The college app is too slow
Students say the college timetable app is slow. The head of IT asks you to fix it.
Do you start changing code? No. First you must know what the problem is. There is a plan that works for almost any problem. It has six steps.
Key words
| Word | What it means | Example |
|---|---|---|
| Strategy | A plan for how to solve a problem. | the six steps |
| Define | Say clearly what the problem is. | the app takes 20 seconds to open |
| Analyse | Study the information to find out what it means. | it is slow only at 9am |
| Implement | Carry out the plan. | add the fix |
| Review | Check if the solution worked. | is it fast now? |
| Top-down | Start with the whole problem and break it into parts. | plan the app, then its screens |
| Bottom-up | Start with small parts and join them. | build the payment tool first |
| Modularisation | Splitting a program into separate modules with one job each. | a login module |
| Root cause analysis | Looking for the real reason a problem happens. | ask why five times |
| Stakeholder | A person who cares about the result. | students and teachers |
The cells students fill in are left blank in this preview.
Can you guess the order?
Have a go
Can you guess the order?
3 questionsYour answers are marked here and are not kept. A college account keeps every student’s marks.
The six steps on the slow app
| Step | What you do | The slow timetable app |
|---|---|---|
| 1 Define the problem | Say what is wrong, in one clear sentence. | The app takes 20 seconds to open at 9am. It should take 2. |
| 2 Gather information | Collect facts: ask people, read records, measure. | Ask 10 students and 3 teachers (the stakeholders). Check the server (the computer that runs the app) records. Time the app at different hours. |
| 3 Analyse the information | Look for patterns and causes. | It is slow only when 500 students sign in together. |
| 4 Make a plan of action | Choose what to do, in what order, and who does it. | Save the timetable so it is not loaded again for each student. |
| 5 Implement a solution | Carry out the plan. | Write the change and put it on a test server first. |
| 6 Review the solution | Check it worked. Ask the same people. | Time it at 9am again. Now it takes 2 seconds. |
The cells students fill in are left blank in this preview.
It is not always a straight line
data-callouts.warning.label
If the review shows the problem is not solved, go back. You may need to define the problem again or gather more information. Many real projects go round more than once.
Many strategies, one problem
You now know several strategies. Top-down and bottom-up decide how to build. Root cause analysis decides why something failed. The six steps give an overall plan.
A good developer chooses the right one for the problem. You often use more than one together.
Which strategy fits?
| Problem | Good strategy | Why |
|---|---|---|
| A new app with a clear list of needs | Top-down | You know the goal and can break it up. |
| You must build with a trusted payment tool you already have | Bottom-up | You start from the part that exists. |
| The same problem keeps coming back | Root cause analysis | You must stop it coming back. |
| A large program with many people working on it | Modularisation | Each person owns a module. |
| A request that is not clear: make it better | The six steps | You must first define and gather information. |
The cells students fill in are left blank in this preview.