Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Seminar #50 - Salesforce Data, Clean, Visualize...

Seminar #50 - Salesforce Data, Clean, Visualize, Analyze, & Dashboard

You can access Salesforce data directly inside Exploratory with the latest release (v6.6). In this seminar, Kan will demonstrate how you can access the data and introduce some of the data visualization and analysis methods to maximize the value of your Salesforce data.

- Access and Query Data
- Transform and Visualize Data
- Create Dashboard
- Forecast Leads with Prophet
- Predict for Customer Win
- Use Parameters to Interact

Subscribe ↓
https://www.youtube.com/channel/UCOVfLaSQBvMRwZCyiccq4Iw

Twitter ↓
https://twitter.com/ExploratoryData

UI Tool: Exploratory(https://exploratory.io/)
Exploratory Online Seminar: https://exploratory.io/online-seminar

Kan Nishida

June 30, 2021
Tweet

More Decks by Kan Nishida

Other Decks in Technology

Transcript

  1. Kan Nishida CEO/co-founder Exploratory Summary In Spring 2016, launched Exploratory,

    Inc. to democratize Data Science. Prior to Exploratory, Kan was a director of product development at Oracle leading teams to build various Data Science products in areas including Machine Learning, BI, Data Visualization, Mobile Analytics, Big Data, etc. While at Oracle, Kan also provided training and consulting services to help organizations transform with data. @KanAugust Speaker
  2. 4 Questions Communication (Dashboard, Note, Slides) Data Access Data Wrangling

    Visualization Analytics (Statistics / Machine Learning) ExploratoryɹModern & Simple UI
  3. Agenda • Get Data • Clean and Transform Data •

    Visualize Data • Create, Publish, Share, and Schedule Dashboard • Parameterize Dashboard • Forecast Leads • Predict Win 6
  4. Agenda • Get Data • Clean and Transform Data •

    Visualize Data • Create, Publish, Share, and Schedule Dashboard • Parameterize Dashboard • Forecast Leads • Predict Win 7
  5. Agenda • Get Data • Clean and Transform Data •

    Visualize Data • Create, Publish, Share, and Schedule Dashboard • Parameterize Dashboard • Forecast Leads • Predict Win 20
  6. Agenda • Get Data • Clean and Transform Data •

    Visualize Data • Create, Publish, Share, and Schedule Dashboard • Parameterize Dashboard • Forecast Leads • Predict Win 25
  7. 28 You can customize the chart by updating the column

    assignments and the chart configuration.
  8. Stage Hearing Conversion Proposal Prospecting Character (Categorical) 35 Stage Level

    Prospecting 1 Hearing 2 Proposal 3 Conversion 4 Factor (Ordinal)
  9. 36

  10. 37

  11. Agenda • Get Data • Clean and Transform Data •

    Visualize Data • Create, Publish, Share, and Schedule Dashboard • Parameterize Dashboard • Forecast Leads • Predict Win 43
  12. 45 Numbers Analytics - Forecasting Analytics - Cohort Analysis Map

    Chart You can bring in various types of numbers and charts.
  13. 52

  14. 56 By clicking the ‘Re-Import Data’ button, you can query

    the latest data from Salesforce and apply all the data wrangling steps that are required to generate the charts.
  15. 64 If they don’t have Exploratory accounts yet they can

    sign up for free ‘Viewer’ accounts.
  16. 70 Take a look at the past seminar for more

    details on the lifecycle of Dashboard!
  17. Agenda • Get Data • Clean and Transform Data •

    Visualize Data • Create, Publish, Share, and Schedule Dashboard • Parameterize Dashboard • Forecast Leads • Predict Win 74
  18. 75 Me Team Mates I want data for Europe. I

    want data for Africa. I want data for Asia. Exploratory Server Everybody has a different interest. Publish
  19. 76 By parameterizing the Dashboard, you can let them choose

    which part of the data they want to see.
  20. You can create Filters in the data import dialog so

    that you can limit the data you query from Salesforce. 78
  21. 80 You can create Filters on top of the imported

    data as data wrangling steps.
  22. You can create Filters inside the Chart. This Chart Filter

    works only for the particular chart. 82
  23. 85 All the parameters that are required to generate the

    chart data will automatically show up in the pane.
  24. For the Dashboard, all the parameters that are required to

    generate data for all the charts inside the Dashboard are automatically added to the Parameter pane. 86
  25. For the published Dashboard, you can open the Parameter pane

    and turn on the interactive mode to start using the parameters. 87
  26. 89 1. Data Source 2. Create Calculation 3. Filter Created

    data wrangling steps and charts. Chart 2 Chart 1
  27. 90 Closed Date 1. Data Source 2. Create Calculation 3.

    Filter Created Filters. Probability Type Chart 2 Chart 1
  28. 91 Closed Date 1. Data Source 2. Create Calculation 3.

    Filter Created Dashboard with Chart 2. Dashboard 1 Probability Type Chart 2 Chart 1
  29. 92 Closed Date 1. Data Source 2. Create Calculation 3.

