crm_analitics_with_rfm/crm_analitics_with_rfm.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a6b0dafd",
   "metadata": {},
   "source": [
    "# Customer Relationship Management (CRM)\n",
    "\n",
    "## Customer Segmentation with RFM\n",
    "___"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "467ca4b8",
   "metadata": {},
   "source": [
    "**Business Problem: There is an e-commerce company and this company want to segments it's customers for develop marketing strategy** \n",
    "___\n",
    "\n",
    "__Variables__\n",
    "\n",
    "Invoice: Unique number for each invoice. If invoice number starts with 'C', it's mean this invoice cancelled.\n",
    "\n",
    "StockCode: Unique number for each product.\n",
    "\n",
    "Description: Prooduct name.\n",
    "\n",
    "Quantity: Indicates how many units of the product were sold.\n",
    "\n",
    "InvoiceDate: Invoice date.\n",
    "\n",
    "Price: Product price for each units. (Sterling)\n",
    "\n",
    "CustumerID: Unique ID for each customer.\n",
    "\n",
    "Country: Customer country."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef39a684",
   "metadata": {},
   "source": [
    "**Import Libraries**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "b8677563",
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime as dt\n",
    "import pandas as pd\n",
    "\n",
    "pd.set_option('display.float_format', lambda x: '%.3f' %x)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6997c4c",
   "metadata": {},
   "source": [
    "**Read Excel File**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "4f04d8fc",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_ = pd.read_excel('crm_analytics/datasets/online_retail_II.xlsx', sheet_name='Year 2009-2010')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "id": "b362c426",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df_.copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d7f9f1cc",
   "metadata": {},
   "source": [
    "## Understanding Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b561c840",
   "metadata": {},
   "source": [
    "**First 15 rows**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "289e96e6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Invoice</th>\n",
       "      <th>StockCode</th>\n",
       "      <th>Description</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>InvoiceDate</th>\n",
       "      <th>Price</th>\n",
       "      <th>Customer ID</th>\n",
       "      <th>Country</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>489434</td>\n",
       "      <td>85048</td>\n",
       "      <td>15CM CHRISTMAS GLASS BALL 20 LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.950</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>489434</td>\n",
       "      <td>79323P</td>\n",
       "      <td>PINK CHERRY LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.750</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>489434</td>\n",
       "      <td>79323W</td>\n",
       "      <td>WHITE CHERRY LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.750</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>489434</td>\n",
       "      <td>22041</td>\n",
       "      <td>RECORD FRAME 7\" SINGLE SIZE</td>\n",
       "      <td>48</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>2.100</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>489434</td>\n",
       "      <td>21232</td>\n",
       "      <td>STRAWBERRY CERAMIC TRINKET BOX</td>\n",
       "      <td>24</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>1.250</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>489434</td>\n",
       "      <td>22064</td>\n",
       "      <td>PINK DOUGHNUT TRINKET POT</td>\n",
       "      <td>24</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>1.650</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>489434</td>\n",
       "      <td>21871</td>\n",
       "      <td>SAVE THE PLANET MUG</td>\n",
       "      <td>24</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>1.250</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>489434</td>\n",
       "      <td>21523</td>\n",
       "      <td>FANCY FONT HOME SWEET HOME DOORMAT</td>\n",
       "      <td>10</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>5.950</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>489435</td>\n",
       "      <td>22350</td>\n",
       "      <td>CAT BOWL</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:46:00</td>\n",
       "      <td>2.550</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>489435</td>\n",
       "      <td>22349</td>\n",
       "      <td>DOG BOWL , CHASING BALL DESIGN</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:46:00</td>\n",
       "      <td>3.750</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>489435</td>\n",
       "      <td>22195</td>\n",
       "      <td>HEART MEASURING SPOONS LARGE</td>\n",
       "      <td>24</td>\n",
       "      <td>2009-12-01 07:46:00</td>\n",
       "      <td>1.650</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>489435</td>\n",
       "      <td>22353</td>\n",
       "      <td>LUNCHBOX WITH CUTLERY FAIRY CAKES</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:46:00</td>\n",
       "      <td>2.550</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>489436</td>\n",
       "      <td>48173C</td>\n",
       "      <td>DOOR MAT BLACK FLOCK</td>\n",
       "      <td>10</td>\n",
       "      <td>2009-12-01 09:06:00</td>\n",
       "      <td>5.950</td>\n",
       "      <td>13078.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>489436</td>\n",
       "      <td>21755</td>\n",
       "      <td>LOVE BUILDING BLOCK WORD</td>\n",
       "      <td>18</td>\n",
       "      <td>2009-12-01 09:06:00</td>\n",
       "      <td>5.450</td>\n",
       "      <td>13078.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>489436</td>\n",
       "      <td>21754</td>\n",
       "      <td>HOME BUILDING BLOCK WORD</td>\n",
       "      <td>3</td>\n",
       "      <td>2009-12-01 09:06:00</td>\n",
       "      <td>5.950</td>\n",
       "      <td>13078.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Invoice StockCode                          Description  Quantity  \\\n",
       "0   489434     85048  15CM CHRISTMAS GLASS BALL 20 LIGHTS        12   \n",
       "1   489434    79323P                   PINK CHERRY LIGHTS        12   \n",
       "2   489434    79323W                  WHITE CHERRY LIGHTS        12   \n",
       "3   489434     22041         RECORD FRAME 7\" SINGLE SIZE         48   \n",
       "4   489434     21232       STRAWBERRY CERAMIC TRINKET BOX        24   \n",
       "5   489434     22064           PINK DOUGHNUT TRINKET POT         24   \n",
       "6   489434     21871                  SAVE THE PLANET MUG        24   \n",
