| A | B | C | D | E | F | |
|---|---|---|---|---|---|---|
1 | Prompt Technique | What it is | What is it useful for? | Limitations | Sample Prompt | Additional Insights |
2 | Zero-Shot Prompting | Gives a single instruction or input to the AI with no context or examples. | Answering straightforward customer inquiries like 'When was Telkomsel founded?' | Struggles with complex reasoning and lacks control over output, leading to hallucinations. | When was Telkomsel founded?' | Simple and fast but lacks depth; best for factual queries. |
3 | Few-Shot Prompting | Provides context, details, and/or examples to enrich the AI’s understanding of the task. | Generating engaging SMS reminders for Telkomsel customers with specific messaging styles. | May have difficulty handling tasks that require deep logical steps and reasoning. | Here are examples of engaging SMS reminders: [example 1, example 2]. Now, write a similar reminder for customers whose credit will expire in 3 days.' | Helps AI understand patterns and context, making it great for structured outputs. |
4 | Chain of Draft Prompting | Generates minimal yet informative intermediate steps to ensure concise reasoning. | Drafting and refining promotional emails for Telkomsel’s new 5G service package. | May not be suitable for highly intricate reasoning that requires extensive elaboration. | First, create a rough draft of a promotional email for Telkomsel’s 5G package. Then refine it for better engagement and optimize the call-to-action.' | Efficient and cost-effective; reduces token usage but may oversimplify reasoning. |
5 | Chain of Thought (CoT) Prompting | Guides AI to break down problems into sequential, logical steps to improve accuracy. | Developing a strategy to reduce customer churn by breaking it into logical steps. | Complex to set up and not suitable for one-off tasks. | Develop a strategy to reduce customer churn for Telkomsel prepaid services. First, identify reasons for churn. Then, suggest targeted retention strategies.' | Best for handling complex reasoning tasks that require step-by-step breakdown. |
6 | Tree of Thoughts (ToT) Prompting | Explores multiple reasoning paths instead of a single linear approach. | Creating different marketing approaches for Telkomsel’s home internet service and evaluating them. | Higher computation cost and slower response time due to multiple reasoning paths. | Generate three different approaches to market Telkomsel's home internet service. Compare benefits and drawbacks for each and choose the best approach for urban areas.' | Explores multiple solutions, making it ideal for decision-making and brainstorming. |
7 | Retrieval Augmented Generation (RAG) | Enhances AI responses by retrieving and integrating external data sources. | Using Telkomsel's customer FAQ database to generate accurate responses to service-related inquiries. | Requires reliable external data sources; incorrect retrieval may lead to misinformation. | Using Telkomsel's customer FAQs, generate accurate responses to questions like 'How do I change my email access?' and 'How long does verification take?' | Improves AI accuracy with external data but depends on the quality of retrieved sources. |
8 | Zero-Shot Chain of Thought Prompting | A variation of CoT where AI is instructed to 'think step by step' without providing examples. | Structuring AI’s reasoning for complex tasks such as launching a new Telkomsel loyalty program. | AI may still produce errors if logical reasoning steps are not well-structured. | Telkomsel is launching a customer loyalty program with exclusive benefits. How can we communicate these benefits effectively? Let’s think step by step.' | Quick way to guide AI reasoning without structured setup; works well for medium complexity tasks. |
9 | ||||||
10 | ||||||
11 | ||||||
12 | ||||||
13 | ||||||
14 | ||||||
15 | ||||||
16 | ||||||
17 | ||||||
18 | ||||||
19 | ||||||
20 | ||||||
21 | ||||||
22 | ||||||
23 | ||||||
24 | ||||||
25 | ||||||
26 | ||||||
27 | ||||||
28 | ||||||
29 | ||||||
30 | ||||||
31 | ||||||
32 | ||||||
33 | ||||||
34 | ||||||
35 | ||||||
36 | ||||||
37 | ||||||
38 | ||||||
39 | ||||||
40 | ||||||
41 | ||||||
42 | ||||||
43 | ||||||
44 | ||||||
45 | ||||||
46 | ||||||
47 | ||||||
48 | ||||||
49 | ||||||
50 | ||||||
51 | ||||||
52 | ||||||
53 | ||||||
54 | ||||||
55 | ||||||
56 | ||||||
57 | ||||||
58 | ||||||
59 | ||||||
60 | ||||||
61 | ||||||
62 | ||||||
63 | ||||||
64 | ||||||
65 | ||||||
66 | ||||||
67 | ||||||
68 | ||||||
69 | ||||||
70 | ||||||
71 | ||||||
72 | ||||||
73 | ||||||
74 | ||||||
75 | ||||||
76 | ||||||
77 | ||||||
78 | ||||||
79 | ||||||
80 | ||||||
81 | ||||||
82 | ||||||
83 | ||||||
84 | ||||||
85 | ||||||
86 | ||||||
87 | ||||||
88 | ||||||
89 | ||||||
90 | ||||||
91 | ||||||
92 | ||||||
93 | ||||||
94 | ||||||
95 | ||||||
96 | ||||||
97 | ||||||
98 | ||||||
99 | ||||||
100 |