A school can buy an excellent digital tool and still teach in much the same way. Logins rise, dashboards fill, and a launch presentation goes well; yet teachers may not have time to redesign a lesson, learners may remain dependent on prompts, and leaders may struggle to tell whether practice improved. That gap is where the next EdTech proposition could be built.
The answer to the title question is possibly—but “selling human change” is too vague to be a sound business promise. Education technology companies may increasingly need to combine software with implementation, professional learning, coaching and evidence of use. The valuable offer is not a promise to transform people. It is a practical way to help a school adopt a tool, adapt it to local teaching, evaluate what changes, and become capable of continuing without permanent vendor dependence.
This is a strategic hypothesis, not a verified forecast of the whole EdTech market. It follows from a basic implementation reality: software makes an action available; people, routines and institutional conditions determine whether that action becomes useful practice.
The product is an opportunity; implementation is the work
A license can provide access, features and technical support. It cannot, on its own, decide which learning problem matters, how a feature fits a curriculum, what teachers should stop doing to make room for it, or how to respond when learners use it in an unexpected way.
The OECD’s Digital Education Outlook 2026 makes a related point about generative AI: educational value depends on clear teaching principles and purposeful design. AI can support learning when its use is guided; without pedagogical guidance, it can make a task look better without producing corresponding learning gains. The implication for product companies is broader than AI: feature availability is not the same as learning design.
A 2026 systematic review of EdTech business models also describes the sector’s familiar emphasis on software-as-a-service, freemium access and subscriptions, while highlighting tensions around equity, data and educational outcomes. The review is a synthesis of literature, not a census of current companies or a forecast. Still, it sharpens a useful question for founders and buyers: if recurring revenue is tied to seats, usage or data, what incentives ensure that the customer’s educational purpose stays central?
An implementation offer answers that question only when it changes the work around the product. Adding more onboarding calls or a generic webinar does not make a product an implementation partner. The service needs to help the institution make better decisions and build its own capacity.
What a credible human-change offer contains
“Human change” should describe observable shifts in practice, not an emotional promise or a claim that a vendor can remake a school culture. A credible offer can be broken into six connected jobs.
Diagnose the learning problem before configuring the tool. Agree on the specific learner or teacher need, the existing practice and the constraints. If the problem is slow feedback on student writing, for example, a school should first understand where feedback is delayed and what teachers already do. A product demo is not a diagnosis.
Co-design the workflow with educators. Teachers and leaders should shape where the tool enters a lesson, what decisions remain theirs, and how students are introduced to it. This is not symbolic consultation after the product is chosen. Co-design can reveal timetable, accessibility, language, assessment and safeguarding requirements that a technical checklist misses.
Develop professional learning around real tasks. A one-off feature tour tends to teach navigation. Teachers also need time to practise with their own materials, compare outputs, discuss when not to use the tool, and adapt the workflow. Support should be role-specific: classroom teachers, school leaders, IT staff and safeguarding leads have different decisions to make.
Support practice during the first cycle. The hard questions often emerge only after use begins: a generated explanation is too advanced, a student follows hints without understanding, or an automated comment conflicts with the teacher’s rubric. A named support route and scheduled review help schools treat these moments as implementation evidence rather than user failure.
Measure more than activity. Logins, completed tasks and time saved can help diagnose adoption. They do not prove that learning improved. A school might also review student work, the quality of teacher feedback, learners’ ability to explain a strategy, or performance on a later task completed without the tool. The right measure depends on the original problem.
Transfer capability and plan an exit. Before the service ends, staff should know how to onboard colleagues, revise local guidance, evaluate the workflow and make a renewal decision. If every adjustment requires a paid vendor consultant, the school has bought dependency as well as software.
The Education Endowment Foundation’s implementation guidance offers a useful structure for this work: Explore, Prepare, Deliver and Sustain. It also emphasizes behaviours and context alongside process. That matters commercially. A responsible vendor should be prepared to recommend a smaller pilot, a different configuration, or no deployment where the educational need and operating conditions do not support the product.
A school example: test transfer, not just assisted performance
Imagine a secondary school exploring a writing-feedback platform. The illustrative goal is not “use AI in English lessons”; it is to help students revise arguments with stronger evidence and clearer reasoning.
