👋 G’day {{first name | DistillED reader!}}

Dr. Haili Hughes here… In this special guest piece, I'm digging into lethal mutations — what happens when a well-evidenced strategy keeps its name and its look, but loses the thing that made it work.

📄 At the end, there’s also an exclusive FREE one-page guide to help you spot and prevent lethal mutations in practice. Sign up or log in to DistillED to get your copy.

At its core, a lethal mutation is a change to an evidence-informed practice that undermines how it is supposed to support learning. The activity may retain its original name and appearance, even though its educational value has been weakened or removed.

Kate Jones and Dylan Wiliam explain that the term was coined by Ed Haertel and describe the danger of adapting techniques without understanding their underlying principles. This can leave an approach ineffective or even counterproductive(Jones and Wiliam, 2022).

Imagine a teacher using mini whiteboards to check whether students understand equivalent fractions. Everyone answers, the teacher notices a shared misconception and models another example before moving on. The boards make students’ thinking available for a teaching decision. Now imagine the same boards being held up, admired and put away while the teacher continues with the original plan, despite widespread errors. The routine survives, but its connection to responsive teaching has disappeared. Buying the equipment was the easy part, but knowing what to do with the answers was always the work.

This does not mean every adaptation is dangerous. Changing the response format so a student can participate may improve the quality of the evidence collected. The question is whether the change preserves the intended learning process. Nor should we use the term whenever we dislike somebody else’s teaching. It describes a particular problem, which requires an explanation of what has changed and why that change matters.

Teachers have finite time, students have finite attention, and meaningful school improvement has a habit of borrowing heavily from both. When an approach loses its purpose, we can spend considerable effort maintaining something that offers little return.

Retrieval practice provides a useful example. Agarwal, Nunes and Blunt’s systematic review examined 50 classroom experiments involving 5,374 students and found benefits across varied educational settings. This gives teachers good reason to consider retrieval, while remembering that a research finding does not validate every activity carrying the label(Agarwal, Nunes and Blunt, 2021).

Consider a starter called labelled as “retrieval” and mandated as a ‘Do Now’ at the start of every lesson. Students answer questions from the slide, a lot of them get it wrong and the teacher moves on with the lesson. If the school then concludes that retrieval has made little difference, it has evaluated something other than the process it intended to introduce.

Feedback offers another warning. The Education Endowment Foundation’s guidance stresses that feedback can support learning, but can also be unproductive or harmful. Its emphasis is on the principles of effective feedback rather than whether comments are delivered verbally or in writing. (EEF, 2021). A school might nevertheless translate “improve feedback” into “write three comments in every book”. Teachers comply, students turn the page, and the visible quantity of feedback becomes the measure of success. Meanwhile, the more revealing questions go unanswered: did students understand the guidance, act on it and improve their work?

There is a professional cost here too. A teacher who spends Sunday afternoon maintaining an elaborate routine without seeing its value has good reason to feel frustrated. Repeated experiences like this can make the next evidence-informed proposal sound suspiciously like another demand for coloured pens.

Why does it happen? Sometimes an idea is compressed as it travels from research to presentation to policy. Sometimes leaders ask for something observable because the underlying learning process is harder to judge. Sometimes teachers make sensible attempts to manage workload or meet students’ needs, without enough support to recognise which changes might compromise the approach.

These are understandable pressures. Blaming individuals will do little to resolve them if the professional learning consists of a brief explanation followed by an expectation of immediate consistency.

The EEF’s implementation guidance treats change as a process, supported by attention to people, context and ongoing reflection. Therefore, distributing a resource is only the beginning of helping colleagues use it well. (EEF, 2024).

For students, the consequences can be substantial. A class may complete every required routine while misunderstandings remain untouched. Students who need a teacher to notice confusion and respond receive the outward signs of support without the help those routines were meant to provide.

Ultimately, preventing lethal mutations protects the relationship between teachers’ effort and students’ learning. It also makes improvement more credible: we can explain what we are trying, why we expect it to help and what would prompt us to change course.

So how can we bring this to life in the classroom? Start with one approach already in use. There is no need to launch a whole-school lethal mutation detection initiative, complete with lanyards and leaders stalking the corridors with clipboards. Instead, choose something worth examining and make its purpose discussable.

