Data collection is no longer a “nice to have” capability. In a modern world filled with smartphones, tablets and other gadgets, reams of data are being collected every second.
Retailers, manufacturers and service brands have been developing ever more sophisticated systems to capture information about every tweet, Like, banner click, product purchase or email read.
But here’s the thing: While many brands have made an effort to capture data, few have succeeded in effectively translating that data into something meaningful. Let’s look at a few well-worn data clichés to help clarify the difference between data and insights.
Data doesn’t lie
We can’t argue with this one. Data doesn’t lie or tell the truth. Data is just a collection of facts. But to turn data into insights, it is important to understand what truth needs to be uncovered.
For example, I can tell you with certainty that in September, 352 people redeemed a coupon for a $20 rebate on a prescription drug related to treating a chronic skin condition. That certainly isn’t a lie. But it’s pretty meaningless without any other information.
Look at what the data says
This is another great cliché. In all my years of analyzing data, I’ve never had information jump off the page and whisper a long-awaited series of revelations into my ear. Unfortunately, data never says anything.
So, to go back to our example, we know that 352 people redeemed a coupon. How do we turn this piece of data into insight?
First, we need to find a way to put it into some kind of context. By dividing the figure over the number of coupons distributed, we can calculate a redemption rate.
By doing this, we learned that the redemption rate for the month was 5.3 percent. To make it more meaningful, we looked at the information over time and realized that the coupon effort had delivered an overall redemption rate of 7 percent over a 12-month period.
Finally, we compared the redemption rate for the current program to results achieved for a similar product. The old product’s redemption rate was 2 percent, so the current program is well ahead of that mark.
There’s a significant difference
Nothing makes clients happier than telling them there’s a significant difference in the data.
“We conducted an x2 regression analysis with a chi-square alpha rating and a neuro-whatsis Monte Carlo simulation and have never seen anything as significant as this before!”
Except… we have one little problem: We still haven’t provided any insight! Let’s retrace our steps.
- The data didn’t lie, there were 352 redemptions.
- The data didn’t say anything, but we figured out there was a 7 percent redemption rate.
- The information was significantly different than what we saw before.
So, what’s the problem?
The problem is that we still need to provide insight into how the redemption rate has affected our goal of educating people about a particular condition and motivating them to take action (by purchasing a product).
What we eventually did was review additional data on what type of people were actually redeeming the coupon versus what our original expectations were. That final comparison revealed the insight that the coupon did two things:
First, it drove people who were already on the drug to go out and buy additional product.
- Insight #1: The coupons helped people stay on the drug and/or stock up on more product in case they needed it in the future.
Second, we were surprised to learn that the second-biggest group to redeem the coupons were folks who were not even diagnosed with the condition.
This was particularly surprising since we had anticipated that only people who were already diagnosed would be interested in redeeming the coupon.
- Insight #2: The coupons drove people to visit their doctor and get diagnosed.
This also has implications on future media strategy, as we now know that it is more important to focus coupon distribution efforts on those who are in the initial stages of learning about the condition rather than those who are already diagnosed and not using this particular product.
That’s the funny thing about data – sometimes the insights you come away with are not what you originally expected.
So the next time you come across a data report, examine whether you are being fed a pile of data or a bundle of insights.