2018/day26_RFM.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How to segment your customers and increase sales with RFM analysis\n",
    "\n",
    "[Source](https://joaocorreia.io/blog/rfm-analysis-increase-sales-by-segmenting-your-customers.html)\n",
    "\n",
    "RFM analysis from customer purchase history data. The sample orders file is Sample - Superstore dataset from Tableau."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "%matplotlib inline\n",
    "sns.set()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "Read the sample orders file, containing all past purchases for all customers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "orders = pd.read_csv('data/sample-orders.csv',sep=',', \n",
    "                     encoding='latin1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>order_date</th>\n",
       "      <th>order_id</th>\n",
       "      <th>customer</th>\n",
       "      <th>grand_total</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9/7/11</td>\n",
       "      <td>CA-2011-100006</td>\n",
       "      <td>Dennis Kane</td>\n",
       "      <td>378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7/8/11</td>\n",
       "      <td>CA-2011-100090</td>\n",
       "      <td>Ed Braxton</td>\n",
       "      <td>699</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3/14/11</td>\n",
       "      <td>CA-2011-100293</td>\n",
       "      <td>Neil Franzsisch</td>\n",
       "      <td>91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1/29/11</td>\n",
       "      <td>CA-2011-100328</td>\n",
       "      <td>Jasper Cacioppo</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4/8/11</td>\n",
       "      <td>CA-2011-100363</td>\n",
       "      <td>Jim Mitchum</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  order_date        order_id          customer  grand_total\n",
       "0     9/7/11  CA-2011-100006       Dennis Kane          378\n",
       "1     7/8/11  CA-2011-100090        Ed Braxton          699\n",
       "2    3/14/11  CA-2011-100293  Neil Franzsisch           91\n",
       "3    1/29/11  CA-2011-100328   Jasper Cacioppo            4\n",
       "4     4/8/11  CA-2011-100363       Jim Mitchum           21"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "orders.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 5009 entries, 0 to 5008\n",
      "Data columns (total 4 columns):\n",
      "order_date     5009 non-null object\n",
      "order_id       5009 non-null object\n",
      "customer       5009 non-null object\n",
      "grand_total    5009 non-null int64\n",
      "dtypes: int64(1), object(3)\n",
      "memory usage: 156.6+ KB\n"
     ]
    }
   ],
   "source": [
    "orders.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create the RFM Table\n",
    "Since recency is calculated for a point in time and the Tableau Super Store dataset last order date is Dec 31 2014, that is the date we will use to calculate recency.\n",
    "\n",
    "Set this date to the current day and extract all orders until yesterday."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import datetime as dt\n",
    "NOW = dt.datetime(2014,12,31)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Make the date_placed column datetime\n",
    "orders['order_date'] = pd.to_datetime(orders['order_date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfmTable = orders.groupby('customer').agg({'order_date': lambda x: (NOW - x.max()).days, # Recency\n",
    "                                        'order_id': lambda x: len(x),      # Frequency\n",
    "                                        'grand_total': lambda x: x.sum()}) # Monetary Value\n",
    "\n",
    "rfmTable['order_date'] = rfmTable['order_date'].astype(int)\n",
    "rfmTable.rename(columns={'order_date': 'recency', \n",
    "                         'order_id': 'frequency', \n",
    "                         'grand_total': 'monetary_value'}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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_value</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>customer</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Aaron Bergman</th>\n",
       "      <td>415</td>\n",
       "      <td>3</td>\n",
       "      <td>887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Aaron Hawkins</th>\n",
       "      <td>12</td>\n",
       "      <td>7</td>\n",
       "      <td>1744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Aaron Smayling</th>\n",
       "      <td>88</td>\n",
       "      <td>7</td>\n",
       "      <td>3050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Adam Bellavance</th>\n",
       "      <td>54</td>\n",
       "      <td>8</td>\n",
       "      <td>7756</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Adam Hart</th>\n",
       "      <td>34</td>\n",
       "      <td>10</td>\n",
       "      <td>3249</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 recency  frequency  monetary_value\n",
       "customer                                           \n",
       "Aaron Bergman        415          3             887\n",
       "Aaron Hawkins         12          7            1744\n",
       "Aaron Smayling        88          7            3050\n",
       "Adam Bellavance       54          8            7756\n",