       "7   489434     21523   FANCY FONT HOME SWEET HOME DOORMAT        10   \n",
       "8   489435     22350                            CAT BOWL         12   \n",
       "9   489435     22349       DOG BOWL , CHASING BALL DESIGN        12   \n",
       "10  489435     22195         HEART MEASURING SPOONS LARGE        24   \n",
       "11  489435     22353   LUNCHBOX WITH CUTLERY FAIRY CAKES         12   \n",
       "12  489436    48173C                DOOR MAT BLACK FLOCK         10   \n",
       "13  489436     21755             LOVE BUILDING BLOCK WORD        18   \n",
       "14  489436     21754             HOME BUILDING BLOCK WORD         3   \n",
       "\n",
       "           InvoiceDate  Price  Customer ID         Country  \n",
       "0  2009-12-01 07:45:00  6.950    13085.000  United Kingdom  \n",
       "1  2009-12-01 07:45:00  6.750    13085.000  United Kingdom  \n",
       "2  2009-12-01 07:45:00  6.750    13085.000  United Kingdom  \n",
       "3  2009-12-01 07:45:00  2.100    13085.000  United Kingdom  \n",
       "4  2009-12-01 07:45:00  1.250    13085.000  United Kingdom  \n",
       "5  2009-12-01 07:45:00  1.650    13085.000  United Kingdom  \n",
       "6  2009-12-01 07:45:00  1.250    13085.000  United Kingdom  \n",
       "7  2009-12-01 07:45:00  5.950    13085.000  United Kingdom  \n",
       "8  2009-12-01 07:46:00  2.550    13085.000  United Kingdom  \n",
       "9  2009-12-01 07:46:00  3.750    13085.000  United Kingdom  \n",
       "10 2009-12-01 07:46:00  1.650    13085.000  United Kingdom  \n",
       "11 2009-12-01 07:46:00  2.550    13085.000  United Kingdom  \n",
       "12 2009-12-01 09:06:00  5.950    13078.000  United Kingdom  \n",
       "13 2009-12-01 09:06:00  5.450    13078.000  United Kingdom  \n",
       "14 2009-12-01 09:06:00  5.950    13078.000  United Kingdom  "
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4bd187c",
   "metadata": {},
   "source": [
    "**Check units and columns**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "d3209586",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(525461, 8)"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7f238e6",
   "metadata": {},
   "source": [
    "**Check null units for each columns**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "0ab5cc26",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Invoice             0\n",
       "StockCode           0\n",
       "Description      2928\n",
       "Quantity            0\n",
       "InvoiceDate         0\n",
       "Price               0\n",
       "Customer ID    107927\n",
       "Country             0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba9f86fd",
   "metadata": {},
   "source": [
    "**Check the unique values for each variables**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "dca25824",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Invoice        28816\n",
       "StockCode       4632\n",
       "Description     4681\n",
       "Quantity         825\n",
       "InvoiceDate    25296\n",
       "Price           1606\n",
       "Customer ID     4383\n",
       "Country           40\n",
       "dtype: int64"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4e6ade7",
   "metadata": {},
   "source": [
    "**Add 'TotalPrice' to dataframe.**\n",
    "\n",
    "TotalPrice = Quantity x UnitPrice"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "916c09d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['TotalPrice'] = df['Quantity'] * df['Price']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "2236a280",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Invoice</th>\n",
       "      <th>StockCode</th>\n",
       "      <th>Description</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>InvoiceDate</th>\n",
       "      <th>Price</th>\n",
       "      <th>Customer ID</th>\n",
       "      <th>Country</th>\n",
       "      <th>TotalPrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>489434</td>\n",
       "      <td>85048</td>\n",
       "      <td>15CM CHRISTMAS GLASS BALL 20 LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.950</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>83.400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>489434</td>\n",
       "      <td>79323P</td>\n",
       "      <td>PINK CHERRY LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.750</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>81.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>489434</td>\n",
       "      <td>79323W</td>\n",
       "      <td>WHITE CHERRY LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.750</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>81.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>489434</td>\n",
       "      <td>22041</td>\n",
       "      <td>RECORD FRAME 7\" SINGLE SIZE</td>\n",
       "      <td>48</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>2.100</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>100.800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>489434</td>\n",
       "      <td>21232</td>\n",
       "      <td>STRAWBERRY CERAMIC TRINKET BOX</td>\n",
       "      <td>24</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>1.250</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>30.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Invoice StockCode                          Description  Quantity  \\\n",
       "0  489434     85048  15CM CHRISTMAS GLASS BALL 20 LIGHTS        12   \n",
       "1  489434    79323P                   PINK CHERRY LIGHTS        12   \n",
       "2  489434    79323W                  WHITE CHERRY LIGHTS        12   \n",
       "3  489434     22041         RECORD FRAME 7\" SINGLE SIZE         48   \n",
       "4  489434     21232       STRAWBERRY CERAMIC TRINKET BOX        24   \n",
       "\n",
       "          InvoiceDate  Price  Customer ID         Country  TotalPrice  \n",
       "0 2009-12-01 07:45:00  6.950    13085.000  United Kingdom      83.400  \n",
       "1 2009-12-01 07:45:00  6.750    13085.000  United Kingdom      81.000  \n",
       "2 2009-12-01 07:45:00  6.750    13085.000  United Kingdom      81.000  \n",
       "3 2009-12-01 07:45:00  2.100    13085.000  United Kingdom     100.800  \n",
       "4 2009-12-01 07:45:00  1.250    13085.000  United Kingdom      30.000  "
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60b6219e",
   "metadata": {},
   "source": [
    "## Data Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3d46b61",
   "metadata": {},
   "source": [
    "**Remove null units from dataframe**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "3099345b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(525461, 9)"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "cf297ccc",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.dropna(inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "e832c7d0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(417534, 9)"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d3aae43",
