In the tutor vs app debate, the research points the same way for adults: apps build reading, listening, and vocabulary at a very low cost, and tutors build speaking, correct your errors, and keep you going long enough to reach a usable level. Apps also struggle with follow-through: on a typical day, fewer than half of Duolingo's monthly users open the app. Most working adults who actually get to conversational Spanish or French use both, in a fixed ratio: a tutor for speaking, an app for daily review.
This guide compares the two options on the questions that matter to an adult with a job: what each one teaches, what each one skips, what each one costs per hour of real progress, and how to combine them. If you want to know how many hours the whole project takes, read our guide on how long it takes to learn Spanish first. If you are an employer looking at options for a team, our guide to corporate language training covers the same choice at company scale.
Are language apps effective for adults?
Yes, for the receptive skills, and the evidence is better than most tutors admit. A 2021 study in Foreign Language Annals tested 225 US adults who had finished Duolingo's beginner Spanish or French course with no other instruction. They reached Intermediate Low in reading and Novice High in listening on the ACTFL scale, which was statistically no different from university students at the end of their fourth semester. In French reading, the app learners scored higher.
Speaking gains are possible too, with enough hours. Michigan State University researchers tracked 54 adults who studied Spanish on Babbel for twelve weeks and found that 59 percent improved their oral proficiency by at least one ACTFL sublevel. Among those who put in fifteen hours or more, the share rose to 75 percent. The strongest predictor of progress on every measure was time spent in the app.
Two cautions. Both studies were funded by the app companies. That does not make them wrong, but it does make them best-case: the participants were volunteers who completed the course, not the general population of people who download an app in January. And "Novice High" listening, after roughly a semester's worth of study, means you can catch familiar phrases in slow speech. It does not mean you can follow a coworker on a call.
Where do language apps fall short?
On the two things adults say they want most: speaking with a real person, and sticking with it. The same Michigan State group followed nine learners through a semester of Turkish on Duolingo. Progress correlated with time on the app, but motivation fell over the semester, and the learners named the repetitive exercise types and the absence of interaction as the reasons.
Dropout is the bigger problem. In a 2025 natural experiment published in Learning and Individual Differences, researchers followed 601 adult beginners in Spanish or French using Duolingo on their own schedule for six months. Many had stopped using the app before the six months were up; the authors describe attrition in app-based learning as high. The trait that best predicted who kept going was perseverance, not talent, prior knowledge, or age. That is a polite way of saying the app did not supply the discipline; the learner had to bring it.
Duolingo's own numbers tell the same story at scale. In its second-quarter 2026 shareholder letter, the company reported 58.7 million daily active users against 140.6 million monthly active users. On any given day, fewer than half of the people who used the app that month opened it. For a habit that needs daily contact to work, that ratio is the product's honest efficacy rate.
Then there is the skills gap that apps rarely test for. Reading runs well ahead of listening in every app study above, and speaking, when it is measured at all, is scored from recordings. Nobody interrupts you. Nobody asks you to rephrase. Nobody notices that you have said "yo soy caliente" to mean you are warm.
What does a private tutor do that an app cannot?
Three things: correct you in the moment, make you produce language you have not rehearsed, and show up at a fixed time so that you do too.
Correction is the one with the strongest evidence. A meta-analysis of 15 classroom studies with 827 learners, published in Studies in Second Language Acquisition, found that oral corrective feedback had significant and durable effects on language development. The effects were largest for prompts, where the teacher signals an error and makes the learner fix it, rather than recasts, where the teacher simply supplies the right form. An app can only recast. A tutor can wait.
Tutoring itself has a large research base, though most of it comes from schools rather than adult language learners. A 2020 review of 96 randomized studies by economists at Northwestern and the University of Toronto found that tutoring raised achievement by 0.37 standard deviations on average, a large effect by the standards of education research. Effects were stronger with trained teachers than with volunteers. Treat that as a signal about the format, not a promise about your Spanish. The mechanism transfers: an expert watching one learner and adjusting in real time.
The third advantage is scheduling. A tutor on your calendar at 7 a.m. Tuesday and Thursday is an appointment. An app on your phone is an option, and options lose to whatever else is on the phone. This is why the attrition numbers look so different between the two formats even when the learners are equally busy.
