Not a generic list. These are 50 errors taken from marked homework by Turkish speakers, each one repaired and then explained by what Turkish actually does differently. The heaviest areas here are pronunciation, articles and determiners, prepositions. Nothing on this page is behind a sign-up.
An error you can name the cause of is an error you can stop making.
Turkish has no who, which or that. It builds a relative clause as a participle placed before the noun: beni muayene eden ebe is literally 'me examine-ING midwife' – 'the midwife who examined me'. The whole clause sits in front, and the linking word simply does not exist. English is the mirror image: the noun comes first, then who / which / that, then the clause after it. To produce an English relative clause correctly, a Turkish speaker has to reverse the word order and insert a word their language does not have. That is two operations at once, in real time – which is why a whole lesson of clean relative clauses is marked as absorbed rather than merely untested.
The takeawayEnglish hangs its describing clause after the noun and joins it with a word Turkish has no equivalent for – the joining word is the part to remember.
Turkish orthography is transparent – every letter is pronounced, one letter to one sound, always. You can read a Turkish word you have never seen before and say it perfectly. English is opaque: letters go silent (bride), clusters collapse (raised = 'rayzd'), and the same letter says different things in different words. So the instinct to say every letter, including the final -e, is exactly right for Turkish and exactly wrong for English. That matters for how the problem is treated: the fix is not more care, because more care makes the transfer stronger, not weaker. It is a small set of rules – silent -e, -ed only doubles after t or d, letters said alone – that override the reading instinct.
The takeawayYour reading instinct is not a weakness; it is the correct instinct for a language English is not. Carry three small suspicions instead.
Turkish has no articles and no free-standing prepositions; it welds those meanings onto the noun as case endings. Ready for the lesson is derse hazır – the -e ending is the 'for' or 'to'. In a flat is dairede – the -de ending is the 'in'. The relation is fully expressed by the time the noun is finished, so nothing is missing from the Turkish speaker's point of view. When the sentence moves into English, the preposition has nowhere to hide: it has to become a separate little word standing in front of the noun. That extra step, and only that step, is what occasionally drops.
The takeawayTurkish says the preposition inside the noun; English says it in front. Nothing is being forgotten – a step is being added.
Turkish forms conditionals with the suffix -sA on the verb – gelir-se-m ('if I come'), gel-se-m ('if I came / were to come') – plus a clear result clause. The system is regular, and it maps well onto the English if-clause: the learner already has the idea of an unreal or hypothetical condition, already distinguishes degrees of it, and already builds a result clause to go with it. That is why the English conditional target produced no errors at all. The one English-side rule left to keep is a purely formal one, with no Turkish counterpart to violate: no would inside the if-clause.
The takeawayIf your first language has a dedicated conditional suffix, English conditionals are a re-packaging, not a new idea – just keep would out of the if-clause.
Turkish marks 'motion towards' with the dative suffix -e/-a attached to the noun – hastane-ye, 'to the hospital'; Londra'ya, 'to London' – so the relation lives inside the word, with no separate preposition anywhere in the sentence. English uses the free word TO instead. Because the first language has already 'said' the to via the suffix, the English word feels redundant and is easy to omit; the sentence sounds complete to the speaker before the preposition has been added. Both came the hospital and move London are this exact drop, and the fix is to store the destination verbs with TO built into them as fixed units.
The takeawayCome to, go to, move to, travel to, return to – store the preposition inside the verb, not as a separate decision.
Turkish negates verbs and non-verbs by completely different means. Verbs take a negative suffix inside the word: gidiyorum ('I am going') becomes gitmiyorum ('I am not going'). But adjectives and nouns – non-verbal predicates – are negated with the separate word değil: endişeli değilim ('worried not-I-am'), hemşire değilim ('nurse not-I-am'). English has exactly the same two-branch arrangement, drawn along a very slightly different line: don't and doesn't negate verbs, while am / is / are not negate adjectives and nouns. So I don't worried is the verbal branch applied to an adjectival predicate. The good news is that no new distinction is being learned – a line the speaker already understands is being re-drawn.
The takeawayIf very fits in front of the word, it is an adjective – so negate it with BE, not with DON'T.