       "Adam Hart             34         10            3249"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfmTable.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Customer **Aaron Bergman** has \n",
    "* frequency:  3, \n",
    "* monetary value:  $887 and \n",
    "* recency:  415 days."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>order_date</th>\n",
       "      <th>order_id</th>\n",
       "      <th>customer</th>\n",
       "      <th>grand_total</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>624</th>\n",
       "      <td>2011-02-19</td>\n",
       "      <td>CA-2011-152905</td>\n",
       "      <td>Aaron Bergman</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>665</th>\n",
       "      <td>2011-03-07</td>\n",
       "      <td>CA-2011-156587</td>\n",
       "      <td>Aaron Bergman</td>\n",
       "      <td>310</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2336</th>\n",
       "      <td>2013-11-11</td>\n",
       "      <td>CA-2013-140935</td>\n",
       "      <td>Aaron Bergman</td>\n",
       "      <td>564</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     order_date        order_id       customer  grand_total\n",
       "624  2011-02-19  CA-2011-152905  Aaron Bergman           13\n",
       "665  2011-03-07  CA-2011-156587  Aaron Bergman          310\n",
       "2336 2013-11-11  CA-2013-140935  Aaron Bergman          564"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aaron = orders[orders['customer']=='Aaron Bergman']\n",
    "aaron"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Inserting the date of Aaron purchase and comparing it to the recency in the rfmTable we verify our RFM table is correct."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(NOW - dt.datetime(2013,11,11)).days==415"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Determining RFM Quartiles"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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_value</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.25</th>\n",
       "      <td>30.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1145.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.50</th>\n",
       "      <td>75.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2257.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.75</th>\n",
       "      <td>183.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>3784.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      recency  frequency  monetary_value\n",
       "0.25     30.0        5.0          1145.0\n",
       "0.50     75.0        6.0          2257.0\n",
       "0.75    183.0        8.0          3784.0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quantiles = rfmTable.quantile(q=[0.25,0.5,0.75])\n",
    "quantiles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Send quantiles to a dictionary, easier to use"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'recency': {0.25: 30.0, 0.5: 75.0, 0.75: 183.0},\n",
       " 'frequency': {0.25: 5.0, 0.5: 6.0, 0.75: 8.0},\n",
       " 'monetary_value': {0.25: 1145.0, 0.5: 2257.0, 0.75: 3784.0}}"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "quantiles = quantiles.to_dict()\n",
    "quantiles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creating the RFM segmentation table"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfmSegmentation = rfmTable"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We create two classes for the RFM segmentation since, being high recency is bad, while high frequency and monetary value is good."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Arguments (x = value, p = recency, monetary_value, frequency, k = quartiles dict)\n",
    "def RClass(x,p,d):\n",
    "    if x <= d[p][0.25]:\n",
    "        return 1\n",
    "    elif x <= d[p][0.50]:\n",
    "        return 2\n",
    "    elif x <= d[p][0.75]: \n",
    "        return 3\n",
    "    else:\n",
    "        return 4\n",
    "    \n",
    "# Arguments (x = value, p = recency, monetary_value, frequency, k = quartiles dict)\n",
    "def FMClass(x,p,d):\n",
    "    if x <= d[p][0.25]:\n",
    "        return 4\n",
    "    elif x <= d[p][0.50]:\n",
    "        return 3\n",
    "    elif x <= d[p][0.75]: \n",
    "        return 2\n",
    "    else:\n",
    "        return 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfmSegmentation['R_Quartile'] = rfmSegmentation['recency'].apply(RClass, args=('recency',quantiles,))\n",
    "rfmSegmentation['F_Quartile'] = rfmSegmentation['frequency'].apply(FMClass, args=('frequency',quantiles,))\n",
    "rfmSegmentation['M_Quartile'] = rfmSegmentation['monetary_value'].apply(FMClass, args=('monetary_value',quantiles,))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "rfmSegmentation['RFMClass'] = rfmSegmentation.R_Quartile.map(str) \\\n",
    "                            + rfmSegmentation.F_Quartile.map(str) \\\n",
    "                            + rfmSegmentation.M_Quartile.map(str)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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_value</th>\n",
       "      <th>R_Quartile</th>\n",
       "      <th>F_Quartile</th>\n",
       "      <th>M_Quartile</th>\n",
       "      <th>RFMClass</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>customer</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>Aaron Bergman</th>\n",
       "      <td>415</td>\n",
