Send weekly post engagement metrics by email
Tags: #linkedin #tool #posts #engagement #metrics #analytics #automation #email #naas #notification
Last update: 2023-05-29 (Created: 2022-05-11)
Description: This notebook automates the process of sending weekly post engagement metrics from LinkedIn via email.
Disclaimer:
This code is in no way affiliated with, authorized, maintained, sponsored or endorsed by Linkedin or any of its affiliates or subsidiaries. It uses an independent and unofficial API. Use at your own risk.
This project violates Linkedin's User Agreement Section 8.2, and because of this, Linkedin may (and will) temporarily or permanently ban your account. We are not responsible for your account being banned.
from naas_drivers import linkedin
import naas
from dateutil.parser import parse
import matplotlib.pyplot as plt
try:
import seaborn as sns
except:
!pip install seaborn --user
import seaborn as sns
import pandas as pd
from datetime import datetime, date
import random
import time
# LinkedIn cookies
LI_AT = "ENTER_YOUR_COOKIE_HERE" # EXAMPLE : "AQFAzQN_PLPR4wAAAXc-FCKmgiMit5FLdY1af3-2"
JSESSIONID = "ENTER_YOUR_JSESSIONID_HERE" # EXAMPLE : "ajax:8379907400220387585"
​
# LinkedIn profile url
PROFILE_URL = "ENTER_YOUR_LINKEDIN_PROFILE_HERE" # EXAMPLE "https://www.linkedin.com/in/myprofile/"
​
# The first execution all posts will be retrieved.
# Then, you can use the parameter below to setup the number of posts you want to retrieved from LinkedIn API everytime this notebook is run.
NO_POSTS_RETRIEVED = 10
EMAIL_TO = "ENTER_RECIPIENT_EMAIL_HERE" # you will receive weekly summary at this email
EMAIL_FROM = None # summary will have this email as sender. Only available for your naas email, otherwise you will receive this email from [email protected]
EMAIL_SUBJECT = "LinkedIn Metrics" # subject of your email
Create CSV to store your posts feed
# Custom Path of your CSV with profile URL
profile_id = PROFILE_URL.split("https://www.linkedin.com/in/")[-1].split("/")[0]
csv_output = f"LINKEDIN_POSTS_{profile_id}.csv"
# Change your remote timezone if needed. By default remote timezone is "Europe/Paris"
# naas.set_remote_timezone("Europe/Lisbon")
​
# the default settings below will make the notebook run at 12:00 each Saturday
# for information on changing this setting, please check https://crontab.guru/ for information on the required CRON syntax
naas.scheduler.add(cron="0 12 * * 6")
# this notebook will run each week until de-scheduled
# to de-schedule this notebook, simply run the following command:
# naas.scheduler.delete()
All your posts will be stored in CSV.
def get_posts(file_path):
try:
df = pd.read_csv(file_path)
except FileNotFoundError as e:
# Empty dataframe returned
return pd.DataFrame()
return df
​
​
df_posts = get_posts(csv_output)
df_posts
It will get the last X posts from LinkedIn API (X = number of set in variable "NO_POSTS_RETRIEVED") and update it in your CSV.
PS: On the first execution all posts will be retrieved.
def update_last_posts(
df_posts, csv_output, key="POST_URL", no_posts=10, min_updated_time=60
):
# Init output
df_new = pd.DataFrame()
​
# Init df posts is empty then return entire database
if len(df_posts) > 0:
if "DATE_EXTRACT" in df_posts.columns:
last_update_date = df_posts["DATE_EXTRACT"].max()
time_last_update = datetime.now() - datetime.strptime(
last_update_date, "%Y-%m-%d %H:%M:%S"
)
minute_last_update = time_last_update.total_seconds() / 60
if minute_last_update > min_updated_time:
# If df posts not empty get the last X posts (new and already existing)
df_new = linkedin.connect(LI_AT, JSESSIONID).profile.get_posts_feed(
PROFILE_URL, limit=no_posts, sleep=False
)
else:
print(
f"🛑 Nothing to update. Last update done {round(minute_last_update, 0)} minutes ago."
