Top News

Edtech has spent years optimizing for outcomes that are easy to measure but increasingly hard to justify. Completion rates, time spent, and certificates issued became convenient stand-ins for learning. They created a system that looked efficient on dashboards, even when the learner walked away unable to use the skill in a real setting.

That gap is no longer easy to overlook. Learners are asking a more direct question. What can I do with this today?

“At SpeakX, we have moved away from completion as a measure of success,” says Arpit Mittal, Founder of SpeakX. “The real question is simple. Did the user actually use English today? That reflects progress far better than finishing a module.”

Once usage becomes the benchmark, the design of learning cannot remain the same. Long pathways that defer value lose relevance. The product has to deliver something usable from the first interaction and continue to reinforce it in ways that fit into everyday life.

From Completion to Capability

Measuring learning through usage changes what progress looks like. It moves away from accumulation and toward action in specific situations.

“The learner is not chasing completion anymore,” Mittal says. “They are chasing confidence, whether that is answering a question in an interview or speaking up at work.”

Confidence is built through repetition in context. It comes from saying something out loud, correcting it, and trying again without the pressure of being evaluated. SpeakX leans into this by structuring learning around small actions that can be practiced and repeated. A sentence delivered clearly, a response given without hesitation, or a short exchange handled smoothly becomes a marker of movement.

When these signals appear early, the product stops feeling like a course and starts functioning as a tool. Engagement follows usefulness. If the platform continues to reflect situations that matter, users return because it fits into what they are trying to improve.

The Real Moat Isn’t AI

The current conversation around edtech often assumes that access to better AI models will define who wins. It is a neat argument, but it misses where most products struggle.

“The idea that AI alone will create a moat is flawed,” Mittal says. “The real advantage comes from how deeply you understand your users and how you apply that understanding to the product.”

Technology is available. Consistent behaviour is not. Getting an adult learner to show up, engage, and improve over time requires more than content delivery. It requires a system that accounts for hesitation, inconsistency, and competing priorities.

SpeakX builds around this reality. Roleplay replaces passive instruction. Repetition is structured so it does not feel mechanical. Visual context grounds conversations in familiar settings. Even gamification reflects how learning is often shared within families, where progress becomes visible and socially reinforced.

Trust is shaped just as deliberately. The AI mentor, Sia, is presented as AI, which lowers the barrier to participation.

“For many users, it removes the fear of judgment,” Mittal says. “It creates a safe space to practice.”

Human input remains central. The team defines how the AI teaches, the tone it uses, and how it responds across different scenarios. AI handles scale, but the experience is directed by decisions that come from observing how users actually behave.

Where Demand Is Most Intentional

A large share of SpeakX’s growth comes from Tier 2 and Tier 3 cities, markets that are often reduced to assumptions around price sensitivity and inconsistent engagement.

The behaviour on the ground suggests otherwise.

“Users in these regions are very clear about why they are learning,” Mittal says. “English is linked to jobs, career growth, and access to opportunity.”

Clarity of purpose changes how users approach the product. Learning is tied to a defined outcome, which keeps engagement steady. The user is not browsing. They are working toward something that affects their next step.

This also reflects in how the business is built. SpeakX chose to anchor itself in profitability early.

“From day one, the entire team was aligned to building a profitable company,” Mittal says. “We stayed lean and focused on clear value creation.”

Reaching profitability before raising capital allowed the company to scale without relying on funding to cover gaps in the model. Growth builds on what already works, rather than compensating for what does not.

Beyond Translation, Toward Confidence

Advances in voice AI and real-time translation have raised a familiar question. If technology can handle communication, why invest time in learning the language.

The answer lies in how communication is experienced, not just how it is transmitted.

“Translation tools can help with basic communication, but they do not replace confidence,” Mittal says. “Language shapes how you are perceived and how you participate.”

Speaking directly carries a different weight. It affects how individuals are assessed in interviews, how they contribute in meetings, and how they navigate everyday interactions without pause.

As SpeakX expands into markets such as Latin America and Southeast Asia, this distinction becomes more pronounced. Localization requires understanding what drives the decision to learn in each context.

“Why someone wants to learn English varies across markets,” Mittal says. “The product has to reflect those aspirations.”

Motivation, social context, and expectations differ across regions. Building for these differences means designing the experience from the ground up, rather than adapting a fixed model.

What replaces older metrics is not another dashboard indicator. It is visible in how people respond when they no longer need to translate in their head, when they speak without second-guessing, and when participation becomes easier.