Turkish expresses the difference between a felt state and a causing property perfectly clearly – endişeli ('worried') beside endişe verici ('worry-giving', worrying); yorgun ('tired') beside yorucu ('tiring'). So no concept is missing here, and it would be wrong to treat this as a conceptual difficulty. It is simply that English builds both members of the pair from the same verb with two different endings, so the two forms look almost identical on the page and have to be stored as a matched pair rather than derived. Learn them in pairs, always aloud, in real sentences: I am concerned, the trend is concerning; she is exhausted, the labour has been exhausting; the mother is reassured, the heart rate is reassuring.
The takeawayPeople take -ed, things take -ing – and the pairs are memorised together, never derived one from the other.
Turkish forms the comparative with daha before the adjective (daha büyük, 'bigger') and the superlative with en (en büyük, 'biggest'). Both are separate words, exactly like English more and most – which is why the analytic half of English comparison gives a Turkish speaker no trouble at all. The English complication is the half Turkish does not have: short adjectives take endings instead (safe → safer → safest; high → higher → highest), and using more with them is an error. The predicted Turkish-speaker error is therefore very specific – more safe, more big – and its absence from a session is meaningful evidence rather than mere quiet.
The takeawayHalf of English comparison is already yours; the half to watch is the short adjective, which takes an ending and refuses more.
Grouped by grammar point, heaviest first. Every line is real.
Pronunciation 8
bride said as bri-de
bride – BRYD, one syllable
The English final -e is usually silent, and its real job is to reach back and change the vowel before it – here turning the i into a long 'eye' sound.
WhyIn Turkish a final -e is always pronounced, so bride looks as though it must be two syllables. Turkish orthography is transparent: every letter is pronounced, one letter to one sound, always.
raised said as ra-i-sed
raised – RAYZD, one syllable
The ai is a single 'ay' sound and the -sed here is just 'zd' – no extra syllable. The -ed ending only adds a whole syllable after a t or d sound.
WhyTurkish spelling is one letter to one sound, so every written letter is expected to be sounded; English collapses clusters that Turkish would pronounce in full.
BMI said as a word
BMI – BEE-em-EYE, said one letter at a time
An initialism is said letter by letter, each letter wearing its own name. The trap is the last letter: English I on its own is EYE.
WhyTurkish i says 'ee', so the English letter-name EYE has to be learned separately – the same 'eye' sound already met inside bride.
SIG-nif-i-cant
significant – sig-NIF-i-cant
The stress falls on the second syllable, not the first, and the ending is '-kant', not '-kaant'.
WhyTurkish stress is broadly regular and usually falls on the final syllable, shifting predictably as suffixes are added. English stress is lexical and unpredictable, so it has to be stored word by word.
oo-TER-us
uterus – YOO-ter-us
Three syllables, starting with 'YOO' as in 'you' – the English u is not the Turkish u.
WhyTurkish spelling is transparent, so the written vowel is expected to give its own sound; English vowels routinely say something the spelling does not show.
WOO-men
women – WIM-in (singular woman = WUM-an)
The plural sounds nothing like its spelling: women is WIM-in, while the singular woman is WUM-an. The spelling barely shows the difference; the sound carries it.
WhyIn a transparent orthography the written o would be read as an o. English writes women and says WIM-in, so the pair has to be learned by ear.
par-tic-ly
particularly – par-TIC-u-lar-ly
Five syllables, with the beat on TIC. Do not collapse the middle of the word.
dis-tin-wish
distinguish – dis-TING-gwish
TING-gw sits in the middle of the word – do not drop the g before the w.
Articles and determiners 6
We recommend vaccine because it protects the baby.
We recommend the vaccine because it protects the baby.
A specific thing already in play takes the. First-mention countable takes a/an. An uncountable or a whole field takes nothing.
We expect it to ease after the week 22nd.
We expect it to ease after week 22.
Numbered units of a schedule – week 22, day 5, room 4 – take no article and no ordinal.
I explained the all options to her.
I explained all the options to her.
All comes before the determiner, not after it: all the options, all my patients, all that time.
I have been working in the big hospital since 2017. I prefered the nursing because I like helping people.
I have been working in a large hospital since 2017. I chose nursing because I like helping people.
First mention of the hospital → a. A whole field – nursing, dentistry, physiotherapy – takes no article at all.
she will go to the theatre
she will go to theatre
A process used as a function takes no article in clinical English: we go to theatre, keep a patient in hospital – the same pattern as at university. A named, specific place still takes the: to the labour suite, to the postnatal ward.
a change would not have been missed
the change would not have been missed
A specific, identifiable thing already under discussion takes the, not a.
