Video summary

ПРОДУКТ в IT. Как рассчитать основные метрики? Просто о сложных формулах

Main summary

Key takeaways

Educational

Main ideas / lessons

  • The video explains what an IT product is and then focuses on core product metrics (“indicators”) used to evaluate how consumer software products perform over time.
  • It emphasizes that metrics help you understand:
    • Audience growth and engagement (e.g., how many users come back)
    • Monetization quality (e.g., conversion, ARPU/ARPPU)
    • Stability and retention (e.g., churn)
    • Financial return (e.g., ROI/ROAS-type metric called “payback”/“ROMI” in subtitles, and LTV)
  • It warns that a “good” metric value depends on strategy and economics: high conversion isn’t automatically good, and low conversion isn’t automatically bad.

The key theme: “Good” depends on the context of pricing, costs, and product strategy—not on a universal threshold.

IT product definition (as described)

An IT product can be:

  • Software
  • Web service
  • Application
  • Hardware

Its purpose is to provide functions and capabilities to users.

Scope of products and what is excluded

  • The focus is on end-consumer software products, such as:
    • Products in the style of VK/YouTube/Instagram
    • App ecosystems like iPhone/Mac
  • The video notes that hardware vs. software metric “laws” differ, so it focuses on software products only.
  • It avoids detailed comparisons of metrics for large companies.

Key metrics and how they’re calculated

1) Daily / Monthly Active Users (DAU / MAU)

  • Meaning: number of users who visit the application/website/service
    • DAU: once per day
    • MAU: once per month
  • What it shows:
    • Audience dynamics (growth or decline)
    • Seasonality (e.g., fitness apps peaking in January)
  • Note from the video:
    • Seasonality can be independently checked using Google Trends (search query statistics).

2) Conversion (purchase conversion)

  • Meaning: ratio between:
    • users who performed the target action (usually purchase)
    • and users who entered/visited the product
  • How it’s calculated (purchase conversion):

[ \text{Conversion} = \frac{\text{number of purchasers}}{\text{number of visitors/users who entered product}} ]

  • Example given:
    • 100 people downloaded the game; 2 bought
    • Conversion = 2 / 100 = 2%
  • Key lesson:
    • “High” vs “low” conversion depends on pricing, average sales, advertising spend, and planning—it’s not inherently good or bad.

3) Retention (return rate)

  • Meaning: shows how many users come back after N time periods (examples use days).
  • How it’s calculated (Y-day retention concept):
    • Take the number of users who visited again on day Y
    • Divide by the number of users who were present Y days earlier

[ \text{Retention (day Y)} = \frac{\text{users who returned on day Y}}{\text{users from day 0}} ]

  • Example given:
    • Day 0: 1000 people download
    • After 7 days: 300 of those visited again
    • Retention (day 7) = 300 / 1000 = 30%
  • Why it matters:
    • Indicates how users live with the product over time
    • Helps predict other metrics like MAU and engagement indicators
  • Value statement:
    • Good retention products are easier to monetize and scale.

4) Churn / Churn rate

  • Meaning: percentage of audience you lose over a period (typically monthly), often referring to paying users/subscribers.
  • How it’s calculated:

[ \text{Churn rate} = \frac{\text{number of clients lost in period}}{\text{number of clients at start (or during period)}} ]

  • Why it matters:
    • Growth in absolute numbers can hide problems if churn is high
    • Helps detect “weak spots”
  • Key lesson:
    • Even while growing, high churn is the enemy of stable, successful growth.

5) ROMI / ROI-like metric (payback on acquisition spend)

  • Meaning: how much money spent to acquire users returns as profit.
  • Important definition detail:
    • The video says this metric typically does not include fixed costs (e.g., salaries, rent).
    • It focuses on cost of attracting users.
  • How it’s calculated (as described):

[ \text{ROMI} = \frac{\text{revenue from attracted audience} - \text{attraction costs}}{\text{attraction costs}} ]

  • Example given:
    • Spend: 1000 rubles on ads
    • Attracted: 100 users
    • Revenue: 1200 rubles
    • ROMI = (1200 − 1000) / 1000 = 0.20 = 20%
  • Why it matters / implied interpretation:
    • Shows whether the product pays for itself via acquisition
    • If the metric were > 0, endlessly increasing ad spend could (in theory) keep generating profit, but in reality returns often drop and the metric becomes volatile
  • Use:
    • Financial planning during product development and evaluation.

6) ARPU and ARPPU

ARPU (Average Revenue Per User)

  • Meaning: average revenue per all users who came
  • How to calculate:

[ \text{ARPU} = \frac{\text{total revenue in a period}}{\text{total number of users who came}} ]

ARPPU (Average Revenue Per Paying User)

  • Meaning: average revenue per paying user
  • How to calculate:

[ \text{ARPPU} = \frac{\text{total revenue in a period}}{\text{number of paying users}} ]

  • Example logic from subtitles:
    • If 1000 users come and revenue is $1000:
      • ARPU = $1000 / 1000 = $1
    • If only 100 paid:
      • ARPPU = $1000 / 100 = $10
  • Why they matter:
    • Predict financial inflow
    • If you know ARPU and expected user volume, you can estimate revenue.

7) Upsell / development of paying audience (strategy example)

  • Meaning (as framed):
    • Use product/commercial mechanics to increase revenue per paying user (i.e., ARPPU)
  • Example given:
    • McDonald’s suggests adding sauce; customers pay extra → increases average revenue per paying consumer.

Presented as a strategy rather than a named metric formula in the subtitles.

8) LTV (Lifetime Value)

  • Meaning: how much each user will ultimately spend in the product over their lifetime.
  • Core idea:
    • It should reflect total expected spend per user (or per user cohort) across time.
  • Challenge noted:
    • Often it’s hard to know exactly when the user’s lifecycle ends, so LTV usually needs approximation.

  • Simplest approximation method described:

    • Average sales per user
    • × average bill
    • × number of periods the user will live (estimated)

  • Why it matters:

    • It supports financial planning and ties into profitability and investment return (as claimed by the video: LTV influences ROI, profit, and product success).
  • Strategy advice:
    • Use better approximations and refine methods to improve planning accuracy.

Overall closing guidance from the video

The speaker says you can design any set of indicators, but what matters is:

  • defining how to relate them to monitor product success
  • using them for financial planning and development
  • getting closer to accurately understanding the true user value / lifecycle (emphasized as very important in the subtitles)

Speakers / sources featured

  • Andrey — host/speaker (creator of the channel discussing IT and startups)
  • Examples / brands mentioned as product examples (not speakers):
    • VKontakte
    • YouTube
    • Instagram
    • Mac / Macintosh
    • iPhone
    • Skype
    • Zoom
    • Candy Crush
    • PlayStation
    • McDonald’s
  • Google Trends — referenced as a tool/source for seasonality via search statistics

Original video