       "      <td>3</td>\n",
       "      <td>887</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Aaron Hawkins</th>\n",
       "      <td>12</td>\n",
       "      <td>7</td>\n",
       "      <td>1744</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Aaron Smayling</th>\n",
       "      <td>88</td>\n",
       "      <td>7</td>\n",
       "      <td>3050</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Adam Bellavance</th>\n",
       "      <td>54</td>\n",
       "      <td>8</td>\n",
       "      <td>7756</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Adam Hart</th>\n",
       "      <td>34</td>\n",
       "      <td>10</td>\n",
       "      <td>3249</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>212</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 recency  frequency  monetary_value  R_Quartile  F_Quartile  \\\n",
       "customer                                                                      \n",
       "Aaron Bergman        415          3             887           4           4   \n",
       "Aaron Hawkins         12          7            1744           1           2   \n",
       "Aaron Smayling        88          7            3050           3           2   \n",
       "Adam Bellavance       54          8            7756           2           2   \n",
       "Adam Hart             34         10            3249           2           1   \n",
       "\n",
       "                 M_Quartile RFMClass  \n",
       "customer                              \n",
       "Aaron Bergman             4      444  \n",
       "Aaron Hawkins             3      123  \n",
       "Aaron Smayling            2      322  \n",
       "Adam Bellavance           1      221  \n",
       "Adam Hart                 2      212  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rfmSegmentation.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Who are the top 5 best customers? by RFM Class **(111)**, high spenders who buy recently and frequently?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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_value</th>\n",
       "      <th>R_Quartile</th>\n",
       "      <th>F_Quartile</th>\n",
       "      <th>M_Quartile</th>\n",
       "      <th>RFMClass</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>customer</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>Sanjit Engle</th>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>12210</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>John Lee</th>\n",
       "      <td>21</td>\n",
       "      <td>11</td>\n",
       "      <td>9801</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pete Kriz</th>\n",
       "      <td>9</td>\n",
       "      <td>12</td>\n",
       "      <td>8647</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Harry Marie</th>\n",
       "      <td>2</td>\n",
       "      <td>10</td>\n",
       "      <td>8237</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Lena Creighton</th>\n",
       "      <td>16</td>\n",
       "      <td>12</td>\n",
       "      <td>7661</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                recency  frequency  monetary_value  R_Quartile  F_Quartile  \\\n",
       "customer                                                                     \n",
       "Sanjit Engle          9         11           12210           1           1   \n",
       "John Lee             21         11            9801           1           1   \n",
       "Pete Kriz             9         12            8647           1           1   \n",
       "Harry Marie           2         10            8237           1           1   \n",
       "Lena Creighton       16         12            7661           1           1   \n",
       "\n",
       "                M_Quartile RFMClass  \n",
       "customer                             \n",
       "Sanjit Engle             1      111  \n",
       "John Lee                 1      111  \n",
       "Pete Kriz                1      111  \n",
       "Harry Marie              1      111  \n",
       "Lena Creighton           1      111  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(rfmSegmentation[rfmSegmentation['RFMClass']=='111']).sort_values('monetary_value', ascending=False).head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# E-mail marketing with RFM segmentation\n",
    "\n",
    "1. Select an RFM segment to focus on (Best Customers, Almost Lost, etc).\n",
    "2. Create an hypothesis as to what would work best for this RFM segment.\n",
    "3. Define a goal for this e-mail campaign and RFM segment.\n",
    "4. Setup email marketing conversion tracking.\n",
    "5. Create an alternate e-mail version, tailored to the RFM segment on which you are focusing.\n",
    "6. Create an A/B testing email campaign, where the control group receives a generic version and the experiment group receives the RFM segment tailored email.\n",
    "7. Analyze results and iterate; do more of what works, less of what doesn’t."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Possible segments \n",
    "1. Best customers\n",
    "2. Big spenders\n",
    "3. Loyal customers\n",
    "4. Recent customers\n",
    "5. Almost lost\n",
    "6. Lost customers\n",
    "7. Loast cheap customers\n",
    "8. Others"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}