   "metadata": {},
   "source": [
    "**Deleting cancalled invoice**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "0b87bc12",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[~df['Invoice'].str.contains('C', na=False)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "6a5ac769",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(407695, 9)"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "83f7a8df",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Invoice</th>\n",
       "      <th>StockCode</th>\n",
       "      <th>Description</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>InvoiceDate</th>\n",
       "      <th>Price</th>\n",
       "      <th>Customer ID</th>\n",
       "      <th>Country</th>\n",
       "      <th>TotalPrice</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>489434</td>\n",
       "      <td>85048</td>\n",
       "      <td>15CM CHRISTMAS GLASS BALL 20 LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.950</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>83.400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>489434</td>\n",
       "      <td>79323P</td>\n",
       "      <td>PINK CHERRY LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.750</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>81.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>489434</td>\n",
       "      <td>79323W</td>\n",
       "      <td>WHITE CHERRY LIGHTS</td>\n",
       "      <td>12</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>6.750</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>81.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>489434</td>\n",
       "      <td>22041</td>\n",
       "      <td>RECORD FRAME 7\" SINGLE SIZE</td>\n",
       "      <td>48</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>2.100</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>100.800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>489434</td>\n",
       "      <td>21232</td>\n",
       "      <td>STRAWBERRY CERAMIC TRINKET BOX</td>\n",
       "      <td>24</td>\n",
       "      <td>2009-12-01 07:45:00</td>\n",
       "      <td>1.250</td>\n",
       "      <td>13085.000</td>\n",
       "      <td>United Kingdom</td>\n",
       "      <td>30.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Invoice StockCode                          Description  Quantity  \\\n",
       "0  489434     85048  15CM CHRISTMAS GLASS BALL 20 LIGHTS        12   \n",
       "1  489434    79323P                   PINK CHERRY LIGHTS        12   \n",
       "2  489434    79323W                  WHITE CHERRY LIGHTS        12   \n",
       "3  489434     22041         RECORD FRAME 7\" SINGLE SIZE         48   \n",
       "4  489434     21232       STRAWBERRY CERAMIC TRINKET BOX        24   \n",
       "\n",
       "          InvoiceDate  Price  Customer ID         Country  TotalPrice  \n",
       "0 2009-12-01 07:45:00  6.950    13085.000  United Kingdom      83.400  \n",
       "1 2009-12-01 07:45:00  6.750    13085.000  United Kingdom      81.000  \n",
       "2 2009-12-01 07:45:00  6.750    13085.000  United Kingdom      81.000  \n",
       "3 2009-12-01 07:45:00  2.100    13085.000  United Kingdom     100.800  \n",
       "4 2009-12-01 07:45:00  1.250    13085.000  United Kingdom      30.000  "
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f15640d3",
   "metadata": {},
   "source": [
    "## Calculating RMF Metrics"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dab4af41",
   "metadata": {},
   "source": [
    "**RFM Metrics**\n",
    "\n",
    "Recency: how long ago they made a purchase. \n",
    "\n",
    "Frequency: how often they make purchases.\n",
    "\n",
    "Monetary: how much money they spend."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 210,
   "id": "6da73c08",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Timestamp('2010-12-09 20:01:00')"
      ]
     },
     "execution_count": 210,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['InvoiceDate'].max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "fc6af309",
   "metadata": {},
   "outputs": [],
   "source": [
    "today_date = dt.datetime(2010, 12, 11)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "id": "dd2083f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfm = df.groupby('Customer ID').agg({'InvoiceDate': lambda x: (today_date - x.max()).days,\n",
    "                              'Invoice': lambda x: x.nunique(),\n",
    "                              'TotalPrice': lambda x: x.sum()})\n",
    "\n",
    "rfm.columns = ['recency', 'frequency', 'monetary']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "id": "fac6af8c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>recency</th>\n",
       "      <th>frequency</th>\n",
       "      <th>monetary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12346.000</th>\n",
       "      <td>165</td>\n",
       "      <td>11</td>\n",
       "      <td>372.860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12347.000</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1323.320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.000</th>\n",
       "      <td>74</td>\n",
       "      <td>1</td>\n",
       "      <td>222.160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.000</th>\n",
       "      <td>43</td>\n",
       "      <td>3</td>\n",
       "      <td>2671.140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12351.000</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>300.930</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             recency  frequency  monetary\n",
       "Customer ID                              \n",
       "12346.000        165         11   372.860\n",
       "12347.000          3          2  1323.320\n",
       "12348.000         74          1   222.160\n",
       "12349.000         43          3  2671.140\n",
       "12351.000         11          1   300.930"
      ]
     },
     "execution_count": 151,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "912c1391",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4314, 3)"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "id": "f7a4c897",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>recency</th>\n",
       "      <td>4314.000</td>\n",
       "      <td>91.270</td>\n",
       "      <td>96.944</td>\n",
       "      <td>1.000</td>\n",
       "      <td>18.000</td>\n",
       "      <td>53.000</td>\n",
       "      <td>136.000</td>\n",
       "      <td>374.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>frequency</th>\n",
