Because "ChatGPT Wrote This" Is Not a Flex
You've dabbled with AI. Maybe you asked ChatGPT to write a blog post, and it gave you something bland, lifeless, painfully written by a robot who has never experienced a single human emotion, that you immediately swore off the whole thing. Or maybe you've seen those cringe LinkedIn posts that start with "As an AI language model, I don't have personal opinions but..." and wondered if we've all collectively lost our minds.
The gap between “AI is amazing” and “AI is useless” is usually just the gap between a good prompt and a lazy one. Specificity matters. Context matters. Examples matter. Telling AI “write like you’re explaining this to a friend over coffee” changes everything. Feeding it your audience personas, recent reviews and asking for content pillars that will engage with them? Game-changer!
So before you write off AI as “not quite there yet,” ask yourself… are you giving it enough to work with? Are you treating it like a collaborator or a vending machine? Are you expecting magic from a single prompt, or are you putting in the effort to get something genuinely useful out?
As a Marketing Manager, I see the same critical mistake happening across the industry daily, treating AI as a final product rather than a force multiplier. The most common failure is using generative AI to write the final, customer-facing copy. Customers have become blind to this content, it doesn’t engage them and they ignore it. Some marketers are hitting “generate” and then “publish” without injecting all of their lived knowledge and learnt understanding of the industry and brand.
Another issue I commonly see, AI models are trained on historical data. It can only work from what has already been done. Using AI without your proprietary insights (sales call transcripts, customer support logs, seasonal trends specific to your niche) guarantees you will create a lagging, not leading, strategy.
Assuming the AI will magically sort messy CRM data or understand the emotional state of a lead from a single click. AI amplifies what you feed it. If you feed it rushed prompts, garbage data, or no brand guardrails, it will generate perfectly polished garbage at scale.
The winning approach isn’t to replace the marketer, it’s to use AI for drafts, clustering, and grunt work, while keeping a human in the loop for judgment, empathy, and strategic red-teaming.
Integrating AI into Your Marketing Plan
Marketers generally use AI across four core service types: research and data analysis, content creation and optimisation, audience targeting and personalisation, and performance automation and reporting. Most marketing departments trying to implement AI into their processes fail not because the models are weak, but because the underlying stack cannot feed them clean, connected, real-time data. You cannot bolt AI onto a broken foundation.
Before working on implementing AI, it’s important to complete a tech stack audit.
Step 1: Inventory Existing Systems
Understanding what applications and technology you have in your marketing department will help you understand where you have data stored, and if you’re looking to invest in AI, understanding what tools are best for your team.
Step 2: Assess Data Structure – Structured vs. Unstructured
When thinking about marketing data that will feed into an AI programme, most marketing data falls into two buckets.
- Structured Data (Easy for ML)
- Tabular, clean schema, numeric or categorical
- Think, purchase amount, email opens, campaign clicks, LTV tier
- Unstructured Data (Rich but Messy)
- Free text, images, voice, video
- Think, support tickets, social comments, call transcripts, and user-generated images.
Step 3: Ensure Customer Identity is Consistent
It’s not uncommon for customers to be fragmented across the CRM, Email Marketing System, Social Media, and anywhere else you have them stored. For your customers to reap the benefits of your new AI systems and processes, they need to become one complete person, with the same customer ID across each system.
Step 4: Evaluate API Readiness
Can your systems actually talk to each other? Which tools work with your current tech stack and are these APIs native or do you need to invest in APIs to help them talk to each other?
What’s Next for your AI in Marketing?
We’ve all had the experience when our manager has seen a new AI demo from a vendor that reached out to them on LinkedIn, and immediately, your manager wants autonomous execution. But skipping foundational levels is like building a skyscraper on sand. You can’t outsource maturity to software. Buying an “AI-powered” tool does not make you AI-mature any more than buying a stove makes you a chef.
Our model gives you an honest assessment of where you are today and a pragmatic roadmap to where you want to be.
| Level | Name | Decision Maker | Learning Loop | Typical Timeline |
| 0 | Manual + Rules | Human | None | Where most start |
| 1 | Assisted Intelligence | Human with AI suggestions | Weekly batch | 3–6 months |
| 2 | Augmented Intelligence | Human + AI (co-decision) | Daily | 6–12 months |
| 3 | Autonomous Execution | AI with human override | Real-time | 12–24 months |
| 4 | Adaptive Optimization | AI self-improving | Continuous | 24+ months |
Most B2B companies should target Level 2. Most B2C companies with high transaction volume can aim for Level 3 in specific channels. Level 4 is the research-stage which we see used mainly by the largest tech companies.
What’s Next for AI in Marketing?
From AI-driven ad campaigns to automated customer service, the possibilities are endless. However, with this potential comes the responsibility to use AI ethically and transparently, ensuring that it enhances rather than replaces human creativity and decision-making.
By embracing AI, marketers can unlock new levels of efficiency, creativity, and insight, driving growth and success in the digital age. Take advantage of this opportunity to question whether what you are doing now is working hard for your business and how marketing can make greater impacts.
Many of us are unsure as to which uses of artificial intelligence are actually useful, but often we are only looking at content generation and not beyond the many genuinely useful tasks in marketing strategy that AI can aid with.
AI Digital Marketing
Unlock the full potential of artificial intelligence in your marketing through SocialB’s structured AI training courses. Designed for marketers at every level, our curriculum delivers practical, hands-on learning tailored to your goals, whether you’re integrating AI into your strategy, optimising campaigns, or scaling content creation.
Learn directly from industry experts who provide clear, actionable guidance in interactive sessions. Each course is built to give you control over your learning journey, with flexible scheduling and a focus on the skills most relevant to your business.