Too much data, wrong data
A bike shop owner asks why sales dropped in May. An analyst has a till file, a weather site, a rival's price list and a folder of reviews.
Using all of them would take weeks. Using the wrong one gives a wrong answer. The skill is to pick the few sources that fit the question.
Key words
| Word | What it means | Example |
|---|---|---|
| Requirement | Something the analysis must do or answer. | Explain the drop in May sales |
| Parameter | A limit or setting that shapes the task: the time, place, group and detail. | Only May and April, this shop |
| Internal source | Data your own organisation holds. | Till file |
| External source | Data another organisation holds. | Weather records |
| Data type | The kind of value a field holds, such as a number or text. | Whole number |
| Numeric | Data that is a number you can calculate with. | Sales of 12 |
| Temporal | Data about time: a date, a time or a length of time. | 2026-05-14 |
| Media | Data that is a picture, sound or video. | A photo of a broken bike |
The cells students fill in are left blank in this preview.
By the end of this lesson
Note
I can read what a question needs, pick internal and external sources for it, and name the data types to look for.
Starter: what you already know
Have a go
Starter: what you already know
3 questionsYour answers are marked here and are not kept. A college account keeps every student’s marks.
Worked example: May sales
| Part of the question | Source chosen | Data type |
|---|---|---|
| How many bikes sold each day | Till file (internal) | Numeric and temporal |
| Was the weather bad | Council weather record (external) | Numeric and temporal |
| What did customers say | Shop review box (internal) | Text |
| What did the broken bikes look like | Repair photos (internal) | Media |
The cells students fill in are left blank in this preview.
How to choose between sources
Ask four things. Does it cover the right time? Does it cover the right group? Is it accurate enough? Can I use it, and what does it cost?
Start with internal data, because it is quick and often free. Add external data only to answer what inside data cannot.
Try it: filter sources by age and cost
sources = [["till", 1, 0], ["census", 60, 0], ["survey firm", 3, 500]]
for one in sources:
if one[1] <= 12 and one[2] <= 0:
print("Keep:", one[0])
else:
print("Skip:", one[0])
# Try it: change 0 to 1000 in the cost test and press Run.