Home Author
Author

Miguel Guhlin

Transforming teaching, learning and leadership through the strategic application of technology has been Miguel Guhlin’s motto. Learn more about his work online at blog.tcea.org, and mguhlin.org. Catch him on Mastodon at @mguhlin@zirk.us

Areas of interest flow from his experiences as a district technology administrator, regional education specialist, and classroom educator in bilingual/ESL situations. Learn more about his credentials online at mguhlin.org.

AI/VRCommunication

Become an AI-Powered Newsletter Architect

by Miguel Guhlin
written by Miguel Guhlin

Another Sunday evening deadline pops up. You have seventeen browser tabs open as you try to put a newsletter together that goes out the next morning. Each one holds a promising article about AI in education, new literacy strategies, or that math approach everyone is talking about. Turning that collection of links into something useful should be easy, but it will take hours you do not have.

Sound familiar? Right now, people process roughly 300% more digital content than in 2019. The challenge isn’t finding content (e.g. Critical Thinking, Education Research). It’s knowing what to do with it. In this blog entry, you’ll see how to build an AI-powered workflow that gets your content curation choices to readers faster, without sacrificing the human touch that makes your communication worth reading. What’s more, you will get access to the custom instructions for building your own AI-powered newsletter.

What Is an AI-Powered Newsletter Architect?

An AI-Powered Newsletter Architect isn’t someone who lets AI write for them. It’s someone who designs a repeatable system. You make the editorial decisions, letting a Gen AI chatbot handle the mechanical work. You are the architect. Gen AI is your construction crew working off the plans you organized. This distinction protects your voice, your accuracy, and your professional judgment.

Step 1: Collect Content in One Place

Before you can curate anything, you need a home for everything you find. Links piling up in browser tabs and random notes apps with no structure is what causes that Sunday evening panic.

Raindrop.io solves this. It’s a free bookmarking and curation tool that lets you save articles, videos, and links into organized collections. You can tag items, add notes, and share collections publicly.

My Learning Loop series shows what a curated Raindrop collection looks like when it’s ready to feed into a newsletter workflow. Once your content lives in one place, the rest of the process moves fast.

In fact, I often try to identify themes in curated content. Those themes (e.g. Health, Environmental Impact, Critical Thinking) set the focus of the newsletter.

Step 2: Build Your Voice into the System

Your favorite newsletter has a recognizable voice. When you build AI-powered workflows, you encode your communication style into the system itself. This means setting up a BoodleBox Bot, Gemini Gem, ChatGPT Custom GPT, or a saved Claude prompt set that enforces your tone, your recurring sections, and your formatting. For example, here is an excerpt from the custom instructions I provided for my The Triple Boot newsletter Gem:

Your voice is “San Antonio warm”—efficient, witty, and communal. You are a “learning-in-public” guide who prioritizes open-source (FOSS) tools and cross-platform workflows. You value technical truth and digital sovereignty over marketing hype.

You are not prompting AI for one-off responses. You are building a set of custom instructions that define the consistent template for every issue generated. My Field Notes newsletter uses a set of custom instructions that enforce structure across every issue. I love the tag line that a bit of back and forth with the AI resulted in: “Coaching: Because ‘bless your heart’ isn’t a feedback model.” I couldn’t stop laughing for a few minutes.

Note that in the example above, Gemini’s Personal Intelligence works in the fact that I love to eat at Thousand Oaks Cafe in San Antonio, as well as my goal of 100 push-ups per day. You can adapt that approach for your own newsletter. For a look at how this logic extends to other content types, From Conversations to Presentations walks through a similar BoodleBox workflow.

Step 3: Practice Cognitive Curation

Most articles and reports weren’t written for busy educators who need to implement something by Monday morning. Cognitive curation is the practice of extracting what’s actionable from complex sources and rewriting it for your specific audience. The process is straightforward:

  1. Feed a dense article or report into your AI workflow
  2. Ask it to extract key findings in plain terms (aim for middle school readability)
  3. Request specific implementation steps
  4. Add your own editorial judgment before you publish

The AI summarizes. You decide what’s worth sharing and why. The Triple Boot series is a good example of this approach in action.

Step 4: Add Visuals and Close the Loop

People scan articles for quick information nuggets and a wall of text can lose readers fast. AI image generators like Adobe Firefly, Canva’s AI image generator, and Google’s Nano Banana Pro can produce polished hero images and infographics quickly. In the past, building an infographic meant opening Canva and starting from scratch. Now, a well-written prompt does the job.

Once you have your draft and visuals, run a quick feedback loop before you publish:

  1. Check facts against original sources
  2. Check your links
  3. Adjust tone to match your voice
  4. Add one or two personal observations only you can provide
  5. Publish

This loop protects you from the two biggest AI risks: inaccuracy and generic writing. For example, when creating images, make sure you provide the Gen AI with all the information it needs. Otherwise, it will go out on the web to fill in the gaps. This image below is accurate:

But if you had seen the original, you would realize everything below the $d=0.40 was WRONG or made up:

What’s worse, some of the instructional strategies mentioned do not exist. Why did this happen? I did NOT list the instructional strategies I wanted to include. I left that up to the Gen AI (remember, this was a “for fun” example to illustrate this point). Make sure you set your Gen AI tool up for success.

The Time Difference

Here’s what this workflow saves you. You won’t realize this the first two or three times, but afterwards, you will. I can generate a newsletter now by pasting in 3-5 links, then go in to clean it up. The process takes me a fraction of the time, and I could (don’t tell my boss) generate one every day.

StepTraditionalAI-Powered
Collect articles with Raindrop.io45 minutesTwo minutes
Read and take notes within Raindrop’s interface90 minutesEight minutes
Write draft60 minutesIncluded above
Find or create images30 minutesFive minutes
Format and edit45 minutesFive minutes
Total270 minutes20 minutes

That’s an 80% reduction…what else could you do with that time? I’d like to think that overall quality improves because you spend your energy on judgment, not the mechanics of newsletter design.

Give It a Go

Why don’t you give it a go? Here’s what your first four weeks might look like:

  • Week 1: Audit your current process. Track how long content creation takes and note where you slow down.
  • Week 2: Set up Raindrop.io. Create collections for your main topics and spend one week saving content there instead of in browser tabs.
  • Week 3: Build your first AI template. Test it with three to five pieces of content and revise based on what works.
  • Week 4: Add visuals and establish your loop. Document your workflow and measure your time savings.

My guess? You will be ready to go a lot sooner since I’m giving you access to all my custom instructions for various newsletter makers. Of course, not my Learning Loop series, but the others I made for you to explore. What’s more, I can easily see using vibe-coding features to generate web-based, one page newsletters in HTML format for easy sharing on an old-fashioned web server (without WordPress or content management systems).

Sustainable Communication Made Easy

You know, I sure wish I had enjoyed access to a quick way to create newsletters back in the day besides Aldus Pagemaker or Quark Express with my students. With AI-powered newsletter makers, I could have celebrated student writing while combining it with web-based content. The juxtaposition would have resulted in powerful creations without all the significant overhead.

Creating sustainable practices means aggressively managing your workload and avoiding burnout. The older I get, the easier it is to feel that burnout coming close. When you build systems that work for you, you create space for what matters most without the burnout. In the future, the question won’t be whether to use AI in your communication workflow. Rather, it will be whether you’ll be intentional about how you design it and then implement it.

0 comments
0 FacebookTwitterPinterestLinkedinEmail
AI/VRArtificial IntelligenceEducational TrendsLeadership

Does Your District Have an AI Plan, or Just an AI Problem?

by Miguel Guhlin
written by Miguel Guhlin

Texas has 5.5 million K-12 students (2024-2025). During the 2024-25 school year, 85% of teachers and 86% of students nationally used Gen AI, and 52% of Texas teachers used AI tools specifically for math instruction. These stats are a signal that educators are curious, motivated, and moving on AI adoption. The adoption is real, and it is happening in classrooms now.

Did You Know? You can join the TCEA Community’s “All About AI” and get regular updates about Gen AI, how it affects education, nonprofits, and businesses via the AI Class Notes publication. Note: To get access to the Generative AI Adoption Checklist featured in this blog entry, you will need the password in the Comments section of this Community post on 4/10/2026 (Issue #2). What could you have done with this information a week ago?

