When applying automation function planning for the first time, we have organized the structures, screens, and priorities that are often blocked for non-majors. We have organized key standards, common mistakes, inspection points, and next actions in one place so that you can directly attach them to the actual planning and execution flow, so apply them right away.
Automation function planning is the main topic of this guide. If you are applying Automation function planning in a real project, start with the structure and checks below.
This article is organized based on the points that often get stuck when attaching automation function planning to actual work flow.
It is safer to check the current environment and official documents before actual application.
When thinking of service ideas these days, automation is almost always mentioned. This is because it seems like a smarter service if it summarizes, automatically classifies, recommends, and sends notifications using AI. The problem is that if you add too much automation from the beginning, the service becomes heavier and verification becomes difficult. Automation may be a cool-looking feature, but in the planning stage, it’s more important to decide where to go.
The tasks that need to be automated first are repetitive and simple tasks
Tasks that are best suited to automation are tasks that humans repeat every time, but the judgment itself is not complicated. For example, functions that create title candidates for long articles, classify articles into tags, or send notifications at certain times have a great automation effect. Conversely, areas with high context and responsibility, such as final approval, sensitive judgment, and customer service, are difficult to fully automate from the beginning.
The structure that can be operated manually comes first
A common pitfall for beginners is to design automation first. However, the service must also be run manually first. Automating processes that cannot be done manually also speeds up the problem. For example, if the classification criteria are still unclear and you start with automatic AI classification, it will be difficult to trust the results. Automation is not a magic that creates something that does not exist, but is closer to a device that makes the already organized flow run faster.
Automation functions must also consider costs and exceptions
Automation is convenient, but it’s not free. AI calls can be expensive, and it takes time to check for incorrect results. You also need to think about how to handle it again if it fails and whether users can modify it themselves. So, at the planning stage, we need to look at “Will this automation really reduce repetition time?” and “Can people correct it right away when something is wrong?”
If you set the automation standard in one line, shaking will be reduced
In the beginning, it is a good idea to set a line of automation principles for each service. For example, criteria such as “automate only repetitive, low-risk tasks” or “only allow structures with final human approval.” If you have these standards, even if a new idea comes to mind, you can judge it without blindly attaching it to it. Automation should not be a competition to put in a lot, but a choice that makes the service flow lighter.
It is safer to attach it as an auxiliary function at first.
The most realistic approach for non-majors is to attach automation as an auxiliary function rather than the main function. For example, a user can write a title and have AI suggest candidates, or a person can post a post first and AI can recommend tags. If you start like this, you can take advantage of the benefits of automation while reducing the risk of failure.
Automation is a feature that stands out and makes us want to see it, but well-designed automation is less noticeable. If it’s less difficult for users and less repetitive for operators, that’s good enough automation.
Automation is not good if you add a lot of it, but it becomes valuable when you put it in the most effective place within the service flow. If you are examining your idea today, ask yourself, “Can this function be done manually by a person, or is it repetitive so it has great automation value?” In the next article, we will continue to look at why documentation becomes more important as functions increase.
As an easy example,
For example, a feature in a blog writing tool that automatically selects title candidates is a good idea to automate. On the other hand, if you let AI make the final decision on whether to publish or not, mistakes may increase. So, at first, it is safer to attach it as an auxiliary function, such as “Recommendation is automatic, final selection is human.”
Quick checklist for Automation function planning
Use this checklist before you apply Automation function planning in an actual post or product flow.
- Is the first action obvious as soon as the user lands on the page?
- Are intermediate steps simple enough that buttons and explanations do not overlap?
- Does the result naturally lead to a next action instead of a dead end?
- Could you explain the structure again later without adding unnecessary screens?
Related posts
- Data structure planning: It is possible to plan an app without knowing the data structure.
- development documentation: Why do collaborating developers consider documentation important?
Things to verify before you apply it
- Tool UI and function configuration may vary depending on the time, so it is safer to check again based on the current version.
- Although this may work well for small examples, in projects with large existing code bases, the scope of modifications can quickly become large if the structure is not broken down first.
