1
00:00:00,930 --> 00:00:04,860
Once we have business knowledge, we need to gather the relevant data.

2
00:00:06,420 --> 00:00:07,590
There are three steps to it.

3
00:00:08,610 --> 00:00:10,530
First is to identify the data needed.

4
00:00:12,730 --> 00:00:15,670
We know what data we need from the research.

5
00:00:17,620 --> 00:00:22,450
Second is requesting this data from relevant people within and outside the organization.

6
00:00:23,590 --> 00:00:28,870
And lastly, when you receive the data from different teams, you need to do a quality check on the

7
00:00:28,870 --> 00:00:29,590
data received.

8
00:00:32,360 --> 00:00:34,910
They are being requested can be of two types.

9
00:00:35,420 --> 00:00:42,670
Internally and externally, the internal data is the data which is collected by you, audio resource

10
00:00:42,680 --> 00:00:44,300
team or your organization.

11
00:00:45,800 --> 00:00:50,450
For example, sales data on monthly money spent on a particular type of promotion.

12
00:00:51,730 --> 00:00:55,480
Is the data available with your organization and is internal data?

13
00:00:57,410 --> 00:01:01,580
External data is data collected and maintained by external data sources.

14
00:01:02,390 --> 00:01:08,630
For example, governments maintain a population centers data which can be used to determine the demography

15
00:01:08,720 --> 00:01:09,740
in a particular region.

16
00:01:10,390 --> 00:01:13,490
Or you can buy data from several third party vendors also.

17
00:01:15,290 --> 00:01:18,490
Now, let's go back to the example of guard abandonment.

18
00:01:19,460 --> 00:01:22,900
We decided to go meet some people and do some secondary research.

19
00:01:23,840 --> 00:01:27,800
Let us see how business knowledge tells us what data to gather.

20
00:01:30,800 --> 00:01:33,160
So first we went to the marketing team.

21
00:01:34,190 --> 00:01:39,860
They tell us that customers who are coming to the website are coming from three different channels.

22
00:01:41,050 --> 00:01:48,520
50 percent come from e-mail marketing, 30 percent from organic search and 20 percent from AdWords marketing.

23
00:01:50,270 --> 00:01:56,570
Probably whether the customer will eventually buy or not buy depends on which channel that customer

24
00:01:56,570 --> 00:01:57,350
is coming from.

25
00:01:58,490 --> 00:02:03,050
So we decided to get this sorted out to our website for all the customers.

26
00:02:04,340 --> 00:02:11,390
Next product team tells us that there are three steps in the buying process after adding a product to

27
00:02:11,410 --> 00:02:11,780
cart.

28
00:02:13,800 --> 00:02:17,940
Card review, adding address or personal details.

29
00:02:18,150 --> 00:02:19,530
And then finally, the payment.

30
00:02:21,020 --> 00:02:23,480
Maybe there is some issue in any particular step.

31
00:02:24,720 --> 00:02:28,850
Let's get a the God abandonment location for all the customers.

32
00:02:30,520 --> 00:02:35,590
Then we did some industry reports regarding the issue and found out the below observation.

33
00:02:36,840 --> 00:02:39,670
Customers who add expensive items in their car.

34
00:02:39,900 --> 00:02:41,790
Keep it in the card for longer duration.

35
00:02:43,610 --> 00:02:48,610
So against each customer order, we should also get the total card value for each customer.

36
00:02:51,680 --> 00:02:57,740
Lastly, when we did our dry run, we saw that there is a small town willing to read the experience.

37
00:02:59,000 --> 00:03:02,980
So let's get these ratings from the customer experience team also.

38
00:03:05,010 --> 00:03:09,600
You can see how business understanding is telling us what data to collect.

39
00:03:11,150 --> 00:03:18,500
Once you have the data, you need to tidy it up and clearly define what each variable stands for.