[Digital Today reporter Jinju Hong (홍진주)] Users who repeatedly use ChatGPT for the same type of work have been advised that it is far more efficient to create and use reusable opening prompts, rather than re-explaining the purpose and conditions each time in a new conversation.
On July 6, local time, IT media outlet TechRadar introduced the habit of creating "reusable prompts" as a way for users who apply generative AI to repetitive tasks to significantly cut working time.
The key is not saving a previous conversation as it is, but organising an effective way of working into a single prompt. In general, users explain the work goal each time they start a new chat, add necessary background information and then request the desired tone and output format again. They then repeat a process of revising the response or refining it to be more concise.
A reusable prompt is a way to reduce that process. If a prompt that lays out the goal, tone, output format and constraints in advance is entered as a new conversation begins, the first response is more likely to be close to the desired direction and the number of later revisions also declines, it said.
Such a prompt can be created easily from a conversation that previously produced a satisfactory result. TechRadar said that after a conversation ends, if a user asks, "Convert this conversation into a reusable prompt I can use in a new chat next time. Include the goal, tone, format, constraints and work steps, and generalise it so it can be used for other similar tasks," ChatGPT will organise an opening prompt.
TechRadar gave meal planning as an example. The writer explained that instead of planning meals from scratch each week, they summarise repeatedly requested conditions such as "quick-to-make menus," "low-cost ingredients," "using leftovers" and a "shopping list by supermarket section" in a single prompt. After that, they can quickly generate a new meal plan by adding only what ingredients remain in the refrigerator or how many days of meals are needed.
A reusable prompt also includes detailed conditions. For example, it can specify in advance that the tool should first ask what ingredients are currently on hand before recommending menus, organise shopping lists by supermarket area, suggest ingredients that can be used across multiple meals and present alternative menus for days when cooking is difficult. The outlet said this method significantly reduced time spent on daily meal planning.
Even so, a reusable prompt does not always guarantee better results. TechRadar pointed out that the purpose of a prompt is to make results more predictable, not to automatically improve a poor way of working. If the content is overly vague or, conversely, contains only excessively detailed instructions, it may simply deliver ordinary results more quickly.
For that reason, it stressed that it is important to select conversations that were actually highly satisfying and turn them into reusable prompts, rather than saving any conversation. The writer said, "A good conversation is not just let go, but turned into a starting point for the next task."
In the industry, this type of use is seen as aligned with a trend of using generative AI not as a simple search tool but as a personalised work assistant. For users who repeat the same type of work, starting a conversation based on a prompt that contains their own working method, instead of explaining from scratch each time, can be a way to boost productivity, it said.