The opportunity that creates is significant. 31 states have already issued formal K-12 AI policy, which means there are frameworks to learn from, adapt, and improve on. Districts that build thoughtful guidance now get ahead of the confusion rather than reacting to it later.

An Interactive Roadmap

Before we get too far along on the conversation, here’s an interactive roadmap (short version) you may want to explore to see the journey your stakeholder committee may need to embark on. It can help you get ahead of the confusion.

TCEA · Stakeholder Committee Tool
AI Adoption Roadmap for Districts and Campuses
Overall Readiness
Not yet started
Phase One · Foundation Policy, integrity, data
Phase Two · Practice Capacity, outcomes
Phase Three · Extension Equity, families, workforce
Stage One
Establish Clear, Permission-Based Guidance
Key Question Does your policy build responsible use, or just document what is forbidden?
Select a maturity level above. Your choice saves to this browser.
Where to Focus Next
Start with Stage One. Review your current policy against the checklist and name the gaps before rushing to fix them.
Read the post

So, how did you do? What’s your campus’/district’s overall readiness? Now that you have a better picture, continue reading.

Closing the Gap

The gap most districts face is not a lack of enthusiasm. Rather, it is the space between teachers who are already using Gen AI and institutions that have not yet caught up with clear expectations, data guardrails, and professional development to match.

If my superintendent asked in March. “Where are we on AI?” I would have had something cooking right away, bringing stakeholders together. It’s like when my superintendent said, “Let’s roll out BYOD at the high school.”

Sure, you can roll something out fast, but are you rolling it out the right way? The right way is more than a policy guide with a title, a table of contents, and a lot of words. It has to be built on community wisdom of the people in your space.

That gap is one you can bridge (unless you, or more likely school leaders sensitive to politics, are afraid of conversations).

A Checklist to Deepen the Conversation

The interactive Generative AI Adoption Checklist for K-16 Educational Institutions gives you a structured way to see exactly where you stand. Only 45% of principals nationally report having any school or district AI policy in place, and more than one in five Texas teachers said their district provided no training on the tools they were already using. The checklist covers 46 items across eight areas, works entirely in your browser, and saves your progress automatically. No account. No download. No license.

Note: To access the Checklist, you will need a password available via the Community post on 4/10/2026 (Issue #2) comments.

What the Checklist Covers

The eight sections map to the decisions that actually keep CTOs and curriculum directors up at night. They are not theoretical. Each one connects to a real gap that tends to appear when districts try to move from “we should do something about AI” to “here is what we expect.”

SectionCore Question
Establish Clear GuidanceDoes your policy build responsible use, or just document what’s forbidden?
Address Academic IntegrityDo students know exactly what is and isn’t allowed in each assignment?
Protect Student DataDo staff know what student information cannot enter any Gen AI tool?
Build Educator CapacityAre educators being equipped to make instructional decisions, or just taught to use tools?
Anchor AI in LearningWould this use of Gen AI produce better-prepared learners, or shortcuts?
Prioritize Equity and AccessWho benefits from your adoption, and who gets left further behind?
Engage FamiliesDo families feel informed and included, or surprised by what they eventually learn?
Bridge K-12 to WorkforceAre your students graduating into environments where their Gen AI skills will hold up?

Use this to frame your review of your campus or district AI plan or CIP/DIP (Improvement plans for campus/district) plan components. The question I ask most often is the one in section five: does this use of Gen AI deepen learning, or replace it? Most plans never get to that question. You can see this question pop up in teaching and learning, too:

How often do we get past Surface Learning strategies to Deep and/or Transfer Learning?

The answer for both questions is a simple “seldom.”

How It Works

Open the checklist in any browser. Check off items as you complete or confirm them. Your progress saves automatically to your browser, so you can return to it across multiple sessions without losing your place.

The sidebar shows your progress by section, and a progress bar at the top tracks your overall completion across all 46 items. Each section ends with a key question designed to surface honest conversation, not just box-checking.

There are no trick items. Each one reflects a decision your institution either has made, needs to make, or has been avoiding.

A Starting Point, Not a Finish Line

The checklist includes a Quick-Start Action Plan at the bottom. It is worth reading before you dive in, because it reframes the tool from a one-time audit to a staged process.

The “this week” item is the most useful one: review your current policy against the checklist and identify gaps. Avoid trying to fix them right away. Instead, simply name them. That alone is a conversation worth having with your team, your board, or your campus leadership.

Most organizations will find gaps in more than one section, and that’s OK for now. If it is still your situation in three to four months, you are moving too slow.

What This Is Not

This checklist is NOT a compliance document. One reason is that tt does not cite specific legislation, TEKS, etc. Privacy and data protection items in section three are written with FERPA and student records in mind. That aside, it does not tell you which Gen AI tools to approve or make the decisions for you. The intent is to assist you and a committee of stakeholders in making those decisions.

The goal is to give you a structured way to see what you have, what you are missing, and where the real risks are. It does this in the absence of state guidance. This checklist assists you in asking tough questions and getting answers from stakeholders. It’s important to do this before people knock on your door asking, “Why haven’t you done anything?” That’s one place no leader wants to be.

Open the Generative AI Adoption Checklist and spend 15 minutes on section one. See what you actually have in place. Note: To access the Checklist, you will need a password available via the Community post on 4/10/2026 (Issue #2) comments.

0 comments
0 FacebookTwitterPinterestLinkedinEmail
iPads
AI/VRGood Teaching

TCEA Responds: Navigating Resistance, Part 2

by Miguel Guhlin
written by Miguel Guhlin

In Part 1 of this series, you explored the “Secret Sauce” of the ARC Protocol. In it, you learned about the neuroscience of psychological safety, the power of the Crucial Conversations‘ authors’ AMPP technique (Ask, Mirror, Paraphrase, Prime) as applied in instructional coaching. You also saw why validating a teacher’s perspective is the essential first step toward instructional change.

AMPP approach is a Crucial Conversations/Confrontations concept that I’ve relied on often in my own work. Be sure to learn more about it in the book links included in this blog entry.

As any instructional leader knows, theory is one thing. The messy reality of a Tuesday morning in a middle school hallway is another. When you’re met with “I don’t have time” or “My students aren’t ready,” you need more than a framework. What you need is a script. You need the exact words that will transform that tangle of dark, knotted resistance into a smooth, golden braid of collaborative action.

In this blog entry, you will see how to put the ARC Protocol to work. Below are five common scenarios you might encounter as an instructional leader, followed by the specific application of Acknowledge, Reframe, and Commit. Each is enhanced with the AMPP technique for deeper psychological safety.

Remember the core principles:

  • A – Acknowledge: Lower the emotional temperature first (Crucial Conversations: “Make it Safe”)
  • R – Reframe: Stop arguing about teacher preference and start talking about student data (HESIS)
  • C – Commit: Don’t leave without a date on the calendar (TCEA EIIR)

Let’s take a look at specific scenarios through this lens.

Scenario 1: The “I Don’t Have Time” for AI Architecture

The Situation: A veteran teacher is resistant to using the AI-powered newsletter architecture, claiming they already spend too much time on parent communication.

A – Acknowledge to Make it Safe (using AMPP):

  • Ask: “How much time are you spending on parent newsletters right now?”
  • Mirror: “I hear the exhaustion in your voice when you talk about the weekend workload.”
  • Paraphrase: “It sounds like you feel that adding a new digital tool is just one more burden on an already overflowing plate.”
  • Prime: “Is it that you’re worried the setup time won’t actually pay off in the long run?”

R – Reframe the Narrative: “What if we looked at this not as ‘one more thing,’ but as a way to reclaim your Sunday evenings? By using the AI architecture, we shift from ‘writing from scratch’ to ‘editing for impact.’ We are moving from a high-labor task to a high-leverage one, ensuring your hard work is actually seen by parents in a mobile-friendly format.”

C – Commit to a Micro-Step: “Let’s just take your existing bullet points for next week and run them through the ‘Mobile-First’ prompt together for five minutes right now. We won’t even send it—just see if the output actually saves you the drafting time we’re hoping for.”