)
else:
df_new = linkedin.connect(LI_AT, JSESSIONID).profile.get_posts_feed(
PROFILE_URL, limit=-1
)
​
# Concat, save database in CSV and dependency in production
df = pd.concat([df_new, df_posts]).drop_duplicates(key, keep="first")
df["DATE_EXTRACT"] = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
df.to_csv(csv_output, index=False)
naas.dependency.add(csv_output)
​
# Return only last post retrieved
return df.reset_index(drop=True)
​
​
df_posts = update_last_posts(df_posts, csv_output, no_posts=NO_POSTS_RETRIEVED)
df_posts
def get_iso_year(row):
return parse(row.PUBLISHED_DATE).isocalendar()[0]
​
​
def get_iso_week(row):
return parse(row.PUBLISHED_DATE).isocalendar()[1]
​
​
def get_iso_day(row):
return parse(row.PUBLISHED_DATE).isoweekday()
​
​
def get_iso_day_string(row):
return datetime.strptime(
str(parse(row.PUBLISHED_DATE).isoweekday()), "%d"
).strftime("%A")
​
​
# week of the year
df_posts["ISO_YEAR"] = df_posts.apply(get_iso_year, axis=1)
# week of the year
df_posts["ISO_WEEK"] = df_posts.apply(get_iso_week, axis=1)
# day of the year
df_posts["DAY"] = df_posts.apply(get_iso_day, axis=1)
# day of the week
df_posts["DAY OF WEEK"] = df_posts.apply(get_iso_day_string, axis=1)
​
# all posts made this week
df_this_week = df_posts[
(df_posts["ISO_WEEK"] == date.today().isocalendar()[1])
& (df_posts["ISO_YEAR"] == date.today().isocalendar()[0])
].sort_values("DAY")
df_this_week.head()
# if we didn't post this week, send a simple email of encouragement and exit
if df_this_week.empty:
post = """\
<h2>It looks like you took a week off from posting...</h2>
<h4>Good for you, you'll crush it next week!</h4>"""
naas.notification.send(
email_to=EMAIL_TO, subject=EMAIL_SUBJECT, html=post, email_from=EMAIL_FROM
)
raise SystemExit("No posts this week, basic email sent")
likes = df_this_week.LIKES.sum()
views = df_this_week.VIEWS.sum()
shares = df_this_week.SHARES.sum()
comments = df_this_week.COMMENTS.sum()
​
# preview of what will be send by email:
print("This week's cumulative statistics:")
print(
"\n\t👀 Impressions\t",
views,
"\n\t👠Likes\t",
likes,
"\n\t💬 Comments\t",
comments,
"\n\tâ© Shares\t",
shares,
)
top_post = df_this_week.sort_values("VIEWS", ascending=False).iloc[0]
​
# get information on most viewed post
top_post_text = top_post["TEXT"]
top_post_url = top_post["POST_URL"]
top_post_views = top_post["VIEWS"]
top_post_likes = top_post["LIKES"]
top_post_comments = top_post["COMMENTS"]
top_post_day = top_post["DAY OF WEEK"]
top_post_text = top_post_text.rstrip()[0:128] + " (...)"
