Why our planning model caps app-only study at B1

The B1 app-only ceiling is a conservative model assumption, not a proven universal limit. Here is its rationale, what the category means, and where it fails.

Updated Sep 1, 2026Kevin Fan

Every estimate on this site can return a refusal. The most common one is this: under this planning model, app-only study is capped at B1.

That is a conservative modelling judgement, not a universal limit established by the NFATC source. It deserves its rationale and boundary in the open.

What the model says

Our estimates multiply an FSI baseline by a factor for how you study. The app-only factor is not one number — it grows with the level you are aiming at:

Aiming at	App-only multiplier
A1	1.3×
C1	3.5×

And above B1 it stops being a multiplier at all, because the model marks the input unreachable under its own ceiling assumption.

Concretely, Spanish to B2 at an hour a day, five days a week:

Method	Hours	Reaches B2?
Living in-country	397	yes
Intensive class (NFATC-style)	504	yes
1-on-1 tutor	529	yes
Structured self-study course	800	yes
App only	1,486	no

The 1,486 is shown struck through on the method chart, because it is not an answer under this model. It is the arithmetic result before the assumed ceiling is applied.

Why the multiplier grows

This is the part worth understanding, because it explains why apps feel so effective at first and so stuck later.

At A1, recognition-heavy practice overlaps more with the task. Vocabulary and basic patterns are bounded enough for drills and spaced repetition. That is the rationale for the estimated 1.3× factor; the percentage itself is not a measured app effect.

At B2, the task has changed. B2 is producing sentences you have not seen before, under time pressure, that are correct — and finding out when they are not. The skill is generating novel output and having it corrected.

The app-only category in this model assumes no sustained open-ended speaking or writing with human-quality feedback. A product or routine that supplies those things no longer matches the input, even if an app is part of it.

Hence the modelled shape: relatively efficient at the bottom, less efficient in the middle, and capped above B1.

What this is not

Not "apps do not work". They are the best tool available for vocabulary and early forms, they are the reason many people start at all, and the free ones are genuinely free. Our recommendations include zero-commission options for exactly this reason.

Not a precise threshold. B1 is where the model places the ceiling; the real edge is fuzzy, depends on the app, and some people push past it by supplementing. A person doing app work plus conversation practice is not an app-only learner, and the model treats them differently.

Not a claim about any specific product. The ceiling describes the model's recognition-heavy, no-corrected-output category. Products and routines vary.

The practical version

If your target is A2 — travel, courtesy, reading signs — an app can take you there and the price is right.

If your target is B2 — work, study, arguing a point — the app is a component and not the plan. What is missing is corrected output, and the cheapest way to buy it is not another subscription. It is a person, for fewer hours than you think.

Work out what your own target costs by method, or read how the method factors were set, including which of them are estimates rather than published figures. Several are, and they are marked.

Selected from the source graph, with fallback links that keep every article connected.

The 1,000-hour Spanish claim closely matches an English exam table
A commercial page gives an unattributed figure of about 1,000 hours for Spanish. Its level table closely matches Cambridge English’s cumulative CEFR guidance, but that correspondence does not prove where the table came from.
Why our model adds 30% at A2 and 70% at C1 for self-study
The modelled self-study penalty grows with level because the task shifts from recall toward production. The percentages are planning assumptions, not measured effects.
What A1 to C2 actually mean, and what they deliberately do not
CEFR describes what you can do at six levels and publishes no study hours, on purpose. The hour figures attached to it come from one exam board, about its own exams. Knowing which is which fixes most confusion in this field.
The cheapest tutor hour is the one you buy last
Our model assigns tutoring 1.05× and makes self-study less efficient as levels rise. The coefficients are disclosed assumptions; this is the spending logic they imply.
What a language you already speak is actually worth, in hours
Our model assigns a 30% Spanish→Portuguese discount and 8% for Russian→German. The phenomenon is documented; the hour coefficients are disclosed estimates.

Sources

Every figure above, and where it came from. A post that traces other people's unsourced numbers does not get to have any of its own.

Foreign Language Training — US Department of State, National Foreign Affairs Training Center
The current NFATC training timelines: a typical week is 23 class hours plus 17 hours of self-study, aimed at an integrated ILR 3 in Speaking and Listening. Category I lists 24–30 weeks; using the page's typical class week, this site derives 552–690 class hours. Categories II–IV print 828, 1,012 and 2,200.
Guided learning hours — Cambridge English
The cumulative shape between CEFR levels. Cambridge publishes guided-learning-hour bands for its English qualifications (A2 180–200, B1 350–400, B2 500–600, C1 700–800, C2 1,000–1,200). Our factors are normalised against C1 rather than fitted to every band; for example, 0.27 × the 750-hour C1 midpoint is 202.5, just above the A2 band.