What does Day and Week Parting mean and should you be using it? (1)
What does Day and Week Parting mean and should you be using it?
One of the approaches that you can take in Affiliate Marketing, is hunting for pockets of profit. So instead of going broad with your campaigns, you would be cherry picking just what works the best for you. This could mean targeting just certain segments, placements, regions, devices … or focusing just on the days, or even hours, that show the best performance with your campaigns.
Today, I would like to write about the last two options I mentioned: targeting specific days or hours. It's called Day-Parting (when targeting certain hours of the day) and Week-Parting (when targeting certain days of the week).
WHY DOES IT EVEN WORK, OR DOES IT AT ALL?
Both dayparting and weekparting works and there are valid reasons for this. Every person goes through various stages during the week or a day … most work from Monday to Friday, most wake-up in the morning and go to bed before midnight, most respond to seasonal trends … I could go on, but I'm sure you already know where I'm going with this.
The habits of people, that we are targeting with our ads, determine what they are most likely to respond to, under what conditions and during what part of the day or week. Sometimes we can identify these patterns easily on our own, by using simple logic, sometimes we have to rely on the data to show us the patterns.
Let's try to come up with an example now, what could be a pattern or habit that we could identify without even looking at some data at all... Ever heard the word PRIME TIME? You might find several definitions for it, but it will always be related to the evening hours, around 7-10 p.m. … At this time, majority of people is already done with all work, everyone is taking some rest, watching TV, browsing the net … these few hours are the most trafficked hours during the day. So if you were to run a Branding campaigns, these are your hours to target for maximum reach.
Let's do another one, shall we? Weekends, past midnight ... those who have partners are already having a good time, the rest is browsing the net, mildly drunk, looking for some fun … bang, these are the perfect candidates for dating offers. And the stats confirms this, these hours are typically the best converting for dating offers, especially adult dating offers 
There are patterns that are repeating every day or every week, like the examples I mentioned … then there are those that are not so frequent – let's call them seasonal. Diet supplements always start selling better few months before the summer season, skiing gear sells better in fall and winter… everything sells where around Christmas. Targeting seasonal or trending offers is also a vital approach, but that's a topic for another article.
Most often than not, the patterns are not so easily identifiable with simple logic and you need a lot of data to spot them. The only way to go about this, is to run consistently for some time and then analyze the data. I will get back to this a bit later on.
SHOULD WE USE SIMPLE LOGIC OR RELY ON DATA?
Thought it might sound tempting to target only certain days or hours straight away, you will have to rely on the data in most cases. Affiliate Marketing is a numbers game, the feelings and assumptions have to go away. But that doesn't mean you could never use the logical approach.
For example the dating example I used, it's a fact that it works best during the weekend. So if you are just starting out, you can focus on weekends in order to reach profits faster. Plenty of “casual” stuff works the best on weekends actually, I had the best results with game apps on weekends, for example. On the other hand, utility apps worked better during the week for me.
The patterns are there, you just need to identify them by testing. If you see that some offers are performing better during certain hours and days, chances are that similar offers will work the same way. This means that you could re-use this info and focus on the same time segments with new offer types that are related to the older ones.
DAY-PARTING OR WEEK-PARTING IS NOT YOUR GOAL, IT'S A SECONDARY OPTION.
We need to set one thing straight here, using day or week parting is not your primary goal, consider it something like a backup solution. Ideally, you want to run 24/7, which gives you the most volume and possible the highest revenue and profit.
There are basically two main reasons for using this technique:
1. To save a losing campaign and make at least some profit with it – by targeting just the best days our hours.
2. To save $$$ on testing – by targeting a logically selected or data-proven time segments, you decrease the amount of $ needed to reach statistical significance.
NOTE: Wrongly chosen time segments will completely screw up your testing, make sure you chose them based on reliable data or by following VERY strong patterns that have been proven again and again in the past.
HOW MUCH DATA DO YOU NEED TO IDENTIFY A PROFITABLE TIME SEGMENT WITH CERTAINTY?
I've seen new affiliates making the same mistake over and over, when selecting time segments to target. They are coming to conclusions too early.
For example : A guy was running some campaign for a week, he was profitable on Tuesday and all the other days lost 20%-30%. He came to me saying that the campaign is a failure and he will only be running it on Tuesdays. The next Tuesday, he lost 20% and abandoned the campaign completely.
This was obviously a wrong decision. First of all -20% is usually optimize-able into profit, and on top of that 1 profitable day doesn't really prove anything.
I personally wait AT LEAST for 3 repetitive signals in order to even consider targeting just a specific time segment. So when talking about dayparting, I want to see the same hours perform better than the rest, 3 days in a row… or 5 times in a week. In case of weekparting, I would want to run the campaing for 2-3 weeks minimum to see whether the same day(s) performs the best, week after week.
By doing it sooner, I'd be risking that my decisions were based on pure coincidence… and that's not something we should be doing in AM. Keep in mind that ONE occurrence of anything doesn't prove shit
NOTE: Keep in mind that by changing significant parts of your funnel, you effectively invalidate the past results and they are not usable for day week parting decisions anymore. Small changes are ok, but if you completely change the logic of the funnel, you need to ignore the past data.
TO DAY-PART OR NOT TO DAY-PART?
As I already mentioned, day or week parting is one of the approaches that can help you reach profits. However, these should be something like backup options and not the primary goals of your efforts.
They can help to save a failing campaign, or cut down the testing costs in case of new offers though, so it's always an option to consider.
Just make sure that you don't jump to conclusions too early … just because something happened once, there are no guarantees for it to happen again.
Thanks for reading!
Home > General > Affiliate Marketing Forum