Is your Team Stuck on the ChatGPT Basics? Unleashing Results (Part 2)

PART 02

Introduction

In the previous article, "Is your Team Stuck on the ChatGPT Basics? Unleashing Results," we discussed how many companies struggle to move past the basic use of ChatGPT. However, there's something beyond just using ChatGPT, and it's the promise of generative AI and the powerful tools like Claude, Gemini, Perplexity, CoPilot, and Anthropic. These tools offer wonderful chatbots that can provide tremendous insights and answers. While there are numerous videos and resources on prompt engineering, these tools are primarily designed for single-purpose tasks. You ask a question, get an answer, and move on, constantly writing your own prompts. This approach falls short of the true promise of generative AI.

The Promise of Generative AI

To unleash the power of generative AI, you need to move beyond using individual tools and figure out how to give substance to your workflows. In our daily, weekly, monthly, or quarterly routines, there are often a series of repetitive tasks that involve manual effort and multiple people. Identifying and cataloging these tasks is crucial to understanding their impact on your business.

Categorizing Tasks with the TEAR Model

Once you have a catalog of repetitive tasks, it's essential to categorize them using the TEAR model (Transform, Enhance, Assist, Revolutionize) - For more info see article: **Optimizing Workflows with the TEAR Model: A Quick Guide to Using AI for Business Productivity** This model helps you understand which tasks are suitable for AI assistance, which ones can enhance your business, and which ones have the potential to revolutionize or transform the way you operate.

Most companies start by focusing on the low-hanging fruits, the routine tasks that AI is perfectly suited for. These include scheduling, data entry, managing expense reports, email management, sales outreach, crafting custom proposals, answering customer questions, document management, inventory management, and a host of other activities that are often considered boring and drudgery. This is an excellent place to start, since it lets the company build it skills and comfort with AI.

Avoiding Common Pitfalls

As you explore automating these tasks, it's essential to avoid two common pitfalls. The first is thinking that you can write a ChatGPT or a slightly smarter version of it, known as GPTs. This approach is limited in its capabilities, difficult to integrate into broader company systems, and challenging to maintain and support.

The second pitfall is attempting custom programming. While this approach allows for complete control and integration, it can be problematic because the world of technology is changing rapidly. As your technology team shifts its focus to AI, all the other projects grind to a crawl as they learn to build a custom solutions, that frankly, will likely become obsolete.

Embracing Low-Code/No-Code Solutions

To truly harness the power of generative AI, you need a tool or system that allows for low-code or no-code development while providing robust tools for maintenance, scalability, and integration with other systems. One such solution is MiningStudio, which we'll assume you've chosen for the remainder of this article.

Task Evaluation and Scoring

The second step is to evaluate and score the repetitive tasks based on specific criteria. Here's a suggested scoring system:

Task Evaluation Criteria (Score each criteria from 1-5, 5 being highest)

  1. Frequency & Magnitude

    • 5 = Daily task taking over 2 hours

    • 4 = Daily task under 2 hours OR Weekly task over 4 hours

    • 3 = Weekly task 1-4 hours OR Monthly task over 8 hours

    • 2 = Monthly task 2-8 hours OR Quarterly task over 12 hours

    • 1 = Quarterly task under 12 hours OR Annual/Infrequent

  2. Number of steps involved

    • 5 = More than 10 steps

    • 4 = 7-10 steps

    • 3 = 5-6 steps

    • 2 = 3-4 steps

    • 1 = 1-2 steps

  3. Data integration requirements

    • 5 = Task requires data from multiple systems

    • 4 = Task requires data from 2-3 systems

    • 3 = Task requires data from 1 external system

    • 2 = Task uses internal data sources only

    • 1 = No data integration needed

  4. Level of human judgment required

    • 5 = Highly rules-based, no judgment needed

    • 4 = Some rules, minimal judgment

    • 3 = Moderate level of human judgment

    • 2 = High level of human judgment

    • 1 = Entirely based on human judgment/creativity

  5. Potential business impact

    • 5 = Critical task, high business impact

    • 4 = Important task, significant impact

    • 3 = Moderate impact

    • 2 = Low impact

    • 1 = Minimal/no business impact

Total the scores for each task. Higher scores (20-25) indicate the task is highly suitable for automation using a low-code/no-code AI workflow tool like MiningStudio. Lower scores (5-12) suggest the task may be better handled manually or using simpler AI assistants.

Additional Considerations

When selecting your first tasks to automate with MiningStudio, consider two additional criteria:

  1. Existing Automation Examples: Look for tasks that others have already automated successfully, as there's evidence they can be easily automated.

  2. Drudgery Factor: Consider the impact on the people performing the task. Tasks that are frequent and time-consuming may be perceived differently in terms of drudgery by the individuals involved. This "drudgery factor" is a subjective judgment call, but it's essential to choose tasks where people will feel positive about the outcome and the elimination of work.

While this decision involves some art and judgment, it's crucial to pay attention to how the automation process unfolds. Not every company has the luxury of engaging with innovation experts or change management professionals, but being mindful of how people adapt to these changes is essential.

Conclusion

In this article, we explored how to move beyond the basics of ChatGPT and unleash the power of generative AI by integrating it into your business workflows. We discussed the importance of cataloging and categorizing repetitive tasks, evaluating them based on specific criteria, and considering existing automation examples and the drudgery factor.

You can either embark on this journey alone, learning to use tools like MiningStudio and developing your staff's capabilities, or consider hiring an expert. Both approaches have proven successful, and the choice depends on your company's resources and needs. Regardless of the path you choose, embracing generative AI and automating repetitive tasks can unlock significant productivity gains and transform the way your business operates.

Need help? The Savvy Founder is a certified MindStudio Partner with certified Mindstudio AI experts. Get your Free AI Assessment Now

Previous
Previous

Stop Learning Prompts, Start Solving Problems: Why SMBs Should Rethink AI Adoption (Part 01)

Next
Next

Is your Team Stuck on the ChatGPT Basics? Unleashing Results (Part 01)