The following five strategies offer a practical starting point. The examples use checking for understanding, but the questions can be applied to other approaches.

Before introducing or reviewing a strategy, finish this sentence: “We need this because our students currently struggle to…” If the answer is vague, spend longer examining students’ work and responses. “We need more retrieval” names a proposed activity; “students cannot recall the vocabulary needed to explain erosion” identifies a learning problem that can guide planning.

This also helps determine what success would look like. In the erosion example, you might look for accurate recall of key terms and their appropriate use in explanations.

When colleagues observe a demonstration, invite them to notice the decisions behind it. Why was that question chosen? Which wrong answer was anticipated? What would have happened if most students had answered correctly? A polished demonstration can conceal exactly the knowledge a novice needs to develop.

Try pausing a rehearsal after the students’ imagined responses appear. Ask colleagues to choose their next move and explain it. Different choices can then be explored against the purpose of the lesson. The discussion becomes more useful than simply agreeing that the teacher used questioning well.

Include examples that look convincing but miss the point. A beautifully organised exit ticket that nobody reads makes an excellent discussion prompt. Ask what information it could provide, when that information would be useful and what would have to happen next for collecting it to be worthwhile.

Teachers will encounter circumstances that a shared plan did not anticipate. A question may depend on unfamiliar vocabulary, a response routine may exclude a student, or the planned timing may be unrealistic. Build a brief opportunity into existing meetings for colleagues to explain adjustments and examine their consequences.

A useful prompt is: “I changed this because… I preserved… I noticed…” It keeps the conversation grounded in reasoning and evidence. It also allows colleagues to spot when a helpful intention has accidentally removed an important learning demand. For example, giving students the first step of a complex calculation might help them practise the remaining steps. Giving them every step to copy would prevent you from finding out whether they could select and apply the method themselves. The judgement depends on what you intended them to learn at that point.

Learning walks, coaching and team meetings can all help, provided the questions go beyond checking whether a prescribed technique appeared. During a visit, notice what students were asked to think about, what their responses revealed and how the teacher used that information. Follow this with a conversation: “What did you notice that influenced your next decision?” The teacher may explain reasoning that was invisible to an observer. Equally, the conversation may reveal that the routine generated information they were unsure how to interpret. That gives coaching a specific and useful focus.

Keep the evidence proportionate. A few carefully chosen questions, examples of work and a later check may be enough to guide the next adjustment. These observations will not prove that a strategy caused improvement, but they can reveal whether the intended process is happening and whether students appear to be benefiting.

Finally, make it professionally safe to say that something has drifted. If acknowledging a problem feels like confessing incompetence, people have a reason to keep performing the routine. Leaders can help by examining their own expectations and removing requirements that reward appearances at the expense of purpose.

Teaching is a learning profession. We should expect our understanding to develop and our practice to need attention. The question worth carrying into the next lesson, coaching conversation or staff meeting is simple: “Is this still doing what we introduced it to do?”

— Dr Haili Hughes

If you want more:

👉 Read Mini-Whiteboards — the routine this edition opens with, and what it takes to keep it doing its job

👉 Read Strategies vs Principles — why understanding the principle behind a technique is what stops it quietly turning into performance

👉 Read The 'Do Now' — the same warning, applied to the retrieval starter most schools already run every lesson

👉 Get the Checking for Understanding CPD Collection — build the five strategies from this edition into a team CPD session

👉 Get The Mini-Whiteboard Handbook — turn the opening example into a whole-department routine that survives contact with a real classroom

👉 Get the Poor Proxies for Learning CPD Collection — more on spotting the gap between an outward signal and real learning

👉 Get Feedback and Responsive Teaching — for the EEF feedback problem this edition raises

Stopping Lethal Mutations One-Pager
To help you put this into practice immediately, download the Stopping Lethal Mutations One-Pager — a practical guide to keeping evidence-informed practices connected to their original purpose. It breaks down what lethal mutations are, why they matter, and five practical steps for preventing good ideas from losing their learning function. Useful for your own reflection, in coaching conversations, or during team PD. 16