Nivelo Guide
The Spanish mistakes learners actually make, by level
Not a syllabus. 70 learners of Spanish made 6,959 graded attempts at real structures in Nivelo's AI conversation partner and writing coach, corrected one sentence at a time. Here is where they slip — and, if you are learning Spanish, where you probably do too.
Which Spanish grammar points do learners get wrong most?
The most common Spanish mistake our learners make is with the present tense — 638 wrong attempts out of 2,843 (22%), across 47 learners. It is the most common because it is the most used: every conversation runs on it. The next two are preterite (35% wrong) and ser vs estar (33% wrong).
| Grammar point | Learners | Attempts | Wrong | Wrong rate |
|---|---|---|---|---|
| Present tense | 47 | 2,843 | 638 | 22% |
| Preterite | 49 | 913 | 320 | 35% |
| Ser vs estar | 46 | 958 | 319 | 33% |
| Object pronouns | 36 | 468 | 201 | 43% |
| Por vs para | 38 | 374 | 172 | 46% |
| Imperfect | 34 | 405 | 135 | 33% |
| Present subjunctive | 45 | 182 | 121 | 67% |
| Present perfect | 28 | 213 | 104 | 49% |
| Future | 26 | 293 | 98 | 33% |
| Conditional | 14 | 136 | 65 | 48% |
Which Spanish grammar points are the hardest?
Frequency is not difficulty. Ranked by how often an attempt goes wrong, the hardest Spanish structure in our data is the present subjunctive: 67% of 182 attempts by 45 learners came out wrong.
| Grammar point | Wrong rate | Attempts | Learners |
|---|---|---|---|
| Present subjunctive | 67% | 182 | 45 |
| Imperfect subjunctive | 57% | 37 | 8 |
| Present perfect | 49% | 213 | 28 |
| Modal speculation | 48% | 93 | 12 |
| Conditional | 48% | 136 | 14 |
| Por vs para | 46% | 374 | 38 |
| Object pronouns | 43% | 468 | 36 |
| Preterite | 35% | 913 | 49 |
What trips learners at each level?
Learners are grouped by the level our free test placed them at. A structure is listed for a level only when at least three learners at that level attempted it twenty times or more between them.
- A1 — 19 learners. Hardest: preterite (100%), present tense (31%), ser vs estar (25%).
- A2 — 24 learners. Hardest: present subjunctive (71%), present perfect (53%), modal speculation (53%).
- B1 — 9 learners. Hardest: present perfect (57%), por vs para (57%), object pronouns (39%).
- B2 — 10 learners. Hardest: present subjunctive (60%), imperfect subjunctive (58%), modal speculation (46%).
- C1 — 6 learners. Hardest: present subjunctive (55%), object pronouns (40%), por vs para (35%).
What do the mistakes actually look like?
These are learners' own fragments, exactly as our correctors marked them — shortened to the part that changed, with nothing that could identify anyone.
| Ser vs estar: what a learner wrote | The fix | Why |
|---|---|---|
| Hoy estuve bien. | Hoy me fue bien. | Minor idiomatic issue: 'estuve bien' sounds like a health-related state; 'me fue bien' better expresses 'my day went well.' |
| En mi día el libre estuve bueno. | Mi día libre fue bueno. | Word order error ('el libre' should come before or without 'el') and ser/estar confusion — use 'ser' to evaluate how a day was overall, not 'estar'. |
| esto estaría un herramienta útil | esto sería una herramienta útil | Ser/estar confusion (should be 'sería' for a permanent characteristic) plus gender agreement error ('una herramienta', feminine). |
| Object pronouns: what a learner wrote | The fix | Why |
|---|---|---|
| pongo mi ropa | me pongo la ropa | 'Vestirse'/'ponerse ropa' is reflexive — you need 'me pongo', not just 'pongo'. |
| Todavía tuve divertido | Todavía me divertí | Para expresar 'pasarlo bien' se usa el verbo reflexivo 'divertirse', no 'tener divertido', que no existe en español. |
| Mi gusta hacer cosas | Me gusta hacer cosas | The correct indirect object pronoun for 'gustar' is 'me', not 'mi' (which means 'my'). |
| Por vs para: what a learner wrote | The fix | Why |
|---|---|---|
| por el desayuno | para desayunar / de desayuno | 'Por' is the wrong preposition here; use 'para desayunar' (to have for breakfast) or 'de desayuno'. |
| necesitan el tiempo a mejora | necesitan tiempo para mejorar | 'A mejora' no es correcto; se necesita el infinitivo 'mejorar' precedido de 'para', y 'tiempo' no lleva artículo aquí. |
| fui solitario por los pasatiempos | estuve solo con mis pasatiempos | 'Ser solitario' describe un rasgo de personalidad permanente; para expresar un estado temporal de estar sin compañía se usa 'estar solo'. |
| Present perfect: what a learner wrote | The fix | Why |
|---|---|---|
| No había hacer mucho hoy | No he hecho mucho hoy | Mixing the imperfect auxiliary 'había' with an infinitive isn't grammatical. |
| todavía desarrollé mis habilidades | he desarrollado mis habilidades | Con 'todavía' en este sentido (aun así) y una acción con relevancia hasta el presente, el pretérito perfecto compuesto es más natural que el pretérito simple. |
| nuestro alcalde ha construido un parque nuevo | nuestro alcalde construyó un parque nuevo | Como se especifica 'el año pasado' (tiempo terminado y delimitado), se debe usar el pretérito indefinido, no el pretérito perfecto compuesto. |
Why does this look different from a textbook's list?
A textbook orders grammar by when it is taught. This page orders it by when it goes wrong. That is why the present tense — taught in week one — tops the first table: it is attempted thousands of times, so even a low rate makes a lot of mistakes. And it is why the present subjunctive tops the second: fewer attempts, but most of them wrong.
How was this measured?
Nivelo's AI conversation partner tags every structure a learner attempts in a session and whether it came out right; the writing coach lists each correction with the fix. The numbers here are those records added up across 70 learners and 6,959 attempts (555 written corrections from 36 of them), last recounted on 2026-09-24. It is a measurement of our learners, not a study: the grader is a language model, levels come from our free test (aligned to the CEFR, not an official certification), and the sample is who happens to use the app.
Start with your level, not the top of the table
Start with the row for your level above, not the top of the first table. If you do not know your level, take the free Spanish level test — it takes five minutes and places you on the same scale this page uses. Then open the grammar reference for the structure your level gets wrong most: the drills there are now built around these exact mistakes.
Is the subjunctive the hardest thing in Spanish?
In our data the hardest structure is the present subjunctive, wrong 67% of the time. Whether that is "the hardest thing" depends on what you count: by sheer number of mistakes, the present tense wins, because it is used far more often.
How often does this page update?
The counts are recomputed from the app's records and republished with the site; the date of the last recount is in the method section above (2026-09-24). As more learners use the app, the tables get more stable and more levels reach the three-learner floor.
Take the first step
Find out your current CEFR level in 5 minutes
Practice on your phone too: Nivelo is on the App Store.