Startup pitch deck

Sodura Speak for IELTS

Mongolian IELTS Speaking practice with AI mock tests, local-language feedback, and a price students can actually try before exam day.

Sodura Speak

IELTS Speaking mock result

Mongolian feedback

Mock band

6.5

Good enough to see progress. Specific enough to know what to fix.

Fluency6.5
Vocabulary6.5
Grammar6.0
Pronunciation7.0
Next drill: answer Part 2 with clearer examples and fewer repeated linking phrases.

01 / Learner pain

Speaking is the hardest skill to practise alone.

IELTS learners can find vocabulary lists and sample answers online, but they cannot easily recreate the pressure of speaking out loud, being timed, and receiving useful feedback in Mongolian.

The official IELTS Speaking test is an 11-14 minute interview with three parts and four assessment criteria.2 That structure rewards repeated, realistic practice.

11-14m

Official IELTS Speaking test length.2

4

Speaking criteria: fluency, vocabulary, grammar, pronunciation.2

898k

MNT listed for IELTS Academic / General Training at the NUM test centre.6

02 / Market signal

A global exam now has a stronger local access story.

12,500+

Organisations accept IELTS across more than 140 countries.4

4M+

IELTS tests were taken globally in 2023.5

NUM

A British Council IELTS Centre opened at the National University of Mongolia in 2025.6

1-5d

The Mongolia centre says digital test results are issued within 1 to 5 days.6

03 / Product

Sodura Speak turns IELTS anxiety into a daily speaking loop.

01

Start a mock test

Part 1, Part 2, and Part 3 in a familiar IELTS Speaking flow.

02

Record answers

The learner speaks instead of only reading tips or memorising templates.

03

Get a mock band

AI estimates an overall score and criterion-level feedback.

04

Read Mongolian coaching

Mistakes, stronger phrases, and next-practice goals are explained clearly.

Live product insight

The current site already sells the right promise: AI-аар, монголоор бэлд.1

04 / Why now

The wedge is not content. The wedge is feedback speed.

Speaking-first

The learner must produce speech, not just consume tips.

Mongolian explanations

Feedback lands faster when the learner understands the why in their native language.

Exam-like timing

Part 1, Part 2, and Part 3 practice builds automaticity under pressure.

Visible progress

Mock band history, criteria trends, and streaks make practice feel measurable.

Replay and compare

Learners can hear their own answers and track stronger versions over time.

Next drill

Every result should recommend one concrete thing to practise next.

05 / Pricing

Affordable compared with the cost of one real exam.

Free level check

0 MNT

1 full IELTS Speaking mock test with Mongolian feedback

Starter pack

29,900 MNT

5 full mocks for learners preparing for a near-term exam

Monthly practice

49,900 MNT

10 full mocks plus free speaking practice for 1 month

Classroom / cohort

Custom

Teacher dashboard, student progress, bulk codes, and reports

Price logic

A 49,900 MNT monthly pack is roughly a low-risk rehearsal purchase before an official IELTS test fee near 900k MNT.

06 / System design

A focused speaking pipeline before a full IELTS platform.

Learner app

Prompt flow
Voice recording
Timer
Progress history

Speech pipeline

Audio capture
Transcription
Pronunciation signals
Quality checks

IELTS coach

Rubric mapping
Band estimate
Mongolian explanation
Next drills

Growth layer

Payment
Streaks
Referral codes
Teacher exports
Audio mock test
AI rubric engine
Mongolian coach report

Quality: sample checks, transcript confidence, rubric constraints, and escalation for unclear audio.

Payments: QPay, cards, class coupons, monthly bundles, and trial-to-paid conversion.

Expansion: Writing feedback, teacher dashboard, and partner school cohorts after Speaking retention.

07 / Competitive map

Mongolian learners choose between local test-centre access and imported practice apps.

Official IELTS test-centre pathway

The official IELTS ecosystem already supports test-centre discovery for Mongolia.12

Where they win

Local learners can already access the official exam pathway without leaving the country.

Gap we can use

Booking a real test is not the same as getting cheap, frequent, low-pressure speaking practice in Mongolian.

NUM IELTS Centre

A newly opened official IELTS centre at the National University of Mongolia.6

Where they win

Strong local legitimacy and makes IELTS feel more accessible inside Mongolia.

Gap we can use

A test-centre presence validates demand, but it does not solve day-to-day speaking practice or instant feedback.

SmallTalk2Me

Imported AI speaking product with an IELTS Speaking simulator and instant feedback.7

Where they win

Strong speaking-first UX and explicit IELTS adjacency for self-study users.

Gap we can use

Broad global English product. It is not clearly built around Mongolian learner context or localized coaching.

Talkface

Imported mobile-first AI English app with dedicated IELTS Speaking and Writing sections.8

Where they win

Clear IELTS positioning and app-style content catalog for quick consumer adoption.

Gap we can use

Looks English-first and generic. There is little evidence of native-language explanation or local exam-prep nuance.

Stimuler

Imported AI speaking coach with IELTS prep entry points.9

Where they win

Good speech-practice loop and strong general speaking-product credibility.

Gap we can use

More general spoken-English coaching than a tightly focused IELTS Speaking score product for Mongolian learners.

Positioning takeaway

The gap is clear: a Mongolia-first IELTS Speaking product that sits between official exam access and imported AI apps, with low-cost repetition, native-language coaching, and instant mock-band feedback.

08 / Trust

A mock score must be useful without pretending to be official.

Clear disclaimer

Sodura gives practice feedback and mock band estimates, not official IELTS results.

Official criteria map

Reports should always map to fluency, lexical resource, grammar, and pronunciation.

Format fidelity

Part lengths, task roles, and scoring language stay close to official public guidance.

AI transparency

Explain when scoring is AI-assisted and when pronunciation confidence is limited.

Evidence-based feedback

Tie every suggestion to transcript evidence, audio evidence, or timing behaviour.

Teacher calibration

Use human review samples to tune score distributions and learner trust.

09 / GTM

Sell confidence first, then study habit.

01

TikTok / Reels

Short before-after speaking clips and common Mongolian learner mistakes.

02

Facebook groups

IELTS Mongolia groups, study-abroad communities, and parent audiences.

03

University partners

NUM IELTS Centre context makes IELTS feel more locally accessible.

04

Teacher affiliates

Give English teachers cohort codes and revenue share.

05

Exam countdown

7-day and 30-day speaking challenge funnels.

Validation target

1,000 free level checks, 100 paid customers, and 35%+ week-two return rate before expanding beyond Speaking.

10 / Roadmap

Own Speaking first. Add Writing when the loop is proven.

Now

Polish free level check, purchase flow, result report, and shareable progress card.

30 days

Daily challenge, reminders, referral codes, and teacher affiliate onboarding.

60 days

Cohort dashboard, score trend analytics, and calibration review set.

Next product

IELTS Writing Task 1 and Task 2 feedback once Speaking retention is proven.

11 / Decision rule

Build only where learners keep coming back to speak.

Paid practice is the proof. Retained practice is the moat.

The next decision should be based on purchase conversion, repeat mock-test usage, and whether learners trust the feedback enough to share it with friends or teachers.

Measure retention

Free check to second mock, week-two return, streak starts, and churn.

Measure payment

Starter vs monthly pack conversion, QPay completion, and refunds.

Measure trust

Feedback usefulness rating, teacher review, and score calibration.