顧客保留

顧客保留gu3 haak3 bou2 lau4（customer retention）係指一間企業或者一件產品有幾能夠「留住啲客」，例如係問「買咗件產品嘅客，一年之後有幾多百分比嘅人仲喺度買緊？」噉嘅問題^[1]。

市場學相當睇重研究顧客保留嘅嘢。

概論[編輯]

事實表明，當班客開始買同用一隻產品嗰陣，佢哋往往會慢慢噉趨向「唔再買同用件產品」－例如想像家陣有隻新嘅智能手機型號 Z，吸引到唔少人買，但出咗街 6 個月之後，間企業嘅競爭對手出咗隻更勁嘅型號 Y，於是啲客就開始慢慢噉趨向「唔再買同用 Z，改為買同用 Y」嘅情況；顧客保留講緊嘅，就係一隻產品有幾能夠長遠噉引啲客買同用^[2]^[3]：

出咗街 6 個月後，有幾多 % 嘅客仲買緊用緊隻產品？
出咗街 12 個月後，有幾多 % 嘅客仲買緊用緊隻產品？
出咗街 18 個月後，有幾多 % 嘅客仲買緊用緊隻產品？

... 如此類推－於是有市場學工作者就諗到條橋，用生還分析做研究，將終結事件設做「唔再買同用件產品」，剖析有邊啲因素能夠影響啲客會買同用件產品幾耐^[4]。事實係有市場學報告話，就算顧客保留率升咗 5%，利潤已經可以升成 25 至 85% 咁多^[3]^{:p. 1}。

統計分析[編輯]

顧客保留可以攞生還分析（survival analysis）嚟理解。

例：2016 年遊戲玩家保留研究

玩家保留（player retention）係遊戲製作上廣受關注嘅課題，指一隻電子遊戲有幾能夠長遠噉引啲玩家一路定時定候返嚟玩，可以想像成「產品 = 電子遊戲」嘅顧客保留分析^{[註 1]}^[5]。

喺 2016 年，有一班法國嘅研究者做咗份研究，想睇吓著名電子遊戲入面嘅玩家保留會受乜因素影響。佢哋搜集咗描述一班玩家嘅數據返嚟，啲數據紀錄咗啲玩家玩隻遊戲嘅紀錄，班研究者^[6]：

量度咗大拃佢哋預咗用嚟做自變數嘅變數，包括玩家响遊戲入面嘅行為（例如「開咗幾多次槍」呀噉）；
然後佢哋計咗個生還模型出嚟；
佢哋用比例危機模型睇吓啲自變數會點樣影響「生還時間」。

研究者發現，（簡化講）玩家响遊戲入面用武器用得愈多，佢哋嘅「生還時間」傾向會愈長。噉亦即係話，電子遊戲工作者可以透過量度玩家响遊戲裏面嘅行為嚟預測佢哋會玩幾耐－而呢點對電子遊戲設計嚟講係樣有用嘅資訊^[6]。

註釋[編輯]

↑ 技術性啲噉講，即係玩家保留係顧客保留嘅狹義化。

睇埋[編輯]

攷[編輯]

↑ Vroman, H. William; Reichheld, Frederick F. (1996). "The Loyalty Effect: The Hidden Force Behind Growth, Profits, and Lasting Value". Academy of Management Perspectives. Harvard Business School Press. 10 (1): 88–90.
↑ Page, M., Pitt, L., Berthon, P., & Money, A. (1996). Analysing customer defections and their effects on corporate performance: the case of Indco. Journal of Marketing Management, 12(7), 617-627.
↑ ^3.0 ^3.1 Zhang, G. (2007, September). Customer retention based on BP ANN and Survival Analysis (PDF). In 2007 International Conference on Wireless Communications, Networking and Mobile Computing (pp. 3406-3411). IEEE.
↑ Mavri, M., & Ioannou, G. (2008). Customer switching behaviour in Greek banking services using survival analysis. Managerial Finance.
↑ Demediuk, S., Murrin, A., Bulger, D., Hitchens, M., Drachen, A., Raffe, W. L., & Tamassia, M. (2018, January). Player retention in league of legends: a study using survival analysis. In Proceedings of the Australasian computer science week multiconference (pp. 1-9).
↑ ^6.0 ^6.1 Allart, T., Levieux, G., Pierfitte, M., Guilloux, A., & Natkin, S. (2016, September). Design influence on player retention: A method based on time varying survival analysis. In 2016 IEEE Conference on Computational Intelligence and Games (CIG) (pp. 1-8). IEEE.

[5] 技術性啲噉講，即係玩家保留係顧客保留嘅狹義化。

[1] Vroman, H. William; Reichheld, Frederick F. (1996). "The Loyalty Effect: The Hidden Force Behind Growth, Profits, and Lasting Value". Academy of Management Perspectives. Harvard Business School Press. 10 (1): 88–90.

[Page1996-2] Page, M., Pitt, L., Berthon, P., & Money, A. (1996). Analysing customer defections and their effects on corporate performance: the case of Indco. Journal of Marketing Management, 12(7), 617-627.

[Zhang2007-3] 3.0 ^3.1 Zhang, G. (2007, September). Customer retention based on BP ANN and Survival Analysis (PDF). In 2007 International Conference on Wireless Communications, Networking and Mobile Computing (pp. 3406-3411). IEEE.

[4] Mavri, M., & Ioannou, G. (2008). Customer switching behaviour in Greek banking services using survival analysis. Managerial Finance.

[Demediuk2018-6] Demediuk, S., Murrin, A., Bulger, D., Hitchens, M., Drachen, A., Raffe, W. L., & Tamassia, M. (2018, January). Player retention in league of legends: a study using survival analysis. In Proceedings of the Australasian computer science week multiconference (pp. 1-9).

[allart2016-7] 6.0 ^6.1 Allart, T., Levieux, G., Pierfitte, M., Guilloux, A., & Natkin, S. (2016, September). Design influence on player retention: A method based on time varying survival analysis. In 2016 IEEE Conference on Computational Intelligence and Games (CIG) (pp. 1-8). IEEE.

[1]

[2]

[3]

[4]

[註 1]

[5]

[6]