WhyTurkish has no articles at all, so both the presence and the choice of an English article are additions made entirely on the English side – which is why the pattern drops in both directions.
Prepositions 6
The team decided a caesarean section.
The team decided on a caesarean section.
Decide takes on before the thing chosen; decide to before the action.
I am ready this lesson
I am ready for this lesson
Ready takes for before the thing you are ready for.
WhyTurkish has no free-standing prepositions; it welds those meanings onto the noun as case endings. Ready for the lesson is derse hazırım – the -e ending is the 'for'. In English the meaning has nowhere to hide and must become a separate little word in front.
I live a flat
I live in a flat
You live in a place – in a flat, in a house.
WhyTurkish carries this inside the noun with a case ending: dairede is 'in a flat', where the -de ending is the 'in'. The relation is already expressed by the time the noun is finished, so the separate English word is easy to leave out.
when I came the hospital
when I came to the hospital
Motion towards a place needs TO. Move, go, come, travel and return all take TO before a destination.
WhyTurkish carries this on the noun as the dative suffix -e/-a: hastane-ye geldim, literally 'hospital-to I-came'. The relation lives inside the word, so the English preposition is easy to leave out.
I would move London
I would move to London
The same motion-to rule: a destination takes TO after move, go, come, travel, return.
WhyTurkish marks 'motion towards' with the dative suffix -e/-a attached to the noun – Londra'ya – so the relation has already been expressed and the separate English word has nothing to prompt it.
I watch to her observations
I watch her observations
Watch takes a direct object with no preposition; listen takes TO. The pair splits in English where it does not split in Turkish.
WhyTurkish governs many objects through the accusative directly: seyretmek ('watch') and dinlemek ('listen') both take a direct object, so English's decision to give one of them a preposition and not the other is unpredictable from the first language. The learner's file shows errors in both directions – listen the radio and watch to the movie.
Comparatives 4
It is more important to document it immediately.
It is important to document it immediately.
A comparative with nothing to compare against. If there is no than, there is usually no more.
her dilation is more slow
her dilation is slower
A one-syllable adjective takes -er, never more.
WhyTurkish forms the comparative with the separate word daha before the adjective (daha büyük, 'bigger'), exactly like English more – so the analytic half transfers free, and the half Turkish does not have, the -er ending on short adjectives, is where the error falls.
The most safe option
the safest option
Short adjectives take -est for the superlative, never most: safe → safer → safest; high → higher → highest.
WhyTurkish forms the superlative with the separate word en (en büyük, 'biggest'), which matches English most exactly – so most is the form that feels natural, and the English -est switch has no Turkish counterpart.
is the more stable it has been
is the most stable it has been
The frame 'the ... it has been' takes a superlative, not a comparative.
Conditionals 4
If I would have more staff, I would feel calmer
If I had more staff, I would feel calmer
No would inside the if-clause. The second conditional puts a past form in the if-clause and would in the result clause.
If the uterus was not contract well, I would commence a fundal massage.
If the uterus were not contracting well, I would commence a fundal massage. (or: did not contract)
The second conditional needs a past form in the if-clause – were not contracting or did not contract, never was not contract. The result half (would commence) was already right.
If there were enough staff, it could have been easier.
If there had been enough staff, it would have been easier.
Match the clocks. The second conditional imagines an unreal present or future (if there were ... I would ...); the third looks back at an unreal past (if there had been ... I would have ...). Here the shift was already over, so both halves belong in the past.
If I am doing it correctly...
If I am doing it correctly, my confidence grows.
An if-clause is only half a sentence – finish the pair with a main clause. This is the same 'complete the sentence' step as the relative clause: the subordinate part cannot stand alone.
WhyTurkish puts the subordinate material first and can stop there in ways English cannot, so the hanging clause appears most often in throwaway lines rather than in drilled work.
Relative clauses 4
The midwives who has been working in the delivery room.
The midwives who have been working in the delivery room will be supervised.
Two things: a plural subject inside a relative clause takes a plural verb (midwives who have), and a relative clause is not yet a sentence – the main clause still needs its own verb.