       "      <td>4314.000</td>\n",
       "      <td>4.454</td>\n",
       "      <td>8.169</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>5.000</td>\n",
       "      <td>205.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>monetary</th>\n",
       "      <td>4314.000</td>\n",
       "      <td>2047.289</td>\n",
       "      <td>8912.523</td>\n",
       "      <td>0.000</td>\n",
       "      <td>307.950</td>\n",
       "      <td>705.550</td>\n",
       "      <td>1722.802</td>\n",
       "      <td>349164.350</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             count     mean      std   min     25%     50%      75%        max\n",
       "recency   4314.000   91.270   96.944 1.000  18.000  53.000  136.000    374.000\n",
       "frequency 4314.000    4.454    8.169 1.000   1.000   2.000    5.000    205.000\n",
       "monetary  4314.000 2047.289 8912.523 0.000 307.950 705.550 1722.802 349164.350"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.describe().T"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68a98d17",
   "metadata": {},
   "source": [
    "**Deleting monetary values equal to zero from rfm dataframe.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "id": "a93b7af1",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfm = rfm[rfm['monetary'] > 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "id": "cf5b158b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(4312, 3)"
      ]
     },
     "execution_count": 155,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "id": "17c97fe7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>recency</th>\n",
       "      <th>frequency</th>\n",
       "      <th>monetary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>4312.000</td>\n",
       "      <td>4312.000</td>\n",
       "      <td>4312.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>91.173</td>\n",
       "      <td>4.456</td>\n",
       "      <td>2048.238</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>96.861</td>\n",
       "      <td>8.170</td>\n",
       "      <td>8914.481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>2.950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>18.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>307.988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>53.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>706.020</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>136.000</td>\n",
       "      <td>5.000</td>\n",
       "      <td>1723.142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>374.000</td>\n",
       "      <td>205.000</td>\n",
       "      <td>349164.350</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       recency  frequency   monetary\n",
       "count 4312.000   4312.000   4312.000\n",
       "mean    91.173      4.456   2048.238\n",
       "std     96.861      8.170   8914.481\n",
       "min      1.000      1.000      2.950\n",
       "25%     18.000      1.000    307.988\n",
       "50%     53.000      2.000    706.020\n",
       "75%    136.000      5.000   1723.142\n",
       "max    374.000    205.000 349164.350"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "445f9b2b",
   "metadata": {},
   "source": [
    "## Calculating RMF Scores"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "id": "522dd20b",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfm['recency_score'] = pd.qcut(rfm['recency'], 5, labels=[5, 4, 3, 2, 1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "id": "7a9aef0f",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfm['frequency_score'] = pd.qcut(rfm['frequency'].rank(method='first'), 5, labels=[1, 2, 3, 4, 5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "id": "d2875b80",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfm['monetary_score'] = pd.qcut(rfm['monetary'], 5, labels=[1, 2, 3, 4, 5])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "id": "d79f3df5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>recency</th>\n",
       "      <th>frequency</th>\n",
       "      <th>monetary</th>\n",
       "      <th>recency_score</th>\n",
       "      <th>frequency_score</th>\n",
       "      <th>monetary_score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12346.000</th>\n",
       "      <td>165</td>\n",
       "      <td>11</td>\n",
       "      <td>372.860</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12347.000</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1323.320</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.000</th>\n",
       "      <td>74</td>\n",
       "      <td>1</td>\n",
       "      <td>222.160</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.000</th>\n",
       "      <td>43</td>\n",
       "      <td>3</td>\n",
       "      <td>2671.140</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12351.000</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>300.930</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             recency  frequency  monetary recency_score frequency_score  \\\n",
       "Customer ID                                                               \n",
       "12346.000        165         11   372.860             2               5   \n",
       "12347.000          3          2  1323.320             5               2   \n",
       "12348.000         74          1   222.160             2               1   \n",
       "12349.000         43          3  2671.140             3               3   \n",
       "12351.000         11          1   300.930             5               1   \n",
       "\n",
       "            monetary_score  \n",
       "Customer ID                 \n",
       "12346.000                2  \n",
       "12347.000                4  \n",
       "12348.000                1  \n",
       "12349.000                5  \n",
       "12351.000                2  "
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 190,
   "id": "6d87def6",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfm['RFM Score'] = rfm['recency_score'].astype(str) + rfm['frequency_score'].astype(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "id": "238facde",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>recency</th>\n",
       "      <th>frequency</th>\n",
       "      <th>monetary</th>\n",
       "      <th>recency_score</th>\n",
       "      <th>frequency_score</th>\n",
       "      <th>monetary_score</th>\n",
       "      <th>RFM Score</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12346.000</th>\n",
       "      <td>165</td>\n",
       "      <td>11</td>\n",
       "      <td>372.860</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12347.000</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1323.320</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.000</th>\n",
       "      <td>74</td>\n",
