Brief Introduction
Welcome to AI for the Rest of Us. This newsletter exists for one reason: to give non-technical leaders the clarity and confidence to win with AI. This is our first post, and we will release an article every Tuesday morning at 9:30 AM. Without further ado, let’s dive in!
There are thousands of AI tools on the market, each promising to revolutionize your business. 99% of them are a waste of your time and money. Today we’ll go over a simple litmus test to spot the 1% that actually matter (it’s easier than you think).
As a former AI Engineer, I can tell you that spotting useless tools is easy once you stop falling for the shiny object the word "AI" has become. It starts by asking the right questions.
The Workflow Question
The biggest hidden cost of any new software is the friction of changing human behavior. A tool might be powerful, but if it requires your team to adopt an entirely new process, it will likely crumble to the ground. The best tools act as a superpower upgrade to a workflow your team already performs.
Here's how to gut-check it:
Integration: Does it plug directly into the software you already use (your CRM, Slack, Google Workspace, Microsoft 365)?
Streamlining: Does it replace a tedious manual step, or does it add a new, separate task? Always favor tools that replace and streamline.
Onboarding: Can a team member understand the core value in 15 minutes, or does it require hours of training? Time is money.
If a tool forces your team to leave their existing software and change their habits, it's starting with a major disadvantage. Once you're confident it fits the workflow, the next question is about risk.
The Data Question
For a business leader, this is the most important question you can ask. Mishandling company or customer data with an unvetted AI tool is a massive security and legal risk.
Many AI models can be trained on the data customers feed them. You would never want your sensitive company information or customer lists stored and used by a third-party model. Before proceeding, you need to know:
Data Policy: Check their terms of service. Do they train their models on your data? If a tool explicitly states that your data is never stored or used for training, that's a great sign.
Security Certifications: If a tool does store data, is it encrypted? Does the company have security certifications like SOC 2?
Compliance: If you handle customer data, does the tool comply with regulations like GDPR and CCPA?
Your company's policies are the ultimate authority. If a tool's data practices don't align with your internal rules, the risk falls on you. This is why many enterprises prefer tools like Microsoft Copilot, which offer enterprise-grade data controls and integrate directly into their existing, secure Microsoft applications.
Once the tool passes the security test, it's time for the final, critical question.
The ROI Question
If you can't state the value simply, it's either not valuable enough or you don't fully understand it. This question cuts through vague marketing promises and forces you to define the bottom-line business impact.
Think about the application from a full business perspective. Here’s some things to consider:
Specific Metrics: Does it save time, increase revenue, or cut costs? Be specific.
Measurability: How will you know if it's working? What key performance indicator (KPI) will change?
Total Cost of Ownership: Factor in not just the subscription price, but the time for implementation and training.
Here’s a little example for how a pitch would look for a tool that does initial screening for candidates on behalf of the company: "This tool will reduce the time our recruiters spend screening the initial 80% of unqualified resumes by half, allowing them to focus exclusively on speaking with top-tier candidates."
It’s clear, it’s measurable, and it delivers the value of the tool without adding fluff or non-sense. You don’t need to know the technicalities behind the product, you just have to explain the tangible outcomes.
Conclusion
AI is the booming industry, so it's easy to get sucked into anything with "AI" slapped on it. Honestly, many applications are just nicely dressed-up "wrappers" that use the powerful models built by major companies. You're paying for the workflow, not the core technology.
Be skeptical of buzzwords. If you run any potential tool through this 3-question test and it passes, you can be confident it's far more likely to save your company effort, time, and money.
P.S. - What's the biggest AI-related pressure you're feeling at work right now? Hit reply and let me know, I read every response! Email: [email protected]