Tutor vs app: the comparison
| Question | Language app | Private tutor |
|---|---|---|
| Cost | Free tier or a low monthly subscription | Roughly 10 to 40 dollars an hour on marketplaces; 55 dollars and up with a vetted, qualified tutor |
| Reading and vocabulary | Strong; comparable to four university semesters after the beginner course | Depends on homework; the hour is better spent talking |
| Listening | Lags reading by one or two ACTFL levels | Built in: every session is live listening at natural speed |
| Speaking | Scored from recordings, if at all; no interaction | The main event; forty minutes of production per hour |
| Error correction | Right or wrong, with the model answer | Prompts you to self-correct, explains the pattern, returns to it next week |
| Accountability | Streaks and notifications; high dropout, because the habit depends entirely on the learner | A person expects you at a set time |
| Fit to your job | Generic course sequence | Your industry vocabulary, your meetings, your emails |
| Flexibility | Any time, five minutes at a stretch | Scheduled; usually 45 to 60 minute blocks |
The marketplace price band comes from italki's own 2026 pricing guide, which puts professional Spanish teachers at about 10 to 40 dollars an hour, with community tutors below that. Cheaper is not always worse, but at the low end you are usually paying a native speaker to chat, not a trained teacher to correct.
Which one costs more per hour of real progress?
The tutor costs more per hour and often less per level. Here is the arithmetic. The US Foreign Service Institute, which has trained diplomats since 1947, estimates that Spanish and French take 24 to 30 weeks of full-time study, or 552 to 690 class hours, to reach professional working proficiency. Those are classroom hours with an instructor, plus homework.
No working adult buys 600 tutor hours. Nobody needs to. The question is which hours need a human. Reading practice, flashcards, and grammar drills do not; an app or a free podcast covers them for nothing. Speaking, listening at natural speed, and being corrected do. If those take two hours a week with a tutor, a year of lessons costs roughly 5,000 to 6,000 dollars at a qualified-tutor rate, and gets most motivated beginners to B1 in conversation. Twelve months of an app at the same effort level costs almost nothing and, on the evidence above, gets a large share of learners nowhere at all because they stopped in month four.
Put differently: the app is cheap per hour and expensive per finished learner. The tutor is the reverse.
What is the best way to learn a language as an adult?
Use both, in a fixed weekly ratio, aimed at a named CEFR level. This is what the completion data and the skills data both point to, and it is what nearly every adult we have watched reach B2 actually did.
- Pick the target before the tool. B1 lets you handle a customer or a coworker; B2 lets you run a meeting. Our guide to CEFR levels explained defines each level in workplace terms so you can choose one.
- Book the tutor first. Two sessions a week, 45 to 60 minutes each, at the same times every week. Speaking only. Bring the vocabulary from your job.
- Add the app for the other five days. Fifteen to twenty minutes of review, reading, and listening. Treat it as homework the tutor assigns, not as the course.
- Ask the tutor to log your errors. A running list of the mistakes you make under pressure is worth more than any streak counter. Recurring items become next week's warm-up.
- Test every twelve weeks. A short speaking assessment against the CEFR scale tells you whether the ratio is working. If reading is racing ahead of speaking, shift minutes from the app to the tutor.
About seven hours a week on this split puts B1 within a year for most beginners in Spanish, and a bit longer for French. Companies apply the same logic when they design programs for staff; our comparison of language training formats shows how blended programs pair live coaching with self-study for the same reasons.
When is an app alone enough, and when is a tutor alone enough?
An app alone is enough if your goal is to read menus, follow signs, and understand slow speech on a trip, or to keep a language you already speak from fading. It is also the right first step if you are not yet sure you will stick with the project: spend a month on the free tier, and if you are still opening it in week five, you have earned the tutor.
A tutor alone is enough if you already read and understand at an intermediate level and your only gap is speaking. Many adults who took Spanish or French in school fall here. They do not need another vocabulary course. They need two hours a week of being made to talk, and a person who notices when they revert to English.
Everyone else, meaning adults who start from zero and need to speak the language at work within a year or two, should plan on both from the first week.
Learn with a private tutor at Edlingo
Edlingo pairs adult learners with qualified, vetted tutors for online Spanish classes, French classes, English, and Arabic, from 55 dollars an hour. Every learner starts with a CEFR placement, sets a target level and a date, and gets a weekly plan that tells you exactly what to do with the app in between sessions. Your tutor keeps the error log, so nothing you say wrong twice goes uncorrected.
Book a Free Placement Session →
Not sure how many hours your goal will take? See how long it takes to learn Spanish. Choosing a level to aim for? Start with CEFR levels explained, or contact us with your situation.