       "      <td>1</td>\n",
       "      <td>222.160</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.000</th>\n",
       "      <td>43</td>\n",
       "      <td>3</td>\n",
       "      <td>2671.140</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12351.000</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>300.930</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>51</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             recency  frequency  monetary recency_score frequency_score  \\\n",
       "Customer ID                                                               \n",
       "12346.000        165         11   372.860             2               5   \n",
       "12347.000          3          2  1323.320             5               2   \n",
       "12348.000         74          1   222.160             2               1   \n",
       "12349.000         43          3  2671.140             3               3   \n",
       "12351.000         11          1   300.930             5               1   \n",
       "\n",
       "            monetary_score RFM Score  \n",
       "Customer ID                           \n",
       "12346.000                2        25  \n",
       "12347.000                4        52  \n",
       "12348.000                1        21  \n",
       "12349.000                5        33  \n",
       "12351.000                2        51  "
      ]
     },
     "execution_count": 191,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bd501206",
   "metadata": {},
   "source": [
    "## Creating and Analysing RFM Segments"
   ]
  },
  {
   "attachments": {
    "image-2.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "id": "6bbe9fef",
   "metadata": {},
   "source": [
    "![image-2.png](attachment:image-2.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 203,
   "id": "51e57e9f",
   "metadata": {},
   "outputs": [],
   "source": [
    "seg_map = {r'[1-2][1-2]': 'hibernating',\n",
    "          r'[1-2][3-4]': 'at risk',\n",
    "          r'[1-2]5': 'cant lose them',\n",
    "          r'3[1-2]': 'about to sleep',\n",
    "          r'33': 'need attention',\n",
    "          r'[3-4][4-5]': 'loyal customers',\n",
    "          r'41': 'promising',\n",
    "          r'[4-5][2-3]': 'potential loyallists',\n",
    "          r'51': 'new_customers',\n",
    "          r'5[4-5]': 'champions'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 204,
   "id": "d6516a15",
   "metadata": {},
   "outputs": [],
   "source": [
    "rfm['segments'] = rfm['RFM Score'].replace(seg_map, regex=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "id": "e333ba4d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>recency</th>\n",
       "      <th>frequency</th>\n",
       "      <th>monetary</th>\n",
       "      <th>recency_score</th>\n",
       "      <th>frequency_score</th>\n",
       "      <th>monetary_score</th>\n",
       "      <th>RFM Score</th>\n",
       "      <th>segments</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12346.000</th>\n",
       "      <td>165</td>\n",
       "      <td>11</td>\n",
       "      <td>372.860</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>25</td>\n",
       "      <td>cant lose them</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12347.000</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1323.320</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>52</td>\n",
       "      <td>potential loyallists</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.000</th>\n",
       "      <td>74</td>\n",
       "      <td>1</td>\n",
       "      <td>222.160</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "      <td>hibernating</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.000</th>\n",
       "      <td>43</td>\n",
       "      <td>3</td>\n",
       "      <td>2671.140</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>33</td>\n",
       "      <td>need attention</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12351.000</th>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>300.930</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>51</td>\n",
       "      <td>new_customers</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             recency  frequency  monetary recency_score frequency_score  \\\n",
       "Customer ID                                                               \n",
       "12346.000        165         11   372.860             2               5   \n",
       "12347.000          3          2  1323.320             5               2   \n",
       "12348.000         74          1   222.160             2               1   \n",
       "12349.000         43          3  2671.140             3               3   \n",
       "12351.000         11          1   300.930             5               1   \n",
       "\n",
       "            monetary_score RFM Score              segments  \n",
       "Customer ID                                                 \n",
       "12346.000                2        25        cant lose them  \n",
       "12347.000                4        52  potential loyallists  \n",
       "12348.000                1        21           hibernating  \n",
       "12349.000                5        33        need attention  \n",
       "12351.000                2        51         new_customers  "
      ]
     },
     "execution_count": 205,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 220,
   "id": "dcc34e94",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"8\" halign=\"left\">recency</th>\n",
       "      <th colspan=\"5\" halign=\"left\">frequency</th>\n",
       "      <th colspan=\"8\" halign=\"left\">monetary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>...</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>segments</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>about to sleep</th>\n",
       "      <td>343.000</td>\n",
       "      <td>53.819</td>\n",
       "      <td>10.286</td>\n",
       "      <td>37.000</td>\n",
       "      <td>45.000</td>\n",
       "      <td>53.000</td>\n",
       "      <td>64.000</td>\n",
       "      <td>71.000</td>\n",
       "      <td>343.000</td>\n",
       "      <td>1.201</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>343.000</td>\n",
       "      <td>441.320</td>\n",
       "      <td>416.673</td>\n",
       "      <td>24.400</td>\n",
       "      <td>194.125</td>\n",
       "      <td>317.760</td>\n",
       "      <td>535.275</td>\n",
       "      <td>3502.480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>at risk</th>\n",
       "      <td>611.000</td>\n",
       "      <td>152.159</td>\n",
       "      <td>69.940</td>\n",
       "      <td>72.000</td>\n",
       "      <td>92.500</td>\n",
       "      <td>131.000</td>\n",