Scenario 2: The “Performance Anxiety” regarding Video Coaching

The Situation: A teacher is nervous about recording their lesson, fearing it will be used for evaluation rather than growth.

A – Acknowledge to Make it Safe (using AMPP):

  • Mirror: “It is completely normal to feel vulnerable when a camera is in the room. Most of us feel like we’re under a microscope when we see ourselves on screen.”
  • Ask: “You want to make sure this stays between us and isn’t part of your formal file, right?”
  • Paraphrase: “It sounds like your biggest concern is that this might be used against you rather than to help you grow.”

R – Reframe the Narrative: “Let’s shift the camera’s focus. Video coaching has an effect size of 0.99. It’s one of the most powerful tools we have because it provides an ‘objective mirror.’ This isn’t about evaluating you but rather it’s about capturing the students’ reactions to the ‘Heavy Hitter’ strategy we tried. The video is just an extra set of eyes to help us see what we miss while we’re busy teaching.”

C – Commit to a Micro-Step: “Record just a 10-minute segment of your direct instruction tomorrow. You don’t have to show it to me. Just watch it yourself and pick one thing you noticed about student engagement. We can talk about that one observation next time we meet.”

Scenario 3: The “Students Aren’t Ready” for Deep Learning

The Situation: A middle school science teacher believes their students lack the discipline for Reciprocal Teaching (d=0.74) and wants to stick to lecture.

A – Acknowledge to Make it Safe (using AMPP):

  • Mirror: “Relinquishing control to middle schoolers can feel like inviting chaos, especially with complex content.”
  • Paraphrase: “It sounds like you’re worried that if we move away from the lecture, the classroom management will fall apart and the learning will stop.”
  • Prime: “Have you tried structured group work before and had it go sideways?”

R – Reframe the Narrative: “Instead of seeing it as ‘giving up control,’ think of it as ‘assigning agency.’ By giving them specific roles (Predictor, Questioner, Clarifier, Summarizer), we are actually providing more structure for their thinking than a standard lecture provides. We are moving from Surface Learning to Deep Learning, which is where the real retention happens. Reciprocal Teaching has an effect size of 0.74. That’s nearly two years of growth in one year.”

C – Commit to a Micro-Step: “Try the ‘Fab Four’ roles with just one paragraph of the text tomorrow. Just one. I’ll even come in and we can do it as a ‘fishbowl’ where you and I model the roles for the students first so they see exactly what the expectations are.”

Scenario 4: The “Colorblind” Approach to Equity & UDL

The Situation: A teacher claims they “treat everyone the same” and doesn’t see the need to adjust lessons for Universal Design for Learning (UDL).

A – Acknowledge to Make it Safe (using AMPP):

  • Mirror: “I know your heart is in the right place and that you care deeply about being fair to every student in this room.”
  • Paraphrase: “You want to ensure that no one gets ‘special treatment’ that might lower the expectations for the class.”
  • Ask: “What does fairness look like to you in your classroom?”

R – Reframe the Narrative: “Fairness doesn’t always mean ‘the same.’ If we look at it through the UDL lens, we’re not changing the goal; we’re just adding multiple paths to get there. It’s about removing the barriers (e.g. heavy reading loads or writing anxiety) that have nothing to do with the actual science content you’re trying to teach. You are removing an unnecessary barrier.”

C – Commit to a Micro-Step: “Let’s look at your next assessment. Can we offer just one alternative, like a verbal explanation or a labeled diagram, for students who struggle with the heavy writing load? We can see if it gives us a more accurate picture of what they actually know about the science.”

Learn more about TCEA’s Instructional Coaching Certification

Scenario 5: The “Data is Personal” Resistance

The Situation: A teacher becomes defensive when student data shows a low growth rate, blaming the students’ lack of effort.

A – Acknowledge to Make it Safe (using AMPP):

  • Mirror: “It is incredibly frustrating when you pour your energy into teaching and the scores don’t reflect that effort.”
  • Ask: “It feels like a personal critique of your hard work, doesn’t it?”
  • Paraphrase: “You’re saying the students just aren’t putting in the effort they need to succeed.”

R – Reframe the Narrative: “Let’s look at the data as a ‘GPS’ rather than a ‘Grade.’ It’s not saying you didn’t teach it; it’s just telling us that the current strategy hasn’t landed yet. This is our cue to try a ‘Heavy Hitter’ like Spaced Practice (d=0.65) or Retrieval Practice (d=0.49) to see if we can make the learning stick longer. The data isn’t judging you—it’s giving us a roadmap for what to try next.”

C – Commit to a Micro-Step: “Pick one specific standard where the gap is widest. Let’s try one 5-minute ‘Brain Dump’ or retrieval activity at the start of class every day this week. We’ll check the exit tickets on Friday to see if that small shift moves the needle.”

Wrapping It Up

Each of these scenarios follows the same pattern of emotions before logic. The AMPP technique in the Acknowledge phase ensures that the teacher feels heard before we ever mention effect sizes or instructional strategies. Only when emotions are acknowledged can you focus on reframing, then obtaining a meaningful commitment.

Your Challenge: Which of these five scenarios sounds most like a conversation you’ve been avoiding? Pick one then try writing out your AMPP statements. Practice them out loud. Then go have the conversation. One ARC at a time.

Read the Full Series:

0 comments
0 FacebookTwitterPinterestLinkedinEmail
Equity, Diversity, InclusionGood TeachingInstructional Coaching

Stop Fixing Students: Coaching for Equity through Universal Design

by Miguel Guhlin
written by Miguel Guhlin

Coaching for every learner requires moving from reactive differentiation to proactive Universal Design for Learning (UDL). By focusing on high-effect strategies like Jigsaw (0.92 effect size), you help teachers anticipate variability and remove learning barriers before instruction begins.

Note: This blog entry is based on my TCEA 2026 Convention and Exposition presentation, Coaching for Every Learner.

The Gap Between Belief and Reality

You’ve been there. You stand in a professional development session and nod your head because you believe, deep down, that every student deserves access to rigorous learning. Then you walk into a classroom and see a different reality. You see teachers struggling to keep up with 30 students who don’t fit the mythical “average.” That average is what many plan for, but that can be problematic.

In fact, Zaretta Hammond warns about the trap of magical thinking. This is the dangerous belief that if a teacher simply follows the steps of a strategy with fidelity, learning automatically happens. But we have to remember that compliance is not engagement. Your job as an instructional coach is to assist teachers in looking past the quiet, orderly room and see the actual cognitive processing taking place (or not taking place). That’s why tools like the SOLO Taxonomy are so helpful, as opposed to tools like Bloom’s Taxonomy.

How the SOLO Taxonomy levels correspond to phases of learning, and heavy hitter instructional strategies.

As Hammond puts it: “Only the learner learns.” You can’t do the learning for them. But you can design an environment where learning becomes possible for everyone.

Reframe the Problem: UDL Is Architecture

Work in a classroom where class size limits have been waived, and you are aware of the uncomfortable reality. It is impossible to differentiate for 30 unique humans if you’re designing for the “average.” The average student is a statistical myth, a point on a bell curve that represents no one in your actual classroom.

When you coach, you must first validate the stress teachers feel. Then pivot the conversation from coaching to fix the child to coaching to fix the barrier.

Consider the difference:

  • Reactive approach: Building a ramp after the building is finished because someone in a wheelchair showed up.
  • Proactive approach: Designing a building with an accessible entrance from day one.

UDL is the architecture of the lesson. It means anticipating that some students will need audio, some will need visuals, and some will need movement. This is true, even before you start teaching. Keep this in mind:

Firm goals, flexible means.

The standard stays the same (everyone analyzes the theme of betrayal). The method is flexible (essay, podcast, debate, or graphic novel).

The Research: High-Effect Equity Strategies

Not all strategies are created equal. Here’s what the Visible Learning research tells us about impact:

StrategyEffect SizeLearning Impact
Jigsaw Method0.92Considerably accelerates learning
Reciprocal Teaching0.74Students own the learning process
Transfer Strategies0.75Applying learning to novel situations
Differentiation0.58Effective, but often reactive

The threshold for “accelerating” learning is 0.40. Notice that Jigsaw doubles that. Compare it to differentiation:

Strategy Spotlight: The Power of Jigsaw (0.92)

The Jigsaw Method outperforms teacher-led differentiation because it creates interdependence. In a lecture, a student can hide. In a Jigsaw, every student holds a piece of the puzzle. If one student doesn’t teach their piece, the group fails. This makes every student valuable, not only those who are the high achieving students.

