ASd

*After we noticed second screen effect we started building TV measurement tool*

• one of challenges that forced us to make TV tool was absence of any measurement of TV in Kyrgyzstan
•  what we can measure now

each channel

each program

cost per user

cost per listing

cost per contact 

LTV life time value

Baseline

creatives effectiveness

all factors of media plan that can affect on ROI like day time or position in block

We also build a prediction system of traffic from TV which is based on analysis of previous campaign

So when we launch new campaign we can compare actual results with a prognosis and make a conclusion whether we stop campaign or continue

we can make such decisions after third day of a campaign

Prediction helps us to build an architecture of campaign that includes weekly weights, length of campaign and TV channel split

We look at four main indicators that show us the performance of the campaign: new listers, new users, new sessions and new listings.

Tools that we use on this stage: Econometric modelling, Reach and Frequency

We came to some interesting conclusions with TV like

• TV pressure should be higher than average noise
• TV offers lower CAS (customer acquisition cost) at high volume
• Entertainment content performs better
• Celebrity works
• Product placement in most popular programs should be for traffic goals
• Competition doesn’t affect conversions
• Cheaper paid online marketing
• Retention boost
• 30% top of mind in 1st year (AZ and KG)
• Market creation

*After we noticed second screen effect we started building TV measurement tool*

• one of challenges that forced us to make TV tool was absence of any measurement of TV in Kyrgyzstan
•  what we can measure now

each channel

each program

cost per user

cost per listing

cost per contact 

LTV life time value

Baseline

creatives effectiveness

all factors of media plan that can affect on ROI like day time or position in block

We also build a prediction system of traffic from TV which is based on analysis of previous campaign

So when we launch new campaign we can compare actual results with a prognosis and make a conclusion whether we stop campaign or continue

we can make such decisions after third day of a campaign

Prediction helps us to build an architecture of campaign that includes weekly weights, length of campaign and TV channel split

We look at four main indicators that show us the performance of the campaign: new listers, new users, new sessions and new listings.

Tools that we use on this stage: Econometric modelling, Reach and Frequency

We came to some interesting conclusions with TV like

• TV pressure should be higher than average noise
• TV offers lower CAS (customer acquisition cost) at high volume
• Entertainment content performs better
• Celebrity works
• Product placement in most popular programs should be for traffic goals
• Competition doesn’t affect conversions
• Cheaper paid online marketing
• Retention boost
• 30% top of mind in 1st year (AZ and KG)
• Market creation

*After we noticed second screen effect we started building TV measurement tool*

• one of challenges that forced us to make TV tool was absence of any measurement of TV in Kyrgyzstan
•  what we can measure now

each channel

each program

cost per user

cost per listing

cost per contact 

LTV life time value

Baseline

creatives effectiveness

all factors of media plan that can affect on ROI like day time or position in block

We also build a prediction system of traffic from TV which is based on analysis of previous campaign

So when we launch new campaign we can compare actual results with a prognosis and make a conclusion whether we stop campaign or continue

we can make such decisions after third day of a campaign

Prediction helps us to build an architecture of campaign that includes weekly weights, length of campaign and TV channel split

We look at four main indicators that show us the performance of the campaign: new listers, new users, new sessions and new listings.

Tools that we use on this stage: Econometric modelling, Reach and Frequency

We came to some interesting conclusions with TV like

• TV pressure should be higher than average noise
• TV offers lower CAS (customer acquisition cost) at high volume
• Entertainment content performs better
• Celebrity works
• Product placement in most popular programs should be for traffic goals
• Competition doesn’t affect conversions
• Cheaper paid online marketing
• Retention boost
• 30% top of mind in 1st year (AZ and KG)
• Market creation