Never be too greedy and chew too much

        There is an American proverb that goes, 'Don't bite off more than you can chew.'. This sentence also applies to the introduction of AI in enterprises: at the beginning, do not be too greedy, set the scope too wide, and stretch the front line too long. In the end, you do not know how to converge and cannot timely display specific results, making it difficult to continue. So my suggestion is to first identify the pain point that can be easily bitten.

                What does' bite 'mean? Enterprises have many pain points, but which problem is the most painful, clear, specific, measurable, and comparable? Leaders need to be able to explain it clearly, clarify the causal relationship between the problem and the result, and provide the team with clear reference and verification mechanisms. This is the pain point that can be bitten.

        When the pain points are identified and the causal relationship is clear enough, everyone can quickly start working and see concrete results, which can generate confidence in the introduction of AI and transformation. Such an effect is more powerful than any propaganda.

        AI is not a panacea, leaders need to spend time understanding where the pain points are first. Every time I encounter a company, I first ask, 'What are your pain points?' instead of asking, 'How do you use AI?' Many organizations have cross departmental issues related to human resources and processes that do not require AI. Some existing tools can solve them, and even communication is poor due to a lack of understanding of end-to-end processes.

        In addition, when clarifying pain points, it is important to avoid treating them as individual projects and ignoring the underlying systems and processes behind many issues.

        Why is it so easy in practice? Because when a company introduces AI or undergoes transformation, in order to let employees know that this is a key "project", it may initially establish a project, followed by project management, tracking schedule and performance. This planned process of advancing and transforming the "company state" includes specific and concrete "outputs", just like the pain points solved by introducing AI, but that is not the final "state" that the transformation aims to achieve.

        For example, everyone in logistics units can use AI to complete their work, IT uses AI to develop software and maintain operations, and core business is optimized and innovated through AI models. These transformations and developments vary, but they cannot be completed in the short term and are difficult to approach with typical project management thinking. The ultimate result of AI transformation still needs to be linked to the medium and long-term key financial indicators of the enterprise.

        However, everything is difficult to start with, and introducing AI is a continuous process of correction and optimization. Moreover, with the rapid development of technology, as long as you start from the pain points that can be bitten, you have already joined the ranks of AI transformation enterprises. (This article is reproduced from the 2006 issue of Business Weekly)

        #