When building out an automation strategy, one of the first and most important questions you need to ask is: what should we automate? Not everything needs to be automated, and trying to do so can lead to wasted effort. Here’s a practical, step-by-step approach to making informed decisions around test automation.
1. Brainstorm Scenarios to Automate
Before jumping into code, take a moment with your team to identify what’s worth automating. A good starting point is to brainstorm possible user scenarios for any new feature. Don’t filter anything out at this stage — the goal is to cast a wide net.
For example, in the context of the Stickerfy application, some potential scenarios might be:
- Viewing products available for purchase on the homepage
- Adding a sticker to the cart
- Adding multiple of the same or different stickers
- Viewing items in the cart and the total
- Removing one or all items from the cart
- Clearing the entire cart
- Completing the checkout process
You don’t need to come up with an exhaustive list immediately — just get all the ideas on the table.
2. Evaluate the Value of Each Scenario
Once you have a list of scenarios, it’s time to assign a value score (1 to 5) to each. This score reflects how important each scenario is for the user and the product.
Factors to consider:
- How critical is the feature to the core functionality?
- How likely is it to be fixed quickly if broken?
- Is the scenario distinct enough to warrant its own test?
Example evaluations:
- View Products for Sale – Score: 5 (Critical and distinct)
- Add Item to Cart – Score: 5 (High value)
- Add Multiple Items to Cart – Score: 4 (Less distinct, but still important)
- Remove Item from Cart – Score: 4 (Useful, but not as vital as adding)
- Checkout – Score: 5 (Revenue-driving feature)
This scoring helps surface the most valuable scenarios to consider for automation.
3. Assess the Risk
Next, evaluate the risk score for each scenario, also on a scale of 1 to 5.
Risk is based on:
- Impact: How bad would it be if this feature broke?
- Frequency: How often is this feature used?
Example risk assessments:
- View Products for Sale – Score: 5 (Frequently used and essential)
- Add Item to Cart – Score: 5
- Add Multiple Items to Cart – Score: 4
- Checkout – Score: 5 (High impact and usage)
Risk scoring helps you understand which features are most likely to affect users if something goes wrong.
4. Estimate the Cost of Automation
Not all scenarios are equal when it comes to the effort required to automate them. Assign a cost score (1 to 5), where 5 means it’s easy and fast to automate, and 1 means it’s complex or time-consuming.
Cost factors:
- How easy is the test to script?
- How quickly can the script be written?
Example cost scores:
- View Products for Sale – Score: 5 (Simple)
- Add Item to Cart – Score: 5
- Remove Item from Cart – Score: 4 (Requires adding an item first)
- Remove All Items from Cart – Score: 3 (Requires multiple additions)
- Checkout – Score: 4 (Depends on the cart state)
Cost scoring helps you estimate effort and allocate resources effectively.
5. Select What to Automate
Now that you’ve scored each scenario for value, risk, and cost, you can total the scores and identify your top candidates for automation.
Let’s say you use this scale:
- 13–15: Automate
- 12 or below: Hold off for now
Using this method, you might decide to automate:
- Viewing products
- Adding items to the cart
- Checkout
These scenarios score high in all three areas and provide the best return on investment.
Final Thoughts
There’s no one-size-fits-all approach to automation. But by evaluating value, risk, and cost, you can make more objective, data-driven decisions. This process helps ensure your test suite focuses on what matters most — improving product quality and minimizing risk.
And remember: this framework is flexible. Adapt it to your team’s goals, product, and available resources. Automation is a tool, not a rule — use it wisely.