WhyTurkish puts the describing part in front of the noun and stops there, so there is no structural prompt to carry on to a main verb once the description is finished.
the birth plan that we agreed it last week
the birth plan that we agreed last week
When that / which / who is the object of the relative clause, it already is the object – do not name it again with it, him or her at the end.
the patient who I reassured her
the patient who I reassured
Say the object once. The relative pronoun who has already done the work; the extra pronoun is one word too many.
My shift that is very busy this night.
My shift is very busy tonight.
If you only want to say something about the noun, you do not need a relative clause at all – drop the that and let the verb do its job. And the fixed adverb is tonight, not this night.
Collocation 3
Her dilatation was higher than at the last check.
Her dilatation was greater than at the last check.
The grammar was never wrong; the lexis was. Higher works for blood pressure and temperature, not for dilatation (greater) or pain (worse, more severe).
She had a birth at 39 weeks and reported stomach pain.
She had a spontaneous vaginal delivery at 39 weeks and reported abdominal pain.
Both were grammatical and both were wrong for the register. OET marks the profession's own words, not the general ones.
there is a long hospital
there is a large hospital
Long describes extent in one dimension; a building of great size is large or big.
Subject-verb agreement 2
Both the mother and the baby was stable.
Both the mother and the baby were stable.
Both A and B is always plural, however singular each half looks.
if everything stay stable
if everything stays stable
Everything is grammatically singular and takes the -s form of the verb.
The copula 2
Many women experience sickness; it usually a normal.
Many women experience sickness; it is usually normal.
Turkish lets an adjective stand alone – o normal. English cannot: every finite clause needs a verb, and where there is no action, be is the verb.
I don't worried about the result.
I was not worried about the result.
Negate be with not, never with do. The Turkish değil sits after the adjective and tempts an auxiliary that English does not want here.
Modal verbs 2
If it continues, you will see a doctor.
If it continues, you should see a doctor.
will predicts; should advises. In clinical writing the difference is not stylistic – it is the difference between a forecast and a recommendation.
When I supervise them, I would be tired
When I supervise them, I will be tired
A real future event takes will. Would is for the unreal or the hypothetical.
participial adjectives 2
The chart was worried.
The chart was worrying.
One test: who is feeling it? People take -ed. Things – charts, traces, trends, observations – take -ing.
this is worried
this is worrying
-ed describes how a person feels; -ing describes what a thing is. A person is worried; a falling heart rate is worrying. Say 'this is worried' and you have accidentally given the observation feelings.
WhyThis is not a gap in Turkish. Turkish distinguishes the two perfectly well – endişeli ('worried') beside endişe verici ('worry-giving', worrying); yorgun ('tired') beside yorucu ('tiring'). There is no concept to acquire, only an English pairing to memorise, because English builds both from the same verb with two nearly identical endings.
Countability 1
among woman
among women
After among the noun must be plural, because among means 'in the middle of several'. And woman pluralises irregularly to women – one of the pairs (woman/women, man/men, child/children, foot/feet) that must be learned as a pair.
WhyTurkish keeps a noun singular after a quantifier – çok kadın is literally 'many woman' – because the quantifier already signals plurality and the noun need not repeat it. English disagrees and marks it twice.
False friends 1
I will have controlled all the wards
I will have checked all the wards
In English, control means to direct or restrain something; to inspect it is to check it.
WhyA false friend from Turkish kontrol, which covers the sense English gives to check – so the English cognate is reached for and lands on the wrong meaning.
have support 1
I haven't neighbours
I don't have neighbours (or, more colloquially: I haven't got any neighbours)
When have means possess, British English makes it negative with don't/doesn't, not by adding -n't to have: I don't have a car, she doesn't have time. I haven't got is the other correct option – but then you need got.
Negation 1
I don't worried
I'm not worried
Worried is an adjective, not a verb, so it is negated with am / is / are not, never with don't. The quick test: if you could put very in front of the word (very worried, very tired, very sure), it is an adjective, and adjectives are negated with BE.
WhyTurkish negates in two different ways depending on what is being negated: verbs take a suffix (gitmiyorum, 'I am-not-going'), but non-verbal predicates – adjectives and nouns – take the separate word değil: endişeli değilim, literally 'worried not-I-am'. English also splits two ways, along a slightly different line. The learner has the same two-branch system already and simply took the verbal branch for an adjectival predicate.
reason clauses 1
The contractions are every five minutes, before the patient come to hospital.