    Filter Dashboard 1 Probability When you open the Parameter pane you will see‘Closed Date’ and ‘Probability’ parameters automatically.
  30. 93 Closed Date 1. Data Source 2. Create Calculation 3.

    Filter Now, we added Chart 1 to the Dashboard. Dashboard 1 Probability Type
  31. Dashboard 1 Type 94 Closed Date 1. Data Source 2.

    Create Calculation 3. Filter When you open the Parameter pane, the ‘Stage’ parameter is automatically added. Probability
  32. Dashboard 1 Type 95 Closed Date 1. Data Source 2.

    Create Calculation 3. Filter Probability If we change the Probability parameter only the Chart 2 gets updated.
  33. Dashboard 1 Type 96 Closed Date 1. Data Source 2.

    Create Calculation 3. Filter Probability If we change the Stage parameter only the Chart 1 gets updated.
  34. Dashboard 1 Type 97 Closed Date 1. Data Source 2.

    Create Calculation Probability If we change the Closed Date parameter all the steps and the charts will be updated. 3. Filter
  35. 99

  36. 100

  37. 101

  38. 102

  39. Agenda • Get Data • Clean and Transform Data •

    Visualize Data • Create, Publish, Share, and Schedule Dashboard • Parameterize Dashboard • Forecast Leads • Predict Win 103
  40. 104 Can we forecast how the number of leads will

    look like in the next 12 months?
  41. 105 Want to forecast the number of leads in the

    next 12 months. Actual Forecasted
  42. • Time interval between data has to be same throughout

    the data • Day with NA is not allowed • Seasonality with multiple periods (Week and Year) is hard to handle • Parameter tuning is hard and requires a forecasting expert level knowledge. Problems with Traditional Time Series Model
  43. • A ‘curve fitting’ algorithm to build time series forcasting

    models. • Designed for ease of use without expert knowledge on time series forecasting or statistics. • Built by Data Scientists (Sean J. Taylor & co.) at Facebook and open sourced. (https:// facebook.github.io/prophet) Prophet Sean J. Taylor @seanjtaylor
  44. Build a model by finding a best smooth line which

    can be represented as sum of the following components. • Overall growth trend • Seasonality - Yearly, Weekly, Daily, etc. • Holiday effects - X’mas, New Year, July 4th, etc. • External Predictors Prophet - Additive Model
  45. 110

  46. 111

  47. 112

  48. Agenda • Get Data • Clean and Transform Data •

    Visualize Data • Create, Publish, Share, and Schedule Dashboard • Parameterize Dashboard • Forecast Leads • Predict Win 113
  49. 115 Algorithm Model Build a Prediction Model. Conversion Age Time

    Country Industry TRUE 60 120 Japan Ad FALSE 45 55 US Medical FALSE 52 20 US Media TRUE 48 140 Japan Ad TRUE 53 80 UK Bank FALSE 35 30 Japan Media
  50. 116 Predict Conversion Age Time Country Industry TRUE 25 120

    Japan Ad FALSE 23 55 US Media FALSE 40 150 US Ad Conversion Age Time Country Industry ? 25 120 Japan Ad ? 23 55 US Media ? 40 150 US Ad Algorithm Model Conversion Age Time Country Industry TRUE 60 120 Japan Ad FALSE 45 55 US Medical FALSE 52 20 US Media TRUE 48 140 Japan Ad TRUE 53 80 UK Bank FALSE 35 30 Japan Media
  51. 117 A model is a definition of a pattern the

    algorithm has captured in the data. Algorithm Model Conversion Age Time Country Industry TRUE 60 120 Japan Ad FALSE 45 55 US Medical FALSE 52 20 US Media TRUE 48 140 Japan Ad TRUE 53 80 UK Bank FALSE 35 30 Japan Media
  52. 118 Not just for the prediction, we can also use

    it to learn a lot about the patterns in data. Insight • Which variables have stronger relationship with the target variable. • How are they related? • Are they significant? • What is the quality if we used this model to predict? Algorithm Model
  53. In order to use the Correlation Mode and build a

    prediction model, we need to convert the StageName variable to Logical (True / False) data type. 122
  54. 124

  55. It shows the relationship between the Is_Win and all other

    variables with the charts along with metrics.
  56. Sort the columns based on the AUC values, which indicate

    the strength of the relationship. 129
  57. Numeric TRUE/FALSE TRUE/FALSE + Time Linear Regression Random Forest /

    XGBoost Statistical Learning Machine Learning Logistic Regression Cox Regression Survival Forest 132 Regression Model Classification Model Survival Model Data Type Statistical Learning Machine Learning Statistical Learning Machine Learning Random Forest / XGBoost
  58. Under the ‘Prediction’ tab, you can see how the probability

    of winning changes as the values in a given variable change. 136
  59. 137 Under the Importance tab, you can see which variables

    are more important to predict the probability of winning.
  60. 139 Under the Data tab, you can see the predicted

    result for all the existing accounts.
  61. But, we want to predict the probability of winning for

    the future opportunities, not the past opportunities.
  62. 144