Overview:
Effective instructional coaching improves teaching when it is built on trust, occurs frequently, and is embedded in real classroom practice, allowing meaningful, sustainable change to spread organically among educators.
It all started in a classroom that looked like everything was going as planned. The students seemed to be engaged. The lesson was moving along. Nothing appeared to be off-track. But when the teacher and I sat down afterward, she paused and said, “I think they’re doing the work, but, honestly, I don’t think they actually understand what they are doing.”
That moment did not come from my formal observation. It came from trust. From enough time spent working together that she did not need to perform the lesson, but she could be honest about how she felt her lesson went. That is the kind of coaching that changes classrooms. And it does not happen by accident.
It Starts With Trust
We often talk about instructional coaching in terms of models, content-focused, pedagogical-focused, and data-focused. And those absolutely matter. But structure alone does not move practice. Trust does.
Research on effective professional learning consistently points to the same conditions: it has to be sustained, collaborative, and embedded in teachers’ daily work (Darling-Hammond et al., 2017). Instructional coaching, when done well, does all three.
In my own research with K–12 educators (n = 307 who experienced coaching), teachers consistently rated relational coaching activities, like co-planning and goal-setting conversations, as some of the most impactful parts of the process (Murugan, 2025). Co-planning (M ≈ 4.5, SD ≈ 1.1) and conferencing/goal setting (M ≈ 4.7, SD ≈ 1.0) emerged as high-impact practices on a six-point scale. Those are not moments of compliance but rather trust-driven ones. They are where change begins.
Frequency Matters More
If there is one thing both research and experience make clear, it’s this that coaching cannot be occasional. Decades of professional learning research have shown that one-time workshops do not lead to sustained instructional change, ongoing support does (Guskey, 2002; Joyce & Showers, 2002). My findings reinforced that. The frequency of coaching interactions was one of the strongest predictors of educators’ perceived instructional improvement, with models explaining approximately 44.7% of the variance (R² ≈ .447) (Murugan, 2025). Not the type of coaching, nor the years of experience effected this.
Frequency.
Educators who experienced more consistent coaching, that is, weekly or biweekly touchpoints, reported stronger instructional growth than those with occasional support. This aligned with broader research showing that sustained coaching had measurable effects on both instructional practice and student outcomes (Kraft et al., 2018). Instructional change is iterative. It requires trying something, reflecting, adjusting, and trying again. A coaching cycle that instructional coaches engage in with educators Without that cycle, even strong strategies are proven ineffective.
The Work Spreads in Ways We Cannot Always Measure
Some of the most powerful shifts that I have seen in schools did not come from mandates, but spread quietly. In the study, teachers who experienced more than one type of coaching (content-focused, pedagogical, or data-focused) reported a broader instructional improvement (Murugan, 2025). On average, educators experienced about 1.39 types of coaching (SD ≈ 0.65), suggesting that while exposure varied, depth mattered.What mattered just as much was what happened between those experiences.
When one teacher tries something new with a coach, and it works, other educators may notice. Conversation may start, and then something subtle begins to happen. The work becomes visible. A strategy that once lived in a single classroom starts showing up in conversations, in shared planning time, in small moments between teachers. Not because it was mandated, but because someone saw it working and wanted to try it too. I’ve come to think of this as a kind of “The Lantern Effect” in schools, where one educator’s experience, supported through coaching, “lights the way” just enough for other educators to see what the possibilities are when they work with an instructional coach. Once that light is there, it is difficult to ignore.
Coaching That Stays Close to the Classroom Wins
The most effective coaching does not live in theory, but stays anchored in the daily realities of teaching. Instructional coaching research has long emphasized the importance of job-embedded, classroom-connected support (Knight, 2007). My findings echoed that clearly.
Educators reported the highest levels of impact when coaching was directly tied to their current instruction, not abstract strategies that PD sessions sometimes had , but real lessons, real students, real challenges (Murugan, 2025). This looks like:Co-planning tomorrow’s lesson; Looking at student work from today; Adjusting instruction for next week. Teachers do not need to come up with more ideas in isolation. They need professional support applying ideas in context.
For School Leaders: What This Means in Practice
If we want instructional coaching to actually improve classrooms, we have to design for what is actually going to move the needle, and not just for what looks good during an observation. Here are a few shifts grounded in both research and practice:
1. Prioritize frequency over coverage.
Creating a space for sustained engagement matters more than one-time reach (Darling-Hammond et al., 2017).
2. Protect time for coaching.
Ongoing, supported practice leads to meaningful change (Guskey, 2002).
3. Invest in relational coaching practices.
Trust and collaboration are foundational, not optional (Knight, 2007).
4. Focus on application, not just strategy.
Practice-based learning drives stronger outcomes (Desimone, 2009).
Creating Instructional Coaching That Lasts
At the end of the day, instructional coaching is not about programs or protocols, but rather about the people and the relationships. In my study, educators reported a high level of overall instructional improvement (M ≈ 5.13, SD ≈ 0.80). Interestingly, that improvement was not tied to a single model of coaching, but instead it was tied to consistency, relational trust, and the opportunities to apply learning in real time.
When that happens, coaching does not stay contained, but spreads. Quietly at first, then more visibly. Until one classroom becomes two. Then five. Then a whole team. That is when you know something is working. Not because it was required, but because it was chosen.