When the contractions are every five minutes, come to the hospital.
Three things at once: the connector names a condition, so when, not a tacked-on before; the patient come → comes; and hospital takes the.
Sentence boundaries 1
There were three options, however Monitoring was the safest.
There were three options. However, monitoring was the safest.
However is not a joining word. It needs a full stop in front of it and a comma after. Same family as therefore and as a result.
verb patterns 1
If the mother needed to support with breastfeeding, we would support her
If the mother needed support with breastfeeding, we would support her
She needs to be supported, not to support – the noun support is what is needed. The conditional frame around it was right.
These are the Turkish corrections only. The full archive holds every language side by side, filterable by grammar point and level: the marking archive. Other languages: Ukrainian, Polish, Russian, Spanish, Chinese, Korean.
A rule earns its place by being produced correctly and unprompted, not by being explained once.
The silent final -e
An English final -e is usually silent, and its real job is to reach back and stretch the vowel before it. See the -e, don't say the -e, and lengthen the vowel.
CautionThis is exactly the wrong instinct for a transparent orthography like Turkish, where a final -e is always pronounced – the reading habit has to be suspended, not sharpened.
The -ed syllable rule
The -ed ending adds a whole extra syllable only when the word already ends in a t or d sound. Everywhere else it flattens onto the end of the word.
CautionOn a clinical ward this covers most of the verbs you use, so the rule earns its keep in speech far more than in reading.
A lone letter says its name
An initialism is not read as a word: it breaks into separate letters, each said with its own English letter-name. The sting is in the letter I, which alone is EYE.
CautionSpell the letter-names long – BEE, JEE – not the short vowels a Turkish reader would give them.
The discharge stress shift
Discharge changes meaning with its stress: as a noun, the stress is on the first syllable (DIS-charge – what you assess and document); as a verb, it moves to the second (to dis-CHARGE a patient – to send them home). Same spelling, two stresses, two meanings.
CautionGetting this wrong is not cosmetic in a clinical setting – a mispronounced discharge at handover is a patient-safety matter, because the two stresses name two different things.
The -TION rule
Every English word ending in -tion or -sion stresses the syllable immediately before that ending, without exception. Ask one question of any unfamiliar long clinical word – does it end in -tion? – and you already know where the stress goes.
CautionAdding a further suffix does not move the stress: gestational is built from gestation and stays jes-TAY-shun-al. And the rule does not reach everything – contemporary (kon-TEM-pruh-ree) belongs to no pattern at all, which is why it is the least urgent word on the sheet.
The -NATAL family
The -natal words are one family with one stress, on NAY. Learning a new member is not learning a new word – it is the same word with a different prefix, and the stress does not move.
CautionThis family sits outside the -tion rule, so do not try to apply the -tion test to it – antenatal is an-tee-NAY-tal, not an-te-NA-tion-style.
The -ic / -ical stress rule
Words ending in -ic or -ical stress the syllable immediately before that ending – the same shape as the -tion rule, applied to a different suffix.
CautionEnglish stress is famously irregular, but the part that is rule-governed happens to include most long medical vocabulary, because those words came from Latin through French and carried their stress rules with them.
The hand test
A relative clause only describes a noun; it cannot stand alone as a sentence. Cover the who / which / that clause with your hand – what is left must still be a full sentence with its own verb. If it is not, the sentence is not finished.
CautionThis gets harder, not easier, as the clauses get fluent: once who and whose come automatically, the temptation is to stop when the description feels complete.
Say the object once
When that / which / who is the object of a relative clause, it already is the object. Do not name it again with it, him, her or them at the end of the clause.
CautionThe extra pronoun always feels like clarity and always reads as an error – the cure is to delete exactly one word, never to rewrite the clause.
Match the clocks – second or third?
Both the second and third conditional imagine something that is not the case; the only question is which clock. Second is an unreal now or future; third looks back at an unreal past. Ask each conditional one question: am I imagining now, or looking back?
CautionThe tell is in the second half: would + verb means present or future; would have + participle means past. Match the if-clause to it – were for now, had + participle for the past.
The clinical article
Clinical English is not fully regular, so the frequent ones are learned as fixed phrases. A process used as a function takes no article; a named, specific place still takes the; and uncountables take nothing at all.