       "      <td>198.000</td>\n",
       "      <td>372.000</td>\n",
       "      <td>611.000</td>\n",
       "      <td>3.074</td>\n",
       "      <td>...</td>\n",
       "      <td>4.000</td>\n",
       "      <td>6.000</td>\n",
       "      <td>611.000</td>\n",
       "      <td>1188.878</td>\n",
       "      <td>1844.624</td>\n",
       "      <td>24.050</td>\n",
       "      <td>449.245</td>\n",
       "      <td>760.190</td>\n",
       "      <td>1376.575</td>\n",
       "      <td>34095.260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cant lose them</th>\n",
       "      <td>77.000</td>\n",
       "      <td>124.117</td>\n",
       "      <td>49.925</td>\n",
       "      <td>72.000</td>\n",
       "      <td>88.000</td>\n",
       "      <td>109.000</td>\n",
       "      <td>137.000</td>\n",
       "      <td>298.000</td>\n",
       "      <td>77.000</td>\n",
       "      <td>9.117</td>\n",
       "      <td>...</td>\n",
       "      <td>9.000</td>\n",
       "      <td>46.000</td>\n",
       "      <td>77.000</td>\n",
       "      <td>4099.450</td>\n",
       "      <td>5304.769</td>\n",
       "      <td>181.350</td>\n",
       "      <td>1698.560</td>\n",
       "      <td>2322.590</td>\n",
       "      <td>4009.660</td>\n",
       "      <td>26286.750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>champions</th>\n",
       "      <td>663.000</td>\n",
       "      <td>7.119</td>\n",
       "      <td>4.620</td>\n",
       "      <td>1.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>7.000</td>\n",
       "      <td>11.000</td>\n",
       "      <td>15.000</td>\n",
       "      <td>663.000</td>\n",
       "      <td>12.554</td>\n",
       "      <td>...</td>\n",
       "      <td>13.000</td>\n",
       "      <td>205.000</td>\n",
       "      <td>663.000</td>\n",
       "      <td>6852.264</td>\n",
       "      <td>21556.377</td>\n",
       "      <td>75.760</td>\n",
       "      <td>1379.675</td>\n",
       "      <td>2508.320</td>\n",
       "      <td>4983.785</td>\n",
       "      <td>349164.350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>hibernating</th>\n",
       "      <td>1015.000</td>\n",
       "      <td>213.886</td>\n",
       "      <td>89.932</td>\n",
       "      <td>72.000</td>\n",
       "      <td>136.000</td>\n",
       "      <td>213.000</td>\n",
       "      <td>284.000</td>\n",
       "      <td>374.000</td>\n",
       "      <td>1015.000</td>\n",
       "      <td>1.126</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>1015.000</td>\n",
       "      <td>403.978</td>\n",
       "      <td>775.890</td>\n",
       "      <td>2.950</td>\n",
       "      <td>148.505</td>\n",
       "      <td>252.480</td>\n",
       "      <td>405.050</td>\n",
       "      <td>11880.840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>loyal customers</th>\n",
       "      <td>742.000</td>\n",
       "      <td>36.287</td>\n",
       "      <td>16.073</td>\n",
       "      <td>16.000</td>\n",
       "      <td>23.000</td>\n",
       "      <td>31.000</td>\n",
       "      <td>50.000</td>\n",
       "      <td>71.000</td>\n",
       "      <td>742.000</td>\n",
       "      <td>6.830</td>\n",
       "      <td>...</td>\n",
       "      <td>8.000</td>\n",
       "      <td>42.000</td>\n",
       "      <td>742.000</td>\n",
       "      <td>2746.067</td>\n",
       "      <td>3256.543</td>\n",
       "      <td>97.400</td>\n",
       "      <td>1072.790</td>\n",
       "      <td>1821.385</td>\n",
       "      <td>3174.050</td>\n",
       "      <td>50291.380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>need attention</th>\n",
       "      <td>207.000</td>\n",
       "      <td>53.266</td>\n",
       "      <td>9.796</td>\n",
       "      <td>37.000</td>\n",
       "      <td>44.000</td>\n",
       "      <td>53.000</td>\n",
       "      <td>60.500</td>\n",
       "      <td>71.000</td>\n",
       "      <td>207.000</td>\n",
       "      <td>2.449</td>\n",
       "      <td>...</td>\n",
       "      <td>3.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>207.000</td>\n",
       "      <td>1060.357</td>\n",
       "      <td>1189.804</td>\n",
       "      <td>101.100</td>\n",
       "      <td>471.470</td>\n",
       "      <td>742.900</td>\n",
       "      <td>1352.330</td>\n",
       "      <td>13544.990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>new_customers</th>\n",
       "      <td>50.000</td>\n",
       "      <td>8.580</td>\n",
       "      <td>4.310</td>\n",
       "      <td>1.000</td>\n",
       "      <td>5.000</td>\n",
       "      <td>8.500</td>\n",
       "      <td>11.000</td>\n",
       "      <td>15.000</td>\n",
       "      <td>50.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>50.000</td>\n",
       "      <td>386.199</td>\n",
       "      <td>493.321</td>\n",
       "      <td>35.400</td>\n",
       "      <td>129.232</td>\n",
       "      <td>258.825</td>\n",
       "      <td>400.688</td>\n",
       "      <td>2945.380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>potential loyallists</th>\n",
       "      <td>517.000</td>\n",
       "      <td>18.793</td>\n",
       "      <td>9.731</td>\n",
       "      <td>1.000</td>\n",
       "      <td>11.000</td>\n",
       "      <td>19.000</td>\n",
       "      <td>26.000</td>\n",
       "      <td>36.000</td>\n",
       "      <td>517.000</td>\n",
       "      <td>2.017</td>\n",
       "      <td>...</td>\n",
       "      <td>3.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>517.000</td>\n",
       "      <td>729.511</td>\n",
       "      <td>836.679</td>\n",
       "      <td>10.950</td>\n",
       "      <td>303.700</td>\n",
       "      <td>524.650</td>\n",
       "      <td>891.020</td>\n",
       "      <td>12079.990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>promising</th>\n",
       "      <td>87.000</td>\n",
       "      <td>25.747</td>\n",
       "      <td>6.035</td>\n",
       "      <td>16.000</td>\n",
       "      <td>21.500</td>\n",
       "      <td>25.000</td>\n",
       "      <td>30.000</td>\n",
       "      <td>36.000</td>\n",
       "      <td>87.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>87.000</td>\n",
       "      <td>367.087</td>\n",
       "      <td>343.886</td>\n",
       "      <td>30.300</td>\n",
       "      <td>194.975</td>\n",
       "      <td>293.740</td>\n",
       "      <td>418.750</td>\n",
       "      <td>2389.620</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 24 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      recency                                                \\\n",
       "                        count    mean    std    min     25%     50%     75%   \n",
       "segments                                                                      \n",