​
# preview of what will be send by email:
print("âœï¸ Your best post this week got 👀 x", top_post_views, "and 👠x", top_post_likes)
print("You posted it on", top_post_day, "\n")
print("This is what it said:\n\n", top_post_text)
print("See your post on LinkedIn : ", top_post_url)
sns.set_style("darkgrid")
fig, axs = plt.subplots(ncols=2)
fig.tight_layout()
sns.lineplot(x="DAY OF WEEK", y="VIEWS", data=df_this_week, ax=axs[0])
sns.lineplot(x="DAY OF WEEK", y="LIKES", data=df_this_week, palette="deep", ax=axs[1])
​
# save image to production as a public asset :
output_image = "week.png"
fig.savefig(output_image)
link_image = naas.asset.add(output_image)
def get_likes(df_posts):
df_out = pd.DataFrame()
for index, row in df_posts.iterrows():
df = linkedin.connect(LI_AT, JSESSIONID).post.get_likes(row["POST_URL"])
df_out = df_out.append(df)
# aditional pause to protect your LinkedIn account
time.sleep(random.uniform(3, 6))
return df_out
​
​
# count likes for each profile
def count_likes(df):
to_group = ["PROFILE_ID", "FIRSTNAME", "LASTNAME"]
df2 = (
df.groupby(["PROFILE_ID"])
.size()
.sort_values(ascending=False)
.reset_index(name="LIKE_COUNT")
)
return df2
df_likes = get_likes(df_this_week)
df_counts = count_likes(df_likes)
​
top_fan = df_likes[df_likes["PROFILE_ID"] == df_counts["PROFILE_ID"][0]]
​
fan_slug = top_fan["PROFILE_ID"].iloc[0]
fan_first_name = top_fan["FIRSTNAME"].iloc[0]
fan_last_name = top_fan["LASTNAME"].iloc[0]
fan_name = f"{fan_first_name} {fan_last_name}"
fan_likes = df_counts["LIKE_COUNT"][0]
fan_url = f"https://www.linkedin.com/in/{fan_slug}"
​
# preview of what will be send by email
print(
"â¤ï¸ This week", fan_name, "was your biggest fan! They left you", fan_likes, "likes"
)
print("Go say hi on their profile:", fan_url)
# You can edit the basic HTML below to change the look and feel of the email
# but be sure to pay attention to the variable names (indicated by $)
# so you don't unexpectedly break anything
​
html = """\
<h2>Here's your weekly LinkedIn report!</h2>
<h4>🚀 You did great this week !</h4>
​
 👀 $VIEWS impressions<br>
 👠$LIKES likes<br>
 💬 $COMMENTS comments<br>
 â© $SHARES shares<br>
​
<h4>â¤ï¸ This week <i>$FAN_NAME</i> was your biggest fan!</h4>
 They left you 👠x $FAN_LIKES<br><br>
 <a href="$FAN_URL">Go say hi on their profile</a><br>
​
<h4>âœï¸ This week your best post got 👀 x $BEST_VIEWS and 👠x $BEST_LIKES</h4>
 You posted it on $BEST_DAY, and this is what it said: <br>
<br>
  <i>$BEST_TEXT</i>
<br><br>
 <a href="$BEST_URL">See your post on LinkedIn</a>
​
<h4>📈 Here, have some charts (we're out of cookies)</h4>
<img src="$CHART" style="width:640px; height:360px;" /><br>
"""
html = html.replace("$VIEWS", str(views))
html = html.replace("$LIKES", str(likes))
html = html.replace("$COMMENTS", str(comments))
html = html.replace("$SHARES", str(shares))
html = html.replace("$FAN_NAME", str(fan_name))
html = html.replace("$FAN_LIKES", str(fan_likes))
html = html.replace("$FAN_URL", str(fan_url))
html = html.replace("$BEST_VIEWS", str(top_post_views))
html = html.replace("$BEST_LIKES", str(top_post_likes))
html = html.replace("$BEST_DAY", str(top_post_day))
html = html.replace("$BEST_TEXT", str(top_post_text))
html = html.replace("$BEST_URL", str(top_post_url))
html = html.replace("$CHART", str(link_image))
post = html
​
# preview final email
print(post)
# sends the email
naas.notification.send(
email_to=EMAIL_TO, subject=EMAIL_SUBJECT, html=post, email_from=EMAIL_FROM
)
Last modified 1mo ago