It looks like JavaScript is disabled, so you are seeing this message.
rotate-phone
Rotate
your phone

Increasing installs while reducing CPI through AI optimization

O’STIN
case-7-image-maincase-7-image-main

Objective:

Improving acquisition efficiency by accelerating creative testing and optimization

Tools:

Programmatic

Challenges

  • High competition and customer acquisition cost in the category

  • Unstable conversion to purchase

  • Fatigue from standard creative approaches

  • High cost of testing a large number of hypotheses

Objectives

  • Increase the number of app newcomers with a focus on first purchases

  • Increase the number of tested hypotheses during the period

  • Reduce CPI during hypothesis testing by 30%

  • Increase overall purchase CR by 15%

Solution

Stage 0
Integrating AI into the manager’s routine
AI role
  • Analysis of the O’STIN app

  • Competitor analysis in spy tools: creatives and media splits

  • Analysis of the agency’s accumulated statistics: creatives and performance metrics

  • Analysis of available source capabilities

  • Generating hypotheses and recommendations

Manager role
  • High competition and customer acquisition cost in the category

  • Unstable conversion to purchase

  • Fatigue from standard creative approaches

  • High cost of testing a large number of hypotheses

Stage 1
Preparatory AI analytics and creative generation
  • Manager: writes an AI brief for category analysis

  • AI agents: run analytics and propose hypothesis options

  • Designer: prepares creatives based on recommendations, using AI to optimize the workflow

  • Manager: selects the most relevant options and passes creatives to launch

Stage 2
Campaign launch
  • AI agents: recommend the “source - creative - targeting” chain

  • Manager: adds new campaigns to rotation and sets KPIs

  • AI agent: analyzes in real time which hypotheses significantly deviate from the target KPIs and sends alerts to a Telegram chat

Stage 3
Manual control and adjustments
  • The manager receives a system report: “Creative A shows 40% lower CPI for the 25-34 audience in Yandex. There is +20% scaling potential.”

  • The manager analyzes the result: audience, settings and scaling potential

  • The manager makes a decision: scale and launch additional hypotheses, or stop

  • The manager feeds into AI: the accumulated knowledge base that trains the AI

Why it works
  • AI is an extremely powerful and fast analyst and executor

  • The manager stays in the loop for strategic decisions

  • The hybrid “AI + human” model creates a synergistic effect greater than the sum of its parts

Results

+35%

Mobile app installs

-32%

CPI

+17%

CR

Increased the speed of creative production and hypothesis testing

Cases

Request a proposal
Request a proposal
Request a proposal
Request a proposal
Request a proposal
Request a proposal
Request a proposal
Request a proposal
Request a proposal

Let us talk

We are always open to new opportunities. Want to launch effective advertising or discuss a partnership? Fill out the form and we will get back to you to discuss the details.
By clicking “Send request”, you agree
to the Personal Data Processing Policy
Thank you, your request has been received
Thank you, your request has been received
Thank you, your request has been received
Thank you, your request has been received
Thank you, your request has been received
Thank you, your request has been received
We will contact you shortly
back