During Explore, teachers review a sample of current writing and agree on what a stronger argument looks like. During Prepare, they select one unit, check data and accessibility requirements, and decide which feedback the platform may suggest and which judgments remain with the teacher. During Deliver, teachers compare platform feedback with their rubric, discuss mismatches, and ask students to explain which suggestions they accept or reject. During Sustain, the school reviews a later writing task completed without the tool and decides whether the routine should continue, change or stop.
This example does not claim a measured result. It illustrates a better evaluation question: can students carry the strategy into a new task when the support is reduced? A product may improve the first draft while leaving that question unanswered. The school therefore needs to distinguish supported performance from independent learning, and measure each for what it is.
For a practical bridge between policy and day-to-day routines, see From AI Policy to Classroom Practice. For a complementary view of company claims and educational evidence, read 10 Rising US EdTech Companies in 2026.
The commercial opportunity—and its limits
For an EdTech company, implementation services can create value in several ways: they can surface product friction earlier, improve the fit between features and real work, and help buyers make a more informed renewal decision. They may also support more durable relationships than a sale measured only by deployment. These are plausible business benefits, not established outcomes for every vendor.
There is a real trade-off. Software can be replicated at low marginal cost; expert support cannot. If a company sells unlimited bespoke consulting, service demand may grow faster than the product business. If it standardizes too aggressively, support may become a scripted rollout that ignores local context. The practical middle ground is a clear core package—diagnosis, role-based learning, implementation milestones and review—with transparent options for additional support.
A company can also make its offer more scalable by designing capability transfer into the product: in-context guidance, reusable lesson examples, configurable review prompts, administrator tools and documentation that a school can adapt. But automation should not be used to disguise the need for human judgment in high-stakes decisions. The point is to make professional expertise easier to apply, not to pretend it is unnecessary.
There are ethical limits too. Vendors should state what they can evidence, separate product activity from educational outcomes, explain data use and avoid placing the burden of adoption entirely on teachers. Schools need to consider staffing, workload, connectivity, accessibility and procurement rules. A service that works only in well-resourced settings may widen the gap it claims to close.
A buyer’s checklist for evaluating the offer
Before signing, school and university leaders can ask vendors:
What specific learning or teaching problem is this implementation designed to address?
Which educators and learners will help shape the workflow, and at what stage?
What professional learning is included, for whom, and how is it tied to actual practice?
What evidence will distinguish access, usage, improved work and independent transfer?
What data is collected, who controls it, and how can the institution export or delete it?
What support ends with the contract, and what capability remains with staff?
What conditions would lead you to recommend a pause, narrower pilot or non-renewal?
The answers should be concrete enough to appear in a statement of work. “We will drive adoption” is not a measure. A stronger agreement defines responsibilities, a baseline, a review point, what evidence will be examined and how the school can change course.
The next product may include a route to independence
Education companies will continue to sell software. The more useful strategic question is what must accompany that software for it to serve learning rather than merely enter a school. A durable offer connects a product to professional judgment, local routines and a way to test whether learners are becoming more capable.
The strongest version of “human change” is therefore not a vendor promising transformation. It is a vendor helping educators make a deliberate change, learn from evidence and own the next decision. The proof is not that a school needs the provider forever. It is that, over time, the school can use, adapt, question or leave the product with greater clarity.
For event organizers, school networks and institutions exploring these questions with their communities, discover Learning Coach’s keynotes.
References
Primary sources
OECD. (2026). OECD Digital Education Outlook 2026: Exploring Effective Uses of Generative AI in Education. Published 19 January 2026. https://www.oecd.org/en/publications/oecd-digital-education-outlook-2026_062a7394-en.html
Education Endowment Foundation. (2024). A School’s Guide to Implementation, third edition. Published 24 April 2024. https://educationendowmentfoundation.org.uk/education-evidence/guidance-reports/implementation/
Maciel, J. O., Pontes, J., Betim, L. M., et al. (2026). “Between Innovation and Inequality: A Systematic Review of EdTech Business Models and Their Contradictions.” Postdigital Science and Education. Published 14 October 2026. https://doi.org/10.1007/s42438-026-00709-6
