What is time-decay attribution?
Time-decay attribution is a multi-touch attribution model that gives some credit to all the channels that led to your customer converting, with that amount of credit being less (decaying) the further back in time the channel was interacted with.
The assumption here is that the first advertising channel your customer interacted with merely planted the seed, and the customer’s interest in committing to a purchase grew over time with repeated exposure to various marketing channels. As such, the way that the time-decay attribution model assigns credit to your different channels can be interpreted as a rising level of interest and commitment from the customer.
Application of the time-decay attribution model
1. First, a friend mentions an old movie, and I think “I wonder if a t-shirt exists for that?” – this leads me to an organic search for the item on my phone during a lull in conversation. I see a few different options. I click one, but when my friend wants my attention again, this passive activity ends as I close the window and return to being present. I would not remember my fleeting desire for this shirt.. unless I was somehow reminded..
2. Two nights later, I’m scrolling through my Instagram feed before bed. (We’ve all seen the studies on electronics usage affecting sleep. That doesn’t mean people don’t do it.) Between pictures of my friends’ pets and what they had for dinner, an ad appears for the site I went to earlier. I go to it again. I wisely decided that I shouldn’t make a purchase while half-asleep, but the item is more firmly lodged in my brain at this point.
3. At work the next day, I scroll through Facebook after a string of meetings to give my brain a break. Surprise! There’s another ad for the site. I go back to show my co-workers the shirt for the social approval I need before making a $20 purchase. They co-sign on it, but as a responsible employee, I then get back to writing articles like this one.
4. The following evening, I’ve decided to pull the trigger on my new t-shirt that will proudly display my knowledge of obscure film. By this point, I remember the site’s name, which would have been in doubt after my first (or even second) encounter with it – thanks, advertising! I make a direct visit to the site while fully lucid, I complete the transaction.
I became a conversion, and my conversion path was:
Organic Search (4 days ago) -> Instagram (2 days ago) -> Facebook (1 day ago) -> Direct visit
Using the time-decay attribution model, this effectively reads as the journey from curiosity to taking action. For any math-inclined readers, the basic formula involved is y = 2(-x/7), with the results then converted proportionately to fit into a neat 100%. Accordingly, the conversion credit break down as:
Organic Search: 19.8% Instagram: 24.1% Facebook: 26.6% Direct: 29.4%
Of note, this illustrates how important the time aspect of this model is: the above example plays out over four days. Many purchases beyond a $20 t-shirt take much more time to consider, so the difference in conversion credit percentages above becomes much more drastic the longer the customer journey is.
Since this t-shirt cost the company $4.50 to make, the profit is $15.50 for the sale. The time-decay model gives Facebook ads spend credit for 26.6% of this profit:
$15.50 \* 26.6% = $4.12
And, Instagram gets 24.1% of the profit:
$15.50 \* 24.1% = $3.74
Using a hypothetical cost of $1.70 per click for Instagram and $1.05 per click on Facebook, we can see that in this case the lower value spent on the Facebook click drove more value. This helps make the decision to maintain or raise the Facebook Ad spend.
To be more accurate, the example would include other business costs, lifetime value of customers and be applied across many purchases and customer touchpoints to measure overall profitability.
Pros and cons of using a time-decay attribution model
- Sharing is Caring. Gives some credit to all touchpoints
- Always Be Closing. Touchpoints closest to the conversion are valued the most, which gives preference to marketing channels that tend to do more of the “closing” work
- Consistency is Key. Applying a standardized formula to all of your campaigns will easily highlight fluctuations in the activity of individual channels
- Didn’t See You Back There. Gives less credit to the first touchpoint, which might be the most difficult to execute
- One Size Doesn’t Fit All. Customer journeys are not always a straight line, so the path they took may not be best represented by lowest to highest value interactions
- Hard Numbers. The math involved, while consistent, could be a little complex
Time-decay attribution is especially helpful to measure longer sales cycles, as time between channel interactions will really serve to highlight the difference in conversion credit they receive. Timed campaigns also do well under this model since the time measurement aspect is what this model is based around. Consistency is also a major benefit, since the y = 2(-x/7) formula is standardized, so fluctuations in activity are easily measurable. By using a set formula that uses the number of days prior to conversion as a key variable, every marketing channel receives credit based on the assumption that the more time that has passed since the first interaction, the closer the customer got to the actual sale.