       "about to sleep        343.000  53.819 10.286 37.000  45.000  53.000  64.000   \n",
       "at risk               611.000 152.159 69.940 72.000  92.500 131.000 198.000   \n",
       "cant lose them         77.000 124.117 49.925 72.000  88.000 109.000 137.000   \n",
       "champions             663.000   7.119  4.620  1.000   3.000   7.000  11.000   \n",
       "hibernating          1015.000 213.886 89.932 72.000 136.000 213.000 284.000   \n",
       "loyal customers       742.000  36.287 16.073 16.000  23.000  31.000  50.000   \n",
       "need attention        207.000  53.266  9.796 37.000  44.000  53.000  60.500   \n",
       "new_customers          50.000   8.580  4.310  1.000   5.000   8.500  11.000   \n",
       "potential loyallists  517.000  18.793  9.731  1.000  11.000  19.000  26.000   \n",
       "promising              87.000  25.747  6.035 16.000  21.500  25.000  30.000   \n",
       "\n",
       "                             frequency         ...                monetary  \\\n",
       "                         max     count   mean  ...    75%     max    count   \n",
       "segments                                       ...                           \n",
       "about to sleep        71.000   343.000  1.201  ...  1.000   2.000  343.000   \n",
       "at risk              372.000   611.000  3.074  ...  4.000   6.000  611.000   \n",
       "cant lose them       298.000    77.000  9.117  ...  9.000  46.000   77.000   \n",
       "champions             15.000   663.000 12.554  ... 13.000 205.000  663.000   \n",
       "hibernating          374.000  1015.000  1.126  ...  1.000   2.000 1015.000   \n",
       "loyal customers       71.000   742.000  6.830  ...  8.000  42.000  742.000   \n",
       "need attention        71.000   207.000  2.449  ...  3.000   3.000  207.000   \n",
       "new_customers         15.000    50.000  1.000  ...  1.000   1.000   50.000   \n",
       "potential loyallists  36.000   517.000  2.017  ...  3.000   3.000  517.000   \n",
       "promising             36.000    87.000  1.000  ...  1.000   1.000   87.000   \n",
       "\n",
       "                                                                            \\\n",
       "                         mean       std     min      25%      50%      75%   \n",
       "segments                                                                     \n",
       "about to sleep        441.320   416.673  24.400  194.125  317.760  535.275   \n",
       "at risk              1188.878  1844.624  24.050  449.245  760.190 1376.575   \n",
       "cant lose them       4099.450  5304.769 181.350 1698.560 2322.590 4009.660   \n",
       "champions            6852.264 21556.377  75.760 1379.675 2508.320 4983.785   \n",
       "hibernating           403.978   775.890   2.950  148.505  252.480  405.050   \n",
       "loyal customers      2746.067  3256.543  97.400 1072.790 1821.385 3174.050   \n",
       "need attention       1060.357  1189.804 101.100  471.470  742.900 1352.330   \n",
       "new_customers         386.199   493.321  35.400  129.232  258.825  400.688   \n",
       "potential loyallists  729.511   836.679  10.950  303.700  524.650  891.020   \n",
       "promising             367.087   343.886  30.300  194.975  293.740  418.750   \n",
       "\n",
       "                                 \n",
       "                            max  \n",
       "segments                         \n",
       "about to sleep         3502.480  \n",
       "at risk               34095.260  \n",
       "cant lose them        26286.750  \n",
       "champions            349164.350  \n",
       "hibernating           11880.840  \n",
       "loyal customers       50291.380  \n",
       "need attention        13544.990  \n",
       "new_customers          2945.380  \n",
       "potential loyallists  12079.990  \n",
       "promising              2389.620  \n",
       "\n",
       "[10 rows x 24 columns]"
      ]
     },
     "execution_count": 220,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm.groupby('segments').describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 223,
   "id": "2f1e77de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>segments</th>\n",
       "      <th>recency</th>\n",
       "      <th>frequency</th>\n",
       "      <th>monetary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12346.000</th>\n",
       "      <td>cant lose them</td>\n",
       "      <td>165</td>\n",
       "      <td>11</td>\n",
       "      <td>372.860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12347.000</th>\n",
       "      <td>potential loyallists</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1323.320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348.000</th>\n",
       "      <td>hibernating</td>\n",
       "      <td>74</td>\n",
       "      <td>1</td>\n",
       "      <td>222.160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349.000</th>\n",
       "      <td>need attention</td>\n",
       "      <td>43</td>\n",
       "      <td>3</td>\n",
       "      <td>2671.140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12351.000</th>\n",
       "      <td>new_customers</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>300.930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18283.000</th>\n",
       "      <td>loyal customers</td>\n",
       "      <td>18</td>\n",
       "      <td>6</td>\n",
       "      <td>641.770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18284.000</th>\n",
       "      <td>about to sleep</td>\n",
       "      <td>67</td>\n",
       "      <td>1</td>\n",
       "      <td>461.680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18285.000</th>\n",
       "      <td>hibernating</td>\n",
       "      <td>296</td>\n",
       "      <td>1</td>\n",
       "      <td>427.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18286.000</th>\n",
       "      <td>at risk</td>\n",
       "      <td>112</td>\n",
       "      <td>2</td>\n",
       "      <td>1296.430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18287.000</th>\n",
       "      <td>loyal customers</td>\n",
       "      <td>18</td>\n",
       "      <td>4</td>\n",
       "      <td>2345.710</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4312 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         segments  recency  frequency  monetary\n",
       "Customer ID                                                    \n",
       "12346.000          cant lose them      165         11   372.860\n",
       "12347.000    potential loyallists        3          2  1323.320\n",
       "12348.000             hibernating       74          1   222.160\n",
       "12349.000          need attention       43          3  2671.140\n",