The Jigsaw also naturally embeds UDL principles:

  • Engagement: Students have a clear role and purpose.
  • Representation: Content is chunked and distributed.
  • Action and Expression: Students must teach, not just receive.

When you coach teachers toward Jigsaw, you’re not adding to their workload; you’re actually reducing the amount of teacher talk time while increasing student cognitive engagement.

Upgrade “Turn and Talk” to Cognitive Apprenticeship

“Unstructured ‘Turn and Talk,’ says Zaretta Hammond, “is often the weakest instructional move.” Without protocols, it can actually reinforce gaps if students lack background knowledge or feel unsafe speaking up.

Your coaching move is to help teachers upgrade unstructured talk into structured protocols, or discourse. Hammond refers to this as “Cognitive Apprenticeship.”

By Grade Band:

  1. Elementary (VOICE): Use sentence stems like “I think… because…” to provide training wheels for thought.
  2. Middle School (ARGUE): Channel natural adolescent energy into analyzing, researching, and evaluating evidence. Give them something to push against.
  3. High School (RHETORIC): Move students into “Transfer” learning where they synthesize multiple viewpoints to create new knowledge.
An example of structured protocol. See more examples.

Getting to Know the Protocols – ARC, N-W-N, VIVA

One of the challenges is coaching others in how to embrace UDL in a 15-minute meeting. To assist with that, here are some protocols along with a variety of scenarios:

One approach is use the ARC Protocol. That involves:

  • Acknowledge the teacher’s reality.
  • Reframe the problem from student-deficit to design-deficit.
  • Commit to one small, high-leverage shift.

The ARC Protocol is intended to keep the conversation moving forward without dismissing the teacher’s feelings. Other protocols include Notice-Wonder-Next and VIVA. To assist you, give this a spin:

Give this tool a try

You will see a “SHOW RANDOM SCENARIO” button. Press the button to get a coaching scenario with guide for each protocol you have chosen. You can see the Notice-Wonder-Next protocol with a scenario below:

Your Next Step

Identify one design barrier in your building this week. Look for a lesson where every student must demonstrate understanding in exactly the same way. Then use one of the protocols, perhaps the Notice-Wonder-Next, to discuss it with a colleague:

  • Notice: “I noticed every student had to write a five-paragraph essay.”
  • Wonder: “I wonder if some students know the content but struggle with written expression.”
  • Next: “What if we offered a podcast or debate option for the next unit?”

Change one thing, then watch what happens when you remove the barrier instead of trying to fix the student.


0 comments
0 FacebookTwitterPinterestLinkedinEmail
AndroidiOS Tips and Tricks

Stop the Spam: Tips for Filtering Unwanted Phone Calls

by Miguel Guhlin
written by Miguel Guhlin

Do you suddenly have a plethora of business-oriented calls landing on your personal device that you didn’t ask for? If you’ve noticed your phone ringing more often with calls from unknown numbers, you’re not alone. According to Truecaller’s 2024 U.S. Spam and Scam Report, 56.2 million Americans were affected by spam and scam calls in 2023. And that was 2023. By all accounts, 2026 could be worse, unless you take action. Some suggestions for what action to take appear in this blog entry.

Unfounded Fears: It’s Not Just You

After every TCEA Convention, members ask, “Did you share my phone number with vendors?” The assurances that TCEA never does this often fall on unbelievers’ ears. The reason why is that everyone is getting increased spam and scam calls. Legitimate vendors don’t like this harassment, either. It makes them look bad, like they hacked into a legitimate member database and are now going wild calling people. That perception couldn’t be farther from the truth.

“TCEA does NOT provide phone numbers on the attendee lists that exhibitors can purchase.”

– TCEA Staff member

The problem has gotten so bad (the FCC has a webpage on the subject!) that 41% of people of a certain age have stopped answering unknown calls altogether. Although I don’t fall into that demographic, that is one of my self-defense strategies. I had to argue with my healthcare provider about it last week. “Why are you calling me? Send me an email with your phone number or text me FIRST.” It is an unfortunate defensive measure that results in legitimate calls sometimes getting lost in the shuffle.

The good news? You don’t have to choose between answering every call or missing important ones. Here’s how to take back control of your phone.

Built-in Tools: Your First Line of Defense

Sporting a new Samsung S25 smartphone, I had to look up how to reduce spam calls without installing anything. As a veteran iPhone user who made the switch, I took advantage of native features both phones have to reduce spam calls. My phone sits blessedly quiet, except for calls from family and friends. And, of course, that annoying healthcare provider who is trying to follow up on a doctor’s visit.

For iPhone Users

Silence Unknown Callers

This feature sends calls from numbers not in your Contacts, recent outgoing calls, or Siri Suggestions straight to voicemail. While your phone may not ring, you still see the call in your Recent calls list.

To enable it: Settings > Apps > Phone > Silence Unknown Callers

Call Screening (iOS 26+)

Apple’s newest weapon against spam is impressive. When enabled, your iPhone will automatically ask unknown callers why they’re calling and send you a real-time transcript. You then decide whether to accept the call. Users have been raving about how effective this feature is at discouraging telemarketer. Most decline to explain themselves to a robot. I can’t wait to see what happens when the telemarketers starting using their own AI bots to talk to Apple’s bot.

To enable it: Settings > Apps > Phone > Ask Reason for Calling

For Android Users

Filter Spam Calls

Open your Phone app, tap the three dots, then go to Settings > Caller ID and spam protection (see image below where this is enabled). Enable both “See caller and spam ID” and “Filter spam calls” to prevent suspected spam from ever ringing your phone.

Call Screen (Pixel and Samsung)
Similar to Apple’s new feature, Google’s Call Screen uses a virtual assistant to answer suspicious calls and ask who’s calling. It’s honestly entertaining to watch telemarketers try to navigate it. You’ll find this option in your Phone app settings.

Third-Party Apps: When You Need More Firepower

If calls are still getting through, you may need a little more assistance from these apps. To be honest, I haven’t needed these types of apps since 2018, but I keep them on my short list. I actually paid for Truecaller ($29.90 a year) for both my wife’s phone and mine. Why? Built-in spam call protection was not as robust in 2018 as it is now in 2026.

  • Truecaller (View video): A popular option with over 450 million monthly users. It features a color-coded system (🔵 Blue = Normal call; 🔴 Red = Spam or fraud; 🟣 Purple = Priority call; 🟢 Green = Verified business). But even if it’s a verified business, it’s up to you as to whether you want to answer the call or not. The free version gives you basic caller ID and spam blocking.
  • Hiya: With 400+ million users worldwide, Hiya offers spam detection, auto-blocking, reverse number lookup, and real-time updates in a clean interface. Hiya offers free spam detection, but paying $24.99 a year gives you enhanced spam blocking and premium caller ID features.
  • RoboKiller: Uses predictive analytics and audio fingerprinting to block up to 99% of spam. Bonus: it deploys “answer bots” that waste telemarketers’ time. Cost is $24.99 a year with a seven day free trial.
  • Simple Call Blocker (iOS) – A completely free option that lets you block entire number ranges. Getting bombarded by calls starting with 407-541? Block the entire range from 407-541-0000 to 407-541-9999.
  • Call Control: Uses crowdsourcing and community reports to identify spam. The premium version ($29.99/year) adds powerful features including landline support.

No matter which solution you choose, you are joining a community of people fighting to keep their phones clear of spam calls. You can also check with your carrier (AT&T Armor, Verizon, T-Mobile, etc.) for similar offerings since most now provide some level of free spam protection.

Pro Tips for Maximum Protection

Two actions you can take right now include searching for your number online. Use Google Search to find your phone number in quotes (“555-555-5555”). This may help you find where it is listed online. Also, check your digital footprint via websites like LinkedIn, email signatures, and other online spaces. You may have inadvertently copied and pasted your phone number into a public space.

