This post was originally featured in the November 7th, 2025 newsletter found here: INBOX INSIGHTS, November 5, 2025: 7 Ways to Get Started with AI, Deterministic vs. Probabilistic
7 Ways to Get Started With AI
Last month I was at an in-person event and I overheard someone saying, “we need to leverage AI across our entire customer journey,” while their partner was frantically scribbling notes, clearly trying to figure out what that actually meant in practice. They looked like they were about to cry.
After the session, she asked me, “My boss wants us to ‘do AI,’ but I don’t even know where to start. Do I need to use it for everything? Just some things? How do I know?”
Here’s what I told her (and what I’m telling you): You don’t need to boil the ocean. You just need to understand that generative AI does seven basic things. That’s it. Seven use cases that cover pretty much everything you’ve been hearing about. Once you know what those are, you can stop drowning in vague directives and start making actual decisions about where AI fits into your workflow.
Let me break it down.
The 7 Use Cases for Generative AI
1. Extraction
This is AI pulling specific information out of larger documents or datasets. Think of it as having an intern who can read through 500 pages of contracts and pull out every mention of pricing terms in about 30 seconds.
Where to start: Have your team identify their most tedious “find the needle in the haystack” tasks. Reading through customer feedback to find mentions of specific features? Pulling data points from reports? That’s extraction.
2. Classification
AI sorts things into categories. It’s like having someone organize your messy email inbox, but they can do it for customer inquiries, support tickets, leads, content types, or whatever you throw at them.
Where to start: Look at anything your team currently sorts manually. Support tickets that need routing to different departments? Customer feedback that needs tagging? Product reviews that need sentiment labels? Start there.
3. Summarization
Taking long-form content and condensing it down to the key points. (If you’ve ever asked ChatGPT to “sum this up,” you’ve used this.)
Where to start: Identify the meetings, reports, or documents where people consistently say, “I don’t have time to read all that.” Meeting notes that need to become action items? Industry reports that need executive summaries? Those are your candidates.
4. Rewriting
Taking existing content and transforming it for a different audience, tone, or format. It’s not creating from scratch; it’s adapting what you already have.
Where to start: Look at content you’re creating multiple versions of. Do you write the same update for executives and then again for your team in different language? Do you adapt blog posts for social media? That’s rewriting work AI can handle.
5. Synthesis
This is where AI combines information from multiple sources to create something new. It’s pulling together insights from different documents, datasets, or inputs to give you a unified view.
Where to start: Find the projects where someone is saying “I need to pull together information from marketing, sales, and customer service to see the full picture.” Quarterly business reviews? Competitive analysis? Those synthesis tasks eat up massive amounts of time.
6. Question Answering
AI acts like a really well-informed colleague who’s read all your documentation and can answer specific questions about it. This is your internal knowledge base on steroids.
Where to start: Think about the questions that get asked repeatedly. New employee onboarding questions? Policy clarifications? “Where did we document that decision?” If you’re constantly Slacking the same people for the same information, that’s a question-answering use case.
7. Generation
Creating new content from scratch based on prompts. This is the use case everyone jumps to first, but honestly, it’s often not where you should start.
Where to start: Look at high-volume, formulaic content creation. First drafts of standard email responses? Social media post variations? Product descriptions that follow a template? Those are generation opportunities.
Identifying Where to Start
If knowing the 7 use cases doesn’t help you with where to start, I’d suggest using the Trust Insights TRIPS Framework. TRIPS helps you figure out which tasks are screaming to be automated with AI. Here’s how it works:
The TRIPS Framework: Your AI Priority Filter
Time – How much time does this task consume? The more time something eats up, the better a candidate it is for AI. If your team is spending hours doing something every week, that’s a flashing neon sign.
Repetition – How repetitive is the task? AI loves repetitive work. Humans? Not so much. The more frequently you’re doing the exact same type of task over and over, the more sense it makes to hand it off.
Importance – How important is the task, and what’s the risk if it goes wrong? Here’s the counterintuitive part: high-importance, high-risk tasks need MORE human oversight. You want to start with tasks that are relatively low-stakes. If AI messes up a social media post draft, you can fix it. If it messes up a regulatory filing? Yeah, that’s not your starting point.
Pain – How much do you hate doing this task? Real talk: if your team groans every time they have to do something, that’s a prime AI candidate. And here’s why this matters beyond just morale: when you can show your team they’ll never have to do [insert soul-crushing task here] again, you get instant buy-in for your AI initiatives.
Sufficient Data – Do you have examples of how this task should be done? The more examples you have (templates, past work, documented processes), the better AI will perform. If you’re already using a template for something, AI should probably be doing it tomorrow.
Putting TRIPS Into Practice
List out everything your team is doing. Then score each task against TRIPS:
Task: Sorting incoming customer inquiries into categories
- Time: 2 hours per day ✓
- Repetition: Every. Single. Day. ✓
- Importance: Low risk (we can review before responding) ✓
- Pain: Everyone hates it ✓
- Sufficient Data: 6 months of categorized emails ✓
Result: This became test #1.
Task: Writing the quarterly executive summary
- Time: 8 hours once a quarter
- Repetition: Only quarterly ✗
- Importance: Very high, CEO reads this ✗
- Pain: Actually kind of satisfying to write
- Sufficient Data: Only 12 past examples ✗
Result: Not a good starting point for AI.
See how that works? Go through your whole list. The tasks that score high across multiple TRIPS criteria? Those become your testing priorities.
The Bottom Line
Your executives aren’t wrong for having big ideas about AI. But big ideas without a framework for execution just create chaos.
Use those seven use cases to understand what AI can do. Use TRIPS to figure out what AI should do for your team first. Then run small tests, measure what works, and scale from there.
You don’t need a grand strategy. You need to identify which repetitive, time-consuming, painful tasks are sitting in your team’s workflow right now and pick one to test this week.
Start there.
How are you choosing where to start with generative AI? Reply to this email or join our free Slack group, Analytics for Marketers.
– Katie Robbert, CEO
|
Need help with your marketing AI and analytics? |
You might also enjoy: |
|
Get unique data, analysis, and perspectives on analytics, insights, machine learning, marketing, and AI in the weekly Trust Insights newsletter, INBOX INSIGHTS. Subscribe now for free; new issues every Wednesday! |
Want to learn more about data, analytics, and insights? Subscribe to In-Ear Insights, the Trust Insights podcast, with new episodes every Wednesday. |
Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.
Newsletter
Podcast
