Do you have enough data, is it good enough?

Do you have enough data, is it good enough?

In this short article, I would like to discuss a question I receive frequently in our client conversations:

Do I have enough data, is it good enough?

We should answer these questions within the context of what do you need the data for and where you need to make it more and better with what urgency?

You need data for the purpose of making your business better.

What you need to do for that depends on what your business looks like right now.

For example you may want to increase retention of customers with high customer lifetime value and in order to increase their stickiness you want to xsell to them a new and helpful product they need.?Your cross sell prediction machine learning models can tell you exactly to whom you should offer the product. This increases customer outreach efficiency, we have seen increases as much as 20x. Especially when the xsell model comes with xsell segment explanations, the model will help you understand why people buy and dont, what is right or wrong about your product and even what your competitor is offering and to whom.?

By examining your model, its accuracy in different segments and the xsell results, you can?go back and decide on your next data investment. Maybe you need more data points to increase the breadth of your data, such as collect information on your product usage, use external data sources. Maybe you need to increase accuracy for some segments and need to collect data from customers in those segments, increasing the depth of your data.?

If you follow this approach, you will first use the existing data to create a machine learning model to help you understand and make your business better. Then you decide on where you need to invest next for data collection and cleaning. And you keep doing this as you create more models and take actions with them, again to make your business better.?

Comments? Do you want to learn what you can do right now using your data? Please reach out to me directly, or email [email protected]

PS: To put this conversation in context:?

At TAZI, we help you to make your business better by providing you with an easy to use SaaS ML platform. The platform does 3 things (well it actually does more):?

1- helps you evaluate and fix your data quality,?

2-allows you to see your data and machine learning model explanations on dynamic and customized dashboards and?

3-allows you to create, productionize, monitor and update machine learning solutions with your teams easily, in weeks, and taking up much little IT time than usual.

You can do these with a platform like us even if you do not have any experienced data scientists in your team right now.?

要查看或添加评论,请登录

Zehra Cataltepe的更多文章

  • Are You Ready to Perform in an AI Concert?

    Are You Ready to Perform in an AI Concert?

    There are some song clips that you not only listen to but also watch every performer who contributed to the song. Those…

    2 条评论
  • Composing AI, GenAI, and Humans: Transforming Customer Complaints

    Composing AI, GenAI, and Humans: Transforming Customer Complaints

    Handling customer complaints in industries like banking, finance, insurance, retail, and tech is a complex process. It…

    6 条评论
  • NRF 2023 Insights!

    NRF 2023 Insights!

    NRF 2023 was great! New York City was as amazing and beautiful as ever. It seems to be getting back on its feet after…

    3 条评论
  • Agile AI and Retail

    Agile AI and Retail

    Retail has always been changing, and even more since Covid-19. While data and data science are being used in retail and…

    1 条评论
  • Machine Learning to Support Decisions not to Make Them

    Machine Learning to Support Decisions not to Make Them

    My dream is to create machine learning solutions, so that when they (or people who use them) have False Positives, you…

  • The best AI to use in the future for decision-making is…

    The best AI to use in the future for decision-making is…

    I attended the Gartner IT Symposium this week (virtual) and felt like sharing my highlights. Especially these two talks…

社区洞察

其他会员也浏览了