The Bottom Line

Between native smartphone features and third-party apps, you can eliminate almost all spam calls on both Android and iPhone. Start with your phone’s built-in settings, add a reputable third-party app if needed, and check what your carrier offers. Share in the comments what approach has worked for you.

0 comments
0 FacebookTwitterPinterestLinkedinEmail
Artificial Intelligence

AI Maker Magic, Part 3: How to Vibe Code

by Miguel Guhlin
written by Miguel Guhlin

In the two previous parts of this blog series, you saw how to make a custom GPT accessible to anyone, whether they have a ChatGPT account or not, as a website. In part one of the AI Magic Maker series, you created a custom GPT-based Fun Personality Quiz. In part two, you saw how to turn a custom GPT into a webpage for easy access via the web. In part three, you will learn how to simply vibe code to collect the data entered via the web version of Fun Quiz into a Google Sheet.

Want to take the Fun, Personality Quiz? Take the Quiz!

About This Series

In this part, we’ll explore how to save data from the interactive webpage to Google Sheets for later analysis. Saving the data from the interactive webpage to Google Sheets takes this interactive page to a whole new vibe code level since you can now create quizzes and more that save the data.

Wait, How Is This Possible?

The journey from custom GPT requiring individual user accounts to a webpage anyone can access is made possible through the use of free tools. Those tools include:

  • A GitHub account. How to get a free account is explained in part two of the series.
  • A Google Drive account with Google Sheets. You can do this with the free, or consumer, version of Google Sheets or you can use your Google Workspace for Education/NonProfit account.
  • Google Apps Script Web App. This is such a game-changer and I suspect you may soon be making your own Web App scripts using Generative AI. It’s the “hidden” lesson in this series.

All these tools are free. Of course, I did rely on a paid Gen AI model (about $20 a month).

Achieving the Impossible

Using Gen AI (for part three, I asked my Claude Pro account for help), I asked for a step by step. Since I’m not a Google Apps script whiz and failed BASIC programming in high school (I hope you appreciate how tough an admission like that would be for a high school student), I can attest to the power of vibe coding to help me achieve this hitherto impossible task.

Yes, with Gen AI at your side, you can leap tall mountains of code. But you first must have a plan.

The Plan

The plan is that you will create a Google Sheet and deploy a Google Apps Script as a Web App (takes approximately five minutes). Then, Gen AI will update the script.js to POST (a specific action, so that’s why it’s capitalized) the result to the Web App URL.

The data logged per quiz submission will be:

  • Timestamp
  • Quiz name (e.g., “Smurfs”)
  • Result character (e.g., “Papa Smurf”)

Step 1: Set up Google Sheets + Apps Script

These are the steps I will follow in the video available to you at the end of this blog post. In the meantime, follow these steps:

  1. Create a new Google Sheet.
  2. Name the first sheet tab Results.
  3. Add headers in row 1: Timestamp, Quiz, Character.
  4. Go to Extensions → Apps Script.
  5. Replace all code with this:

function doPost(e) { var sheet = SpreadsheetApp.getActiveSpreadsheet().getSheetByName("Results"); var data = JSON.parse(e.postData.contents); sheet.appendRow([data.timestamp, data.quiz, data.character]); return ContentService.createTextOutput("OK"); }

  1. Click Deploy → New deployment
  • Type: Web app
  • Execute as: Me
  • Who has access: Anyone
  1. Copy the Web App URL. You will need it next.

Step 2 — Updated script.js

Now that you are ready with the Google Sheets Web App URL, you will need to add it to the script.js file at the top of the file. This will result in rows like the ones in this Google Sheet:

This is the only file you need to replace. All four HTML files already load script.js, so nothing else changes. Once you have your Apps Script Web App URL, replace “YOUR_WEB_APP_URL_HERE” at the top of the file.

Each completed quiz will log a row like:

TimestampQuizCharacter
2026-03-11T14:32:00ZThe SmurfsPapa Smurf

That’s pretty much all there is to it. Not sure what to do?

Watch the Video

This video walkthrough will get you started. You are also going to get the download file (a compressed zip file linked below the video) with all the web pages and the generic script.js so that you won’t be tempted to simply use mine.

In this video, you get a tour of the process, from start to finish:

You can also download a copy of the entire FunQuiz website from GitHub using the link below:

Get the FunQuiz HTML/web files with both versions of the script.js file

Remember, if you decide to try this yourself, you will need to update the script.js file to point to YOUR Web App URL, rather than mine.

Since you have both versions of the file, you can compare them:

From left to right: Script.js file that saves results to Google Sheets, script_NoWebAppScript.js that simply runs the quiz and no results are saved. Make sure to rename the “script_NoWebAppScript.js” to “script.js” if that’s what you decide to use.

Wrapping Up

Vibe coding has become so easy, and with free hosting by GitHub or any web server you have access to, you should be able to create web/HTML pages with ease and share them. Here are some potential applications in school environments, even though it may drive technology directors crazy with the need to host teacher-generated HTML pages (but it’s super inexpensive, so why not on an intranet server or off a USB flash drive?).

Here are practical, high-impact ideas organized by who benefits most.

https://mguhlin.github.io/funquiz/examples.html

A simple, bare bones example of one of each of the following ideas is available. It took about thirty seconds to make, and that shows. I could have spent more time on it, but this example should give you an idea of what’s possible with minimal time investment.

🧑‍🏫 For Teachers

  • Lesson Timer and Agenda Display. A full-screen HTML page showing the day’s agenda with a countdown timer per activity. No app needed, just a browser tab.
  • Exit Ticket Generator (example included). Input a topic, get a randomized exit ticket question. Teachers could build a bank of prompts tied to SOLO levels (unistructural → relational questions).
  • Seating Chart Tool. Drag-and-drop HTML interface for arranging desks. Exportable as an image. Far lighter than any app.
  • Vocabulary Flashcard Set. A simple spaced-repetition card flipper built around a unit’s word list. Teacher edits one JSON file to update the deck.

🎓 For Students

  • Portfolio Landing Page (example included). Students build a personal HTML page showcasing work samples, reflections, and goals. Hosted free on GitHub Pages. Real-world skill, real audience.
  • Study Guide/Reference Page. A single-page interactive reference sheet for a unit — definitions, diagrams, examples — that students help build as a class project.
  • Peer Feedback Form. A lightweight HTML form that structures feedback using a protocol (e.g., “I notice/I wonder/What if”). Submissions could go to a Google Sheet via a free form endpoint.

🏫 For Campus/Department Use

  • Club or Team Hub Page. A simple landing page for a club: meeting times, announcements, photo gallery. Students own and update it.
  • Professional Development Resource Library (example included). A searchable, filterable page of links, one-pagers, and videos organized by topic. Way faster to update than a LMS page.
  • Data Dashboard for a PLC. A static HTML page that visualizes pre/post assessment data from a CSV. No backend needed if you use a SheetJS that will read the file client-side. (No, I don’t know what SheetJS is either, but ask Gen AI to explain it to you).

For example, Dr. Aida Guhlincozzi runs a Geosciences camp in Missouri (wouldn’t this be great to have in Texas?). Here are some sample projects that camp participants (all students) could create:

1. Interactive Disaster Response Map. Students place SVG/CSS pins on a campus or regional map, tag each with resource type (water, food, shelter), and add pop-up notes explaining why that location matters. Demonstrates spatial reasoning and GPS concepts.

2. “Before & After” Disaster Slider. A CSS clip-path slider showing satellite imagery or hand-drawn maps before and after a natural disaster. Students write captions explaining the geoscience behind the change.

3. Personal Field Journal. A scrollable, styled “expedition log” with sections for hypothesis, observation, data collected, and reflection. Could include a hand-drawn map image upload area. Demonstrates scientific writing and process documentation.

4. Mini-Data Dashboard. Students input their own GPS coordinates from a field activity into a table that auto-calculates the distance between points. Demonstrates data literacy and real-world math application.

5. “Meet a Geoscientist” Profile Card. A styled bio card students build about a scientist they researched; includes a quote, field of study badge, and “why this matters” section. Great for identity/representation work (ties directly to Guhlincozzi’s story).