CautionThe trap runs both ways: the same speaker who drops the before postnatal ward will add it before theatre. Decide per phrase, not per habit.
Silent final -e: bride = BRYD, not bri-de. The -e stretches the vowel before it.
-ed adds a syllable ONLY after t or d: admitted yes, raised no.
To possess: don't / doesn't have. I don't have neighbours – or I haven't GOT any.
DIS-charge the noun. Dis-CHARGE the verb. Stress carries the meaning.
Came TO. Move TO. Go TO. Motion always takes TO.
No WOULD in the if-clause: if I HAD more staff, I would feel calmer.
Watch her observations. Listen TO the heartbeat. The TO flips by verb.
Before -tion, the stress. Dy-LAY-shun. Medi-CA-tion. Every time.
Post-NAY-tal. Nee-oh-NAY-tal. An-tee-NAY-tal. One family, one stress.
I am not worried. Adjectives take BE, never DON'T.
I am worried. The trend is worrying. People -ed, things -ing.
Safer, safest. Short adjectives never take MORE or MOST.
A rule scales. A word list does not.
A report that only lists faults teaches you nothing about what to keep.
The departments that the team covers at night are often very busy.
CreditedA defining relative clause built correctly and, crucially, finished: the main clause has its own verb (are often very busy). This is the exact structure that was left hanging elsewhere in the same piece of writing, so it shows the target is available and not yet automatic.
Then, the second ward, which is called the infant ward, is where nurses work.
CreditedA non-defining relative clause with both commas correctly placed, embedded mid-sentence – the harder of the two relative types, and produced unprompted in free writing rather than in a controlled exercise.
If we had not acted quickly, the patient's condition could have worsened.
CreditedA clean third conditional produced unprompted in free writing, with the negative in the if-clause and a modal perfect in the result – and it stands in a piece where the second/third distinction was the identified fault line, so it shows the structure is there when the writing is unhurried.
If I had encouraged her sooner, she would feel more confident by now.
CreditedThe mixed conditional, exactly right: an unreal past in the if-clause and a present consequence in the result. It is the hardest of the five types and it was produced correctly in a controlled exercise where the two simpler types were also in play.
This is the marking talking to itself. It is here because it is the honest evidence that the rest of the page came from somewhere.
Defining and non-defining relative clauses were built across the whole lesson – who, which, that, whose, where – and not one came out wrong. Eight errors were logged in the session; none was a relative clause.
The report calls this "inferred from absence, and it's a meaningful absence", and then argues why the absence carries weight rather than being merely uneventful. Turkish builds relative clauses back to front from English, placing the whole clause before the noun as a participle with no linking word at all. To produce an English relative clause the learner must reverse the word order and insert a word their first language does not possess. A whole lesson of that, error-free, is graded absorbed and not merely 'not tested'.
No conditional errors appear anywhere in the session. Two grammar strings were logged – "when I came the hospital" and "I would move London" – and both are missing prepositions, charged under a separate count. The conditional frame inside the second of them (I would move) is perfectly built.
The interesting move is what the report refuses to count against the verdict. Two faulty sentences occurred inside conditional contexts, and it would have been easy to let them contaminate the grade; instead the report states explicitly what they are not – they are preposition drops, not faults in the conditional system – and charges them elsewhere. The verdict is also supported from the typology: Turkish builds conditionals cleanly with the suffix -sA (gelir-se-m, 'if I come') plus a clear result clause, so the paradigm is predicted to transfer well, and today it transferred without a mark.
Not one comparison error across the session, and specifically none of the error that was predicted.
This absence-inferred verdict is argued from a prediction rather than from silence. Turkish builds comparison the way English builds half of it – daha plus adjective for the comparative, en plus adjective for the superlative – so the analytic pattern with more and most transfers straight across, and the predicted Turkish-speaker error is the mirror image: over-using more and most where English switches to -er and -est. None appeared. The absence of the specific predicted error, rather than the absence of errors in general, is what licenses the verdict, and the Error Hunt was written to test the switch deliberately.
Knowing the pattern your first language produces is most of the work. Knowing which of them you are still making takes someone reading your writing.
Every lesson comes with a written report: the marking line by line, the criteria, and one priority for next time. A worked example shows exactly what comes back. Lessons are £20 for fifty minutes, one to one, in proper British English; the first step is a discounted trial lesson.