See below an example of an Analyticbridge email campaign that was monitored over a period of about **600 days**. It clearly shows that 20% of all clicks originate after day #5. Yet most advertisers and publishers ignore clicks occurring after day #3. Not only 20% of all clicks occurred after day #3, but **the best clicks** (in terms of conversions) occurred several weeks after the email blast. Also, note an organic spike occurring on day #23 in the chart below - this could be due to our vendor (iContact) featuring the newsletter in question on their website, without charging us an extra fee.

This brings interesting points:

- If you advertise a webinar taking place in 5 days, also add in your email blast a short note about webinars happening in 20, 40 or 80 days.
- If you don't renew a campaign because ROI is < 0 after 3 days, you are going to lose the "long tail" distribution that is possibly making your ROI > 0, and you might erroneously kill a profitable campaign.
- You can't just look at days #1 to #3 to estimate the lifetime value of an email blast. You are dealing with rigt-censored data, and you have to use statistical attribution and survival analysis models to measure the true impact.
- Even if you monitor clicks over a 60-day time period, you'll still miss 5% of clicks, and much more than 5% in revenue.

Thus, there's a systemic methodology flaw and bias when measuring half-life of your campaign (unless you use ad-hoc statistical methodology): the data is right-censored. How can you be sure that 97% of all clicks occur in the first 3 days? Maybe as many clicks will arrive between day 11 and day 300. But since your window of observation is only 3 days (or at best 14 days), you just can't answer the question. You can compute good estimates for half-life though, if you use a statistical model based on (say) exponential distributions, together with statistical inference, and integrate the fact that the data is right-censored, in your statistical model.

Below is a chart showing that even 60 days worth of historical data covers only 95% of your campaign in terms of clicks - and indeed much less in terms of revenue:

Here's an example where the data and conclusions are biased and wrong due to ignorance of the "right censorship" principle that applies to time series: http://blog.bitly.com/post/9887686919/you-just-shared-a-link-how-lo...

**Conclusion**: You don't need to track your email campaign for 600 days to measure ROI, you can monitor your campaign for 28 days, and then make interpolation using statistical models. Of course, if you started your campaign just before a great event (like Christmas shopping), then you need to take into account seasonality. That's where a statistician can help you. The above chart represents a campaign generating about 6,000 clicks.

© 2020 Data Science Central ® Powered by

Badges | Report an Issue | Privacy Policy | Terms of Service

**Most Popular Content on DSC**

To not miss this type of content in the future, subscribe to our newsletter.

- Book: Statistics -- New Foundations, Toolbox, and Machine Learning Recipes
- Book: Classification and Regression In a Weekend - With Python
- Book: Applied Stochastic Processes
- Long-range Correlations in Time Series: Modeling, Testing, Case Study
- How to Automatically Determine the Number of Clusters in your Data
- New Machine Learning Cheat Sheet | Old one
- Confidence Intervals Without Pain - With Resampling
- Advanced Machine Learning with Basic Excel
- New Perspectives on Statistical Distributions and Deep Learning
- Fascinating New Results in the Theory of Randomness
- Fast Combinatorial Feature Selection

**Other popular resources**

- Comprehensive Repository of Data Science and ML Resources
- Statistical Concepts Explained in Simple English
- Machine Learning Concepts Explained in One Picture
- 100 Data Science Interview Questions and Answers
- Cheat Sheets | Curated Articles | Search | Jobs | Courses
- Post a Blog | Forum Questions | Books | Salaries | News

**Archives:** 2008-2014 |
2015-2016 |
2017-2019 |
Book 1 |
Book 2 |
More

**Most popular articles**

- Free Book and Resources for DSC Members
- New Perspectives on Statistical Distributions and Deep Learning
- Time series, Growth Modeling and Data Science Wizardy
- Statistical Concepts Explained in Simple English
- Machine Learning Concepts Explained in One Picture
- Comprehensive Repository of Data Science and ML Resources
- Advanced Machine Learning with Basic Excel
- Difference between ML, Data Science, AI, Deep Learning, and Statistics
- Selected Business Analytics, Data Science and ML articles
- How to Automatically Determine the Number of Clusters in your Data
- Fascinating New Results in the Theory of Randomness
- Hire a Data Scientist | Search DSC | Find a Job
- Post a Blog | Forum Questions

## You need to be a member of Data Science Central to add comments!

Join Data Science Central