6. Decision Tree: What Would You Do?. A branching HTML page presenting a disaster scenario. Each choice leads to a new page/section with consequence text. Students author the branches, not just navigate them.

The Pedagogical Angle Worth Naming

These aren’t just tech projects. Each student item maps directly to Transfer learning in Hattie’s framework. Students applying content knowledge to build something real for a real audience is one of the highest-leverage moves in the Surface → Deep → Transfer progression.

GitHub Pages hosting + a basic HTML/CSS/JS file = zero cost, real publication. Oh, the places we’d go!

Code Reject

Google Gemini 3 Pro coded this page below. It’s amazing, but not quite as good as the Claude Pro generated version included above. Still, for comparison’s sake, it’s worth including below:

Access this example online
0 comments
0 FacebookTwitterPinterestLinkedinEmail
Artificial IntelligenceEnglish Language Arts Reading (ELAR)Writing

News-Style Pet Flyers: Teaching Writing for Real

by Miguel Guhlin
written by Miguel Guhlin

When a pet goes missing (or a “loose dog followed us home”) need facts and next steps. Local news writing is designed to get that job done. Pair that writing style with a phone-friendly flyer and a single image that includes both the photo, key details, and you can get your message across. That is a writing task that can work well for teaching students in grades 5-6, two of my favorite grades, how to write.

This blog entry suggests encouraging students to ground new learning in real life. That means drafting a short news brief, converting it into flyer copy, and then using a Generative AI image generator to create a polished flyer with a headline, facts, and what to do next. The focus is not “AI magic.” The focus is audience, clarity, structure, and verification.

Image generated by author with ChatGPT

Why News Writing Makes Pet Flyers Better

Too much text, missing time/location, multiple contact methods, or emotional language can cause a message to fail. So teach students news-style writing that fixes that. Require writing to have these components, at minimum:

  • A headline that names the situation (MISSING / FOUND)
  • A lede that answers who/what/when/where quickly
  • A brief timeline (last seen, found, actions taken)
  • A “What you can do” section with one clear action

The purpose of this type of bare-knuckle writing is to get things done, to get action.

The Classroom Setup

Relying on fictional scenarios enables students to craft compelling stories without risking privacy. You can get groups of students (I like to group mine a la Atwell in four desks) to work on scenarios. Give each group a “scenario card” (missing cat, found dog, loose pet seen near a park). Include a short list of facts they must keep consistent.

Scenario card includes:

  • Status: Missing / Found / Seen loose
  • Description: size, color, markings, collar
  • Time window: “between 4:00–6:00 PM”
  • Location: neighborhood or cross streets (not an address)
  • Behavior: skittish/friendly, approaches people or runs
  • Microchip note: scanned/not scanned (optional)
  • One contact method: a classroom-safe placeholder (example: “Text the school office”)
  • One action for readers: “Check cameras,” “Report sightings,” “Share”

A Workflow Students Can Follow

Here are a few steps you can put to good use in the classroom with students:

Step 1: Gather facts (5–8 minutes)

Students complete a simple fact checklist. This is their “source note.”

Step 2: Write the news brief (15 minutes)

Students produce a 150–220 word local-news-style story:

  • Headline
  • Lede
  • Timeline
  • “What you can do”

You might, along with this step, bring in a few flyers and ask students to critique them. Consider having them craft a rubric to assess what worked and what did not while protecting the private data on the flyers (black marker for redactions). Then, after they have it down, they can take a stab at writing the news brief.

Step 3: Convert to flyer copy (10 minutes)

With their news brief in hand, students then reformat the same information into:

  • 3–6 word headline
  • 5–7 bullet “facts box”
  • 40–70 word summary
  • One call-to-action + one contact method

You might also rely on the VIVA framework to ensure students have a clear grasp of the facts.

Step 4: Verify and revise (10 minutes)

Before moving on to Gen AI tools, encourage students to do an “accuracy sweep:”

  • Is MISSING/FOUND unmistakable?
  • Did we make up anything not on the scenario card?
  • Is location general (neighborhood/cross streets), not an address?
  • Is there one contact method only?
  • Is the tone calm and helpful?

Only after verification do students move to design. Provide some news flyer templates that they can create with hand-drawn images and handwriting. Make it fun.

Where Gen AI fits

Whether you are using Google’s Nano Banana Pro or ChatGPT’s Image Generator, Gen AI tools can help in two useful ways:

  1. Drafting from notes (students still verify)
  2. Generating the flyer image layout (clean typography, scannable design)

The important classroom message: AI can draft and design, but it cannot be trusted to verify. Students must keep control of facts. What’s more, you can also ask students to draft sample prompts to get the visual result they are seeking from the Gen AI tool. This can be managed via Google Docs or similar shared documents. Student groups submit the writing and the teacher uses the Gen AI tool students have chosen to get the result. That result is then shared back with students.

Prompt Set for Writing

Below, please find a few prompt ideas to facilitate writing. Again, I encourage you to have students write the first two drafts, then ask the Gen AI tool to fine tune it, and/or compare their writing with what the Gen AI tool comes up with.

News brief prompt

Write a short local news brief (150–220 words) about this fictional incident. Include a headline, lede, short timeline, and a “What you can do” section. Keep it factual and calm. Do not include exact addresses. Facts: [paste the scenario facts].

Flyer copy prompt

Create flyer text for a fictional missing/found pet. Format it as: (1) headline (3–6 words), (2) 5–7 bullets, (3) 40–70 word summary, (4) one call-to-action with one contact method. Keep it easy to scan on a phone.

Image Generation Prompts

If you want the final product to be an image students can post (or print), have them “pour” the verified flyer text into an image prompt like the one below. Keep the design requirements strict for readability.

Image generation prompt (copy/paste)

Create a clean, phone-friendly flyer infographic. White background, simple grid layout, high contrast, large readable text.
Title at top: “[MISSING DOG]” or “[FOUND CAT]” (all caps).
Include a “FACTS” box with 5–7 bullets.
Include a short “WHAT TO DO” box with 2–3 bullets.
Include one contact line at bottom: “TEXT: (xxx) xxx-xxxx”.
Use clear sans-serif fonts only. No stylized lettering. No typos.
Include a simple, non-photorealistic pet illustration that matches: [describe pet].
Do not include any address. Use only neighborhood/cross streets: [text].
Use calm, professional wording.

If you have a real pet photo, you can use image editing instead of generating a new pet image; but for classroom scenarios, a simple illustration is safer and avoids accidental “look-alike” issues. Consider requiring students to proofread the generated image text. If it contains errors, they must regenerate with corrected text. You can also model branching chats, and if you have access to tools like BoodleBox, consider taking advantage of bot stacking.

A Final Word: Assessment

Since assessment remains an ever-present concern, recall that in the writing workshop, it’s all about the process and steps students have taken to get to this point. It’s not the final product where the assessment happens so much as the significant steps along the journey. Consider grading on each component of these major areas listed below:

  • Structure: headline/lede/timeline/next steps
  • Clarity: scannable flyer, short bullets, one action
  • Accuracy: matches scenario card, no invented details
  • Audience + safety: no private info, calm tone, clear contact method

Why News-Style Pet Flyers Work

As a grown-up, I’ve had to put together my own news-style pet flyers. It’s a valuable skill that scales up and teaches authentic purpose-driven writing. The purpose is to get community response for the benefit of saving a life. Students practice informational writing, revision, and media literacy in a format they recognize. The AI piece becomes a controlled design tool, not a shortcut for thinking. This empowers students to learn to write like a newsroom, then publish like a community helper. It’s a powerful skill to learn as journalism fades and Gen AI takes away opportunities to teach that news-style writing can save lives.

0 comments
0 FacebookTwitterPinterestLinkedinEmail
AI/VRCTO/CIO

KPI Dashboards Built with Gen AI

by Miguel Guhlin
written by Miguel Guhlin

We have all been there. The board meeting is in three days. Someone just sent you a spreadsheet with device deployment numbers that don’t match the ones in last month’s cabinet deck. You open a browser tab to find the help desk ticket totals, then another for PD completion rates, then another for the budget utilization report your finance director exported as a PDF. It’s 7:45 in the morning and you are already tired.