       "12351.000           new_customers       11          1   300.930\n",
       "...                           ...      ...        ...       ...\n",
       "18283.000         loyal customers       18          6   641.770\n",
       "18284.000          about to sleep       67          1   461.680\n",
       "18285.000             hibernating      296          1   427.000\n",
       "18286.000                 at risk      112          2  1296.430\n",
       "18287.000         loyal customers       18          4  2345.710\n",
       "\n",
       "[4312 rows x 4 columns]"
      ]
     },
     "execution_count": 223,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfm = rfm[['segments', 'recency', 'frequency', 'monetary']]\n",
    "rfm"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "66b71d2b",
   "metadata": {},
   "source": [
    "## Functionalization of all steps"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "id": "30c752cf",
   "metadata": {},
   "outputs": [],
   "source": [
    "def create_rfm(dataframe, csv=False):\n",
    "    \n",
    "    # Data preparation      \n",
    "    dataframe['TotalPrice'] = dataframe['Quantity'] * dataframe['Price']\n",
    "    dataframe.dropna(inplace=True)\n",
    "    dataframe = dataframe[~dataframe['Invoice'].str.contains('C', na=False)]\n",
    "    \n",
    "    # Calculating RMF metrics    \n",
    "    max_day = dataframe['InvoiceDate'].max()\n",
    "    today_date = dt.datetime(max_day.year, max_day.month, max_day.day+2)\n",
    "    rfm = dataframe.groupby('Customer ID').agg({'InvoiceDate': lambda x: (today_date - x.max()).days,\n",
    "                              'Invoice': lambda x: x.nunique(),\n",
    "                              'TotalPrice': lambda x: x.sum()})\n",
    "\n",
    "    rfm.columns = ['recency', 'frequency', 'monetary']\n",
    "    rfm = rfm[rfm['monetary'] > 0]\n",
    "    \n",
    "    # Calculating RMF scores      \n",
    "    rfm['recency_score'] = pd.qcut(rfm['recency'], 5, labels=[5, 4, 3, 2, 1])\n",
    "    rfm['frequency_score'] = pd.qcut(rfm['frequency'].rank(method='first'), 5, labels=[1, 2, 3, 4, 5])\n",
    "    rfm['monetary_score'] = pd.qcut(rfm['monetary'], 5, labels=[1, 2, 3, 4, 5])\n",
    "    rfm['RFM Score'] = rfm['recency_score'].astype(str) + rfm['frequency_score'].astype(str)\n",
    "    \n",
    "    # Creating and analysing RFM segments    \n",
    "    seg_map = {r'[1-2][1-2]': 'hibernating',\n",
    "          r'[1-2][3-4]': 'at risk',\n",
    "          r'[1-2]5': 'cant lose them',\n",
    "          r'3[1-2]': 'about to sleep',\n",
    "          r'33': 'need attention',\n",
    "          r'[3-4][4-5]': 'loyal customers',\n",
    "          r'41': 'promising',\n",
    "          r'[4-5][2-3]': 'potential loyallists',\n",
    "          r'51': 'new_customers',\n",
    "          r'5[4-5]': 'champions'}\n",
    "    \n",
    "    rfm['segments'] = rfm['RFM Score'].replace(seg_map, regex=True)    \n",
    "    rfm = rfm[['segments', 'recency', 'frequency', 'monetary']]\n",
    "    rfm.index = rfm.index.astype(int)\n",
    "    \n",
    "    # Convert to csv\n",
    "    rfm.to_csv('rfm.csv')\n",
    "    \n",
    "    return rfm  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 228,
   "id": "1d29211b",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_df = df_.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "id": "41c68345",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>segments</th>\n",
       "      <th>recency</th>\n",
       "      <th>frequency</th>\n",
       "      <th>monetary</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Customer ID</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>12346</th>\n",
       "      <td>cant lose them</td>\n",
       "      <td>165</td>\n",
       "      <td>11</td>\n",
       "      <td>372.860</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12347</th>\n",
       "      <td>potential loyallists</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>1323.320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12348</th>\n",
       "      <td>hibernating</td>\n",
       "      <td>74</td>\n",
       "      <td>1</td>\n",
       "      <td>222.160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12349</th>\n",
       "      <td>need attention</td>\n",
       "      <td>43</td>\n",
       "      <td>3</td>\n",
       "      <td>2671.140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12351</th>\n",
       "      <td>new_customers</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "      <td>300.930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18283</th>\n",
       "      <td>loyal customers</td>\n",
       "      <td>18</td>\n",
       "      <td>6</td>\n",
       "      <td>641.770</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18284</th>\n",
       "      <td>about to sleep</td>\n",
       "      <td>67</td>\n",
       "      <td>1</td>\n",
       "      <td>461.680</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18285</th>\n",
       "      <td>hibernating</td>\n",
       "      <td>296</td>\n",
       "      <td>1</td>\n",
       "      <td>427.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18286</th>\n",
       "      <td>at risk</td>\n",
       "      <td>112</td>\n",
       "      <td>2</td>\n",
       "      <td>1296.430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18287</th>\n",
       "      <td>loyal customers</td>\n",
       "      <td>18</td>\n",
       "      <td>4</td>\n",
       "      <td>2345.710</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4312 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                         segments  recency  frequency  monetary\n",
       "Customer ID                                                    \n",
       "12346              cant lose them      165         11   372.860\n",
       "12347        potential loyallists        3          2  1323.320\n",
       "12348                 hibernating       74          1   222.160\n",
       "12349              need attention       43          3  2671.140\n",
       "12351               new_customers       11          1   300.930\n",
       "...                           ...      ...        ...       ...\n",
       "18283             loyal customers       18          6   641.770\n",
       "18284              about to sleep       67          1   461.680\n",
       "18285                 hibernating      296          1   427.000\n",
       "18286                     at risk      112          2  1296.430\n",
       "18287             loyal customers       18          4  2345.710\n",
       "\n",
       "[4312 rows x 4 columns]"
      ]
     },
     "execution_count": 229,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "create_rfm(test_df, csv=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}