision.

Apple Podcasts  ·  Spotify  ·  Buzzsprout