Open School District KPI Dashboard

As district leaders, we are all tracking more data (see example) than ever. Turning that data into something readable for a board update still means hours of manual work, or a tool license nobody budgeted for. One of my Aha! moments as a technology director? Make a KPI dashboard. But in the past, making digital KPI dashboards cost money and involved complex tools.

Before we jump into KPIs, let’s take a moment to define them. A well-defined KPI can make it easier to gauge success. Here’s a definition that I use with folks attending my webinar for TCEA’s IT/Director Certification:

KPIs are an actionable scorecard. They focus your strategy. They are metrics that make management, control and achievement of  desired results possible.

The purpose of KPIs is to create an easy-to-understand visual representation of metrics, automating data analysis from a few key systems, then share it an easy way.” This makes it simple for staff to update data and for the end user to see it, too (source).

There’s a simpler path, and it runs through Claude or BoodleBox Unlimited. You can also use Gemini or ChatGPT, but I will refer to BoodleBox Bots for the purposes of this blog entry. That’s because I created a demo you can use once you get your BoodleBox Unlimited account (see note below).

Note: Get a BoodleBox Unlimited account via referral code (one month) or use this free code MGFREE123 (two months). Want to learn more about BoodleBox? Take The AI-Amplified Educator Accelerator, an online learning certification opportunity. TCEA also offers other opportunities to learn and make with Gen AI.

Answer Three Questions, Get a Working Dashboard

Using BoodleBox Unlimited with a purpose-built set of Bot instructions, you can generate a fully functional, browser-based KPI dashboard by answering three questions:

  1. What are you tracking? (Technology, Student Achievement, Curriculum, Budget, Professional Development, or Registrations)
  2. What’s your district name and school year?
  3. Do you have real data to drop in, or should BoodleBox Bot generate sample data to start?

BoodleBox will then produce two files. One HTML file runs the dashboard in any browser. One Markdown file holds all the data.

That Markdown file is the only thing anyone on your team ever edits again.

Put both files on a shared hosting folder or internal web server, open the HTML file in a browser, and the dashboard builds itself from whatever is in the Markdown file. You can see one example here.

See live version online

What Your Leaders Actually See

Every dashboard follows a scoreboard-style layout: a fixed left sidebar for navigation, animated hero stat cards showing top-level numbers, and a detailed breakdown table with inline progress bars by campus or department. While it may seem it took a long time to build, it did not. After all, the point of this is to focus on the key performance indicators and the data, not spend hours on the look.

Each KPI category arrives with sensible defaults. A Technology dashboard tracks devices deployed, open help desk tickets, network uptime, and staff trained, broken down by campus. A Budget dashboard shows allocated vs. spent by category, with a utilization bar for each line item. Student Achievement surfaces passing rates, at-risk counts, and attendance.

Goals are optional but useful. Add one line to the Markdown file, something like “Goal Devices Deployed: 2500,” and the corresponding stat card gains an animated progress bar. Remove the line and it disappears. No one touches the HTML file to make that happen. For me, that is the beauty of this system. Markdown is easy, HTML…not so much.

Sustainable Dashboards

What usually kills KPI dashboards is cost of expensive solutions, as well as the onerous task of maintaining the dashboard numbers. When the data goes stale, the Dashboard becomes useless.

To avoid stale data, expensive dashboard software or web services, use this setup. The Markdown file looks like a simple table. Your designated data coordinator opens it, changes the numbers, saves, and reloads the browser. In fact, you can drop the markdown file into a Google Doc, and then copy-n-paste the markdown version of that information into the text file.

That’s the complete workflow. No dashboard platform login. No export-import cycle. No ticket to IT.

## Technology

| Campus | Devices | Tickets Open | Tickets Closed |
|--------|---------|-------------|----------------|
| Central High | 842 | 12 | 47 |
| North Middle | 614 | 8 | 31 |

Change a number, save, refresh. That’s all it takes. Pretty easy. Want to get fancy? If you are using a Gen AI tool that can pull data from Google Drive (e.g. Gemini, Claude, ChatGPT), then see if you can set this up with a Google Sheet.

Who This Fits

The audience for this isn’t everyone, although you could customize the instructions for a classroom or campus. If you have metrics that involve measuring things that can be counted, you can easily adapt this for your use.

RoleWhat They Track
Technology directorDevice rollouts, help desk volume, network uptime
Curriculum coordinatorPD completion, program adoption, teacher participation
Finance and operationsBudget utilization, allocated vs. spent by category
Cabinet-level leaderA single URL for board meetings that reflects current numbers

The best thing? There’s no database, special server-side code, or monthly license (well, none except for a Gen AI tool). All you need is a markdown text file and a browser.

Adding Sections as Your Needs Change

Need to add a Project Management section mid-year? Tell your Gen AI tool of choice the section name and the columns you want to track. It updates both files together and explains what changed. The sidebar navigation adds the new section automatically.

Multiple campuses or departments can each have their own color-coded section inside the same dashboard, visually distinct but fed from a single file.

The TCEA registration dashboard, for example, has a general course registration section and a separate district section side by side, each with its own color scheme and stat cards. When registration data needed to be added mid-project, it went in as pasted text. The Gen AI tool formatted it and updated both files in one exchange.

Getting Started

The Project instructions include a catalog of KPI categories with pre-defined columns, sample data, and suggested goal targets. When you describe what you need, even informally, BoodleBox will ask you clarifying questions, generate both files with realistic sample data, and deliver a plain-language guide on how to maintain it.

The sample data is realistic enough to use in a demo or a training session immediately, which means you can put a working dashboard in front of your cabinet on day one, then swap in real numbers when you’re ready.

What to prepareWhy it matters
A list of the data points you track most oftenHelps Gen AI generate columns that match your actual reporting needs
Your district name and current school yearPopulates the dashboard header automatically
One real data table, even a rough oneGets you real numbers instead of sample data from the start

If you’ve been pulling together board updates from five different spreadsheets every month, this is worth an afternoon to set up.

Give It a Go!
Try describing your district’s data needs to the District Dashboard Generator and see what it generates. Share what you build in the comments below.

How To Get The Dashboard

Ready to get give it a go? Explore this Google Doc with link to the BoodleBox Bot, District Dashboard Generator. It includes an example of the data.md file (that’s the markdown file) in Google Doc format, and custom instructions you can copy and use on your own AI tool.

0 comments
0 FacebookTwitterPinterestLinkedinEmail
iPads
Educational ResearchGood TeachingInstructional Coaching

TCEA Responds: Navigating Resistance, Part 1

by Miguel Guhlin
written by Miguel Guhlin

Dear TCEA Responds:

I’m an instructional coach, and lately, I feel like I’m hitting a brick wall. Every time I suggest a high-effect size strategy or mention video coaching, I get “the look.” You know the one: the crossed arms and the “I don’t have time for this” sigh. How can I have these tough conversations without ruining my relationship with my teachers?

Thank you!
-Frustrated in Frisco

Dear Frustrated:

Thanks so much for reaching out. First, take a deep breath. Resistance isn’t a sign that you’re a bad coach. It’s a sign that your teachers are human.

In my blog entry on Coaching for Results, I talk about the “vulnerability gap.” When you ask teachers to change, you are asking them to step out of their comfort zone and into a space where they might fail. That’s terrifying, especially when their professional identity is on the line.

Did You Know?

According to John Hattie’s research, the “hinge point” for instructional strategies is an effect size of 0.40. Anything above that is considered significantly impactful. But here is the catch: no strategy works if the teacher is too defensive to implement it.

The real question isn’t “What strategy should I recommend?” It’s “How do I create the psychological safety needed for that strategy to even have a chance?”

Learn more about TCEA’s Instructional Coaching Certification

The Secret Sauce: The ARC Protocol

To move past the “crossed arms,” we need a communication framework that prioritizes psychological safety. That’s where the ARC Protocol comes in. It’s a three-step process designed to lower the “affective filter” and get to the work that matters.

A – Acknowledge to Make it Safe

Before you can talk about data or strategies, you must validate the person. When people feel heard, their brain’s amygdala (the “fight or flight” center) calms down.

The AMPP Technique:

  • Ask — Open with questions to understand their perspective.
  • Mirror — Reflect their emotions back to show you’re listening.
  • Paraphrase — Restate their concerns in your own words to confirm understanding.
  • Prime — If they’re holding back, offer a guess to get the conversation flowing.

