{"id":4774,"date":"2010-10-06T11:55:52","date_gmt":"2010-10-06T11:55:52","guid":{"rendered":"http:\/\/www.smartdatacollective.com\/index.php\/post\/practical-data-analytics-when-close-enough-good-enough\/"},"modified":"2010-10-06T11:55:52","modified_gmt":"2010-10-06T11:55:52","slug":"practical-data-analytics-when-close-enough-good-enough","status":"publish","type":"post","link":"https:\/\/www.smartdatacollective.com\/practical-data-analytics-when-close-enough-good-enough\/","title":{"rendered":"Practical Data Analytics \u2013 When is \u201cclose enough\u201d good enough?"},"content":{"rendered":"

\"amazing<\/a><\/p>\n

<\/p>\n

\"amazing<\/a><\/p>\n

Data analytics<\/a> isn\u2019t always about getting the right<\/em> answer \u2013 it\u2019s often about getting useful<\/em> answers that help make the best decisions. There are many instances where the right answer doesn\u2019t even exist. An example is if we\u2019re using social data or a predictive model<\/a>.  So how do we know when \u201cclose enough\u201d is good enough?<\/p>\n

Let\u2019s say you are using data analytics to help prevent undesirable turnover of high potential employees in your organization. You have a model that predicts which high potential employees will quit next. So, the idea is to alert management so they can intervene.  In this case, there is no right answer \u2013 until someone quits, and then it\u2019s too late. <\/span>The model provides an indication along some continuum of the likeliness to quit.  You will draw a line on that continuum and intervene for every employee who falls above the line.  Hopefully you have the model inputs and an understanding of how the model works.<\/p>\n

If the answer from the model is low, medium or high, is that \u201cclose enough\u201d to help you decide where to place the line? What if the model produces a whole number from 0 to 10? Can you make a decision about where to intervene?  What if it provides half-steps (8.5, 9.0, 9.5)? What about tenths (8.1, 8.2, 8.3)? Hundredths (8.01`, 8.02, 8.03)? Pretty soon you\u2019ll reach a point where an increase in precision doesn\u2019t really affect your decision, and you\u2019ve found your definition of \u201cgood enough\u201d.<\/p>\n

If the model doesn\u2019t provide acceptable precision, you can assess the cost of increasing precision against the cost and consequence of intervening unnecessarily. Once you get acceptable precision, where you place the line will be a balance between your tolerance for risk (someone quits without intervention) and the cost and consequence of intervening unnecessarily.<\/p>\n

As with any kind of change management, your success will be heavily influenced by the way you communicate this to others. If you think some users won\u2019t find the results \u201cgood enough\u201d, you should manage their expectations and either discuss the process you\u2019ll take to improve the results, or discuss the economics of why improving the results will cost more than the problem is worth. In either case, it\u2019s more likely that a bad decision will be due to a lack of understanding than a lack of information.<\/p>\n

Steve McDonnell
Spotfire Blogging Team<\/p>\n

Image Credit:  thechive.com<\/p>\n

<\/p>\n

<\/a><\/div>\n

\"\"<\/p>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":33,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_seopress_robots_primary_cat":"","_seopress_titles_title":"","_seopress_titles_desc":"","_seopress_robots_index":"","footnotes":""},"categories":[2],"tags":[234],"class_list":{"0":"post-4774","1":"post","2":"type-post","3":"status-publish","4":"format-standard","6":"category-business-intelligence","7":"tag-data-analytics"},"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/posts\/4774","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/users\/33"}],"replies":[{"embeddable":true,"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/comments?post=4774"}],"version-history":[{"count":0,"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/posts\/4774\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/media?parent=4774"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/categories?post=4774"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.smartdatacollective.com\/wp-json\/wp\/v2\/tags?post=4774"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}