Predict which leads have the most propensity to convert.

We will train a machine learning (ML) model based on historical lead conversion data, in order to predict likelihood of conversion for new leads.

Determining the value of each lead is a critical factor for businesses that want to maximize their sales efforts. It helps them identify which leads are most likely to convert and generate revenue over time.

Here is an example on how to build an ML model in Peliqan.io with a few lines of Python code.

Import required modules

from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score
from sklearn.preprocessing import LabelEncoder
from collections import defaultdict
from joblib import dump
import pandas as pd

Load a dataset

Load data from a table into a dataframe (df). The table needs to contains lead data, including an indication if these leads converted (historical data).

# Load Data
df = pq.load_table('leads', df=True)

Using Streamlit to build an app

We use the Streamlit module (st), built into Peliqan.io, to build a UI and show data.

# Show a title (st = Streamlit module)
st.title("Lead Conversion")

# Show some text
st.text("Predict Hot Leads")

# Show the dataframe
st.dataframe(df.head(), use_container_width=True)

This is what the output looks like:

Here’s our code in the Peliqan low-code editor, with a preview:

Prepare the data

We remove unwanted columns (features) from our leads and convert categories (e.g. lead source) to a numerical value:

# Drop unwanted features
drop_features = ['Prospect ID', 'index', 'Prediction'] 
df = df.drop(drop_features, axis=1)

encoder = defaultdict(LabelEncoder)

# Apply Label Encoding to convert categorical variables to Numerical
cat_cols = df.select_dtypes('object').columns
df[cat_cols] = df[cat_cols].apply(lambda x: encoder[x.name].fit_transform(x))

# Save the label encoder for future predictions
dump(encoder, '/data_app/encoder_lead_conversion')

Train and save the model

Once the data is ready we split it into a training set and a testing sets to evaluate the model. We save the model to make more predictions later on.

# Training data to train model on
X = df.drop('Converted', axis=1)

# Feature to predict
Y = df['Converted']

# Split data into train test and train model with .fit()
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=.30, random_state=0)
model = LogisticRegression().fit(X_train,Y_train)

# Make prediction
pred = model.predict(X_test)

# Evaluate the model
accuracy = accuracy_score(pred, Y_test)
st.text("Accuracy: " + str(accuracy))

# Save the model for future real-time predictions
dump(model, '/data_app/model_lead_conversion')
st.success('Model saved successfully!')

‣

Expand this to see the full code

from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score
from sklearn.preprocessing import LabelEncoder
from collections import defaultdict
from joblib import dump
import pandas as pd

# Load Data
df = pq.load_table('leads')

# Show a title (st = Streamlit module)
st.title("Lead Conversion - predict hot leads")

# Show some text
st.text("Predict Hot Leads")

# Show the dataframe
st.dataframe(df.head(), use_container_width=True)

# Drop unwanted features
drop_features = ['Prospect ID', 'index', 'Prediction'] 
df = df.drop(drop_features, axis=1)

encoder = defaultdict(LabelEncoder)

# Apply Label Encoding to convert categorical variables to Numerical
cat_cols = df.select_dtypes('object').columns
df[cat_cols] = df[cat_cols].apply(lambda x: encoder[x.name].fit_transform(x))

# Save the label encoder for future predictions
dump(encoder, '/data_app/encoder_lead_conversion')

# Training data to train model on
X = df.drop('Converted', axis=1)

# Feature to predict
Y = df['Converted']

# Split data into train test and train model with .fit()
X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=.30, random_state=0)
model = LogisticRegression().fit(X_train,Y_train)

# Make prediction
pred = model.predict(X_test)

# Evaluate model
accuracy = accuracy_score(pred, Y_test)
st.text("Accuracy: " + str(accuracy))

# Save the model for future real-time predictions
dump(model, '/data_app/model_lead_conversion')
st.success('Model saved successfully')

Next Steps

You can make predictions on real-time incoming data using the saved model. Learn more about making real-time predictions on new incoming data.
You can make real-time predictions on new incoming data and send alerts to slack if the model makes a prediction above a certain threshold.
Using Peliqan you can create an app for business users to consume the model you have made. Learn more about creating apps for users to consume your ML model.