What to say: “I hear how exhausted you are” or “I understand why that feels overwhelming.”

The Goal: Build a “brave space” for dialogue. As the research in Crucial Conversations reminds us, you must lower the emotional temperature first.

R – Reframe the Narrative

Once the teacher feels safe, you shift the perspective. Most resistance comes from a “problem-centered” view. Your job is to pivot to a “growth-oriented” or “solution-focused” view.

The Pivot: Move from “This is a new burden” to “This is a tool to solve the problem you just mentioned.”

The AMPP approach is a Crucial Conversations/Confrontations concept that I’ve relied on often in my own work. Be sure to learn more about it in the book links included in this blog entry.

Example: If they hate the noise of group work, reframe Classroom Discussion (d=0.82) as a way to let students expend that energy productively on the content. Stop arguing about the teacher’s preference and start talking about student data. Connect the strategy to High-Effect Size Instructional Strategies (HESIS) so the conversation shifts from opinion to evidence.

C – Commit to a Micro-Step

This is where many coaching conversations fail: they stay too big. A “Commitment” shouldn’t be a whole new lesson plan. It should be a micro-step.

The Rule: If it takes more than 10 minutes to plan or execute, it’s too big for a first step.

The Goal: Immediate, actionable success. Don’t leave without a date on the calendar. This aligns perfectly with the TCEA EIIR coaching model (Empower & Engage, Investigate & Innovate, Implement & Improve, Reflect & Realize).

Key Idea: Focus on the “Heavy Hitters”

When you use the ARC Protocol, make sure you are leading them toward strategies that actually pay off. Don’t waste your “coaching capital” on low-effect strategies. Refer to the High-Effect Size Instructional Strategies list often. Consider how concepts from Crucial Conversations, Crucial Confrontations are helpful in this.

If you’re going to ask a teacher to take a risk, make sure it’s for a strategy like:

  • Reciprocal Teaching (d=0.74)
  • Jigsaw (d=0.92)
  • Feedback (d=0.92)
  • Video-Based Coaching (d=0.99)
  • Conceptual Mapping (d=0.64)

Wrapping It Up

Coaching is 20% strategy and 80% relationship. The ARC Protocol ensures that the 80% stays intact so the 20% can actually happen.

Your Next Step: What’s one conversation you’ve been putting off? Try applying the ARC steps to your planning today. Remember:

  1. Acknowledge: Lower the emotional temperature first (Crucial Conversations: “Make it Safe”).
  2. Reframe: Stop arguing about teacher preference and start talking about student data (HESIS).
  3. Commit: Don’t leave without a date on the calendar (TCEA EIIR).

The tangled strands of resistance can become a smooth, golden braid of collaborative growth. One conversation at a time. In the second part of this blog entry, you will see five scenarios that illustrate the ARC Protocol in action.

Make a Gen AI Connection

To take this further, consider creating a “Coaching Conversation Generator” prompt:

Using the ARC Protocol (Acknowledge with AMPP, Reframe with HESIS, Commit with micro-steps), generate a coaching conversation for the following scenario: [describe teacher resistance]. Include specific effect sizes for recommended strategies and ensure the micro-step takes less than 10 minutes to implement.

Feel free to use this blog entry to provide context your Gen AI prompt needs to deliver relevant responses. Combine it with part two of this series (to appear soon) to get examples of desired output from your Gen AI tool of choice.

2 comments
0 FacebookTwitterPinterestLinkedinEmail
AI/VRArtificial Intelligence

AI Maker Magic, Part 2: A Fun Personality Quiz Webpage

by Miguel Guhlin
written by Miguel Guhlin

How do you make a custom GPT accessible to anyone, whether they have a ChatGPT account or not? You turn it into a website! In part one of the AI Magic Maker series, you created a custom GPT-based Fun, Personality Quiz. The only problem is that this personality quiz lives inside a Custom GPT. If you did not have access to that GPT, or a free ChatGPT account, you could not use the quiz. Instead of a custom GPT, I wanted a web address or uniform resource locator (URL). In this way, I would be able to drop the link into a slide deck, blog post, or whatever for easy access. A tool I knew little about (i.e. GitHub) made this possible.

GitHub Repository for the website files generated by ChatGPT and the author

Let’s go through the process together. Towards the end, you will find a video that walks you through the GitHub set up as well.

Step 1: Capture the Quiz as a Web Page

First, I asked ChatGPT to help me translate the quiz into a web format. What I am looking for has to include:

  • A simple page titled “Fun Personality Quiz for Educators.”
  • The same instructions as the GPT version.
  • A place to show the quiz image at the top.

Also, I wanted to get an interactive version using HTML and Javascript with buttons, tallies, and the response with the result. To get ready, you will need:

  • A GitHub account (for hosting your quiz for free )
  • A GitHub repository (which you will create)
  • The files you will need (e.g. index.html, style.css, script.js)

Your next step is to sign up for a free, GitHub account.

Step 2: Create a Dedicated GitHub Repository

After creating an account using GitHub, you are ready to create a repository. Rather than mix this quiz with other projects, I created a new GitHub repository just for it. Something like:

funquiz

Then, I prompted ChatGPT to create the files I would need with a simple prompt:

I want to turn this personality quiz into a github page that people can fill out online, then get the result right then and there. How would that work?

It then offered a step-by-step for me to follow, and provided the required files (list appears below).

Did You Know?

In part three of this series, you will modify that prompt to get a version that saves the results into a Google Sheet. For now, this is a simple process to put the quiz online.

With my new repository, funquiz, you will see the following:

As you can see, this image offers a folder structure for placement of files. I did not include the assets (I’ll do that in part three). Or, to be more exact, I included them but they did not work. I will ask ChatGPT to fix that in the final version.

You can prompt ChatGPT to do the following prompt, which appears below:

Begin prompt

Create a complete GitHub Pages project for "Fun Personality Quizzes for Educators" called `funquiz`. 

It should include:
- One HTML page per quiz (Pooh, Smurfs, Dwarfs, Looney Tunes)
- Shared CSS for layout and styling
- Shared JS logic for quiz progression and results
- One question set per quiz in separate JS files under `/data/`
- A dropdown menu to choose a quiz from any page
- An `/assets/` folder with character images or placeholders
- A README.md file explaining how to deploy and customize

Package it all as a downloadable ZIP I can use or upload to GitHub Pages.

End Prompt

Now the quiz exists as a small website folder, ready to publish. Before putting it into the GitHub repository, you can test it on your device. I created a folder to house these files, then uploaded (put) them into the GitHub repository once they had the look I wanted.

Step 3: Turn on GitHub Pages

In the repository:

  1. Go to Settings.
  2. Scroll to Pages (sometimes under “Code and automation”).
  3. Under Build and deployment, set:
  • Source: Deploy from a branch
  • Branch: main
  • Folder: / (root)

4. Save the settings.

    GitHub will build the site and generate a public URL, usually in this format:

    https://<username>.github.io/funquiz/

    This is your quiz’s public home. For my example, the web address is https://mguhlin.github.io/funquiz/.

    As you can see from the screenshot below, I’m able to pick the quiz I want users to see. I can also upload images so that each quiz has its own custom image. I’ll save that for the final version in part three of this series.

    Step 4: Test the Live Quiz

    Next, I opened the URL in a browser:

    • Did the title display correctly?
    • Did the image load?
    • Were instructions clear for educators?
    • If interactive, did clicking answers produce a final result?

    Any changes needed went back into index.html or one of the other files. You can watch a short video of the steps I followed below:

    Step 5: Share, Embed, and Extend

    For the final step, I’m ready to share this with colleagues as a link that will go in a blog post (like this one), or use it in a future workshop. In part three of this series, I will add images, instructions, and save the results of the quiz to a Google Sheet.

    The big benefit of all this is that we not only end up with a fun personality quiz, but also, have a process for vibe coding our way from a custom GPT to a website hosted for free online. Of course, GitHub Pages may be blocked by your school district, but that’s another blog entry.

    0 comments
    0 FacebookTwitterPinterestLinkedinEmail
    Newer Posts
    Older Posts