BeginnerLesson 5 of 305 min read

Social Media Basics: Choosing a Channel and Measuring It

Choose where to start with a simple weighted scorecard, understand what social media is good for, and judge your posts fairly with engagement rate using a worked example.

In this lesson you will

  • Know what social media is good for
  • Score channels with a weighted scorecard
  • Work out an engagement rate
  • Point every channel to something you own

About 22 minutes

Noor wonders which app to use

Noor has an account on four different social networks and neglects all of them. A friend says she should be on every one. Another says to ignore social media and focus on email. Both are partly right and neither tells her which to choose. A better question is: where do my customers already spend time, and what can I keep up? This lesson gives you a simple way to choose a channel and a simple way to judge your posts.

What social media is good for

  • Being found by people who were not looking. A good short video or image can reach people who never searched for you.
  • Showing the human side. Pouring, packing, answering questions.
  • Proof. Reviews, tagged photos, customer comments.
  • Conversation. Replies and messages, which build trust faster than adverts.

It is less good as your only home. You do not own your following, and a platform can change its rules or how many people see your posts. Use social media to introduce people, and then invite them to your email list or website, which you do own.

Choose with a scorecard

The program below scores four channels on three questions, using weights that say how much each question matters to Noor: is the audience there, does the format suit candles, and can she keep up the effort? The scores and weights are made up for the example. Your own answers might be completely different, and that is the point of doing it yourself.

import statistics

# MADE-UP scores 1-5 for how well each channel fits the shop, with weights
weights = {"audience is there": 3, "format suits candles": 2, "effort I can keep up": 2}
channels = {
    "Instagram": {"audience is there": 5, "format suits candles": 5, "effort I can keep up": 3},
    "Facebook":  {"audience is there": 3, "format suits candles": 3, "effort I can keep up": 4},
    "LinkedIn":  {"audience is there": 1, "format suits candles": 2, "effort I can keep up": 3},
    "Short video": {"audience is there": 4, "format suits candles": 4, "effort I can keep up": 2},
}
top = sum(weights.values()) * 5
ranked = sorted(channels, key=lambda c: -sum(weights[k] * v for k, v in channels[c].items()))
for c in ranked:
    pts = sum(weights[k] * v for k, v in channels[c].items())
    print(f"{c}: {pts} of {top}")
print("start with:", ranked[0])

# MADE-UP post results: reach, likes, comments, shares, saves
posts = {"how to trim a wick": (900, 60, 9, 12, 30), "new scent": (1500, 40, 5, 3, 6),
         "pouring day": (700, 55, 14, 8, 11), "sale post": (1100, 20, 1, 1, 2)}
rates = {}
for name, (reach, *acts) in posts.items():
    rates[name] = 100 * sum(acts) / reach
    print(f"{name}: {rates[name]:.1f}%")
print("best:", max(rates, key=rates.get))
print("median:", round(statistics.median(rates.values()), 1), "%")
Instagram: 31 of 35
Short video: 24 of 35
Facebook: 23 of 35
LinkedIn: 13 of 35
start with: Instagram
how to trim a wick: 12.3%
new scent: 3.6%
pouring day: 12.6%
sale post: 2.2%
best: pouring day
median: 8.0 %

Instagram scores 31 of 35, well ahead of the others, and LinkedIn scores only 13 because her customers are simply not there. The program says to start with Instagram. Notice that a lower-scoring channel is not bad, it is just not the best place to begin. A scorecard does not make the decision for you. It makes your reasons visible, so you can argue with them.

Judging your posts fairly

Likes alone can mislead. A post that reaches 1,500 people and gets 40 likes feels bigger than one that reaches 700 and gets 55, but it is not better. A fairer measure is the engagement rate: the number of people who did something, such as liking, commenting, sharing or saving, divided by the number the post reached. In the second half of the program, ‘how to trim a wick’ scores 12.3 per cent, ‘new scent’ 3.6, ‘pouring day’ 12.6 and ‘sale post’ 2.2. The best is pouring day, and the median is 8.0 per cent. The useful pattern is that the two helpful and behind the scenes posts do far better than the post that asked for a sale. Saves and comments are especially valuable, because they show real interest. Compare like with like, and treat a single post as a hint, not proof.

  1. Score two or three channels honestly with the three questions.
  2. Pick one and commit to it for at least a month.
  3. After each post, note the reach and the actions, and work out the rate.
  4. Make more of what scores well and less of what does not.
  5. Always include a way to join your email list.
Common mistake: chasing follower counts

Buying followers or chasing a big number. Followers who do not care do not buy, and they drag your engagement down. A small, interested audience is worth more.

Check the platform’s own rules

Platforms rename features and change what they show people. For anything technical, such as sizes, formats or advertising rules, check the platform’s own help pages. If you work with paid partnerships or gifts, check the current disclosure rules for your country and platform.

Write three questions that matter for your own scorecard and weight them. Score two channels. Then compute the engagement rate for your last five posts.

Practice for this lesson

Check what you learned

Five quick questions. Pick an answer and you will see straight away whether it is right and why.

Why is social media a poor only home?

  1. It is free
  2. It needs a logo
  3. You do not own your following
  4. It is old

Platforms can change rules and reach, so invite people to a list you own.

What does a channel scorecard make visible?

  1. Your reasons for choosing
  2. The winner automatically
  3. Competitors’ plans
  4. Platform secrets

It shows your reasoning so you can question it.

What is an engagement rate?

  1. Followers divided by likes
  2. Actions divided by reach
  3. Posts per day
  4. Shares only

It compares what people did with how many saw the post.

Why compare engagement rates, not likes?

  1. Likes are banned
  2. Rates are bigger
  3. It is easier
  4. Reach differs between posts

A post with more reach naturally gets more likes.

What should every channel point to?

  1. A rival
  2. Your email list or website
  3. Nothing
  4. A different channel only

Owned channels keep the relationship if a platform changes.

Summary

  • Social media introduces you to people who were not looking.
  • Choose a channel with a scorecard that shows your reasoning.
  • Engagement rate compares actions with reach.
  • Invite people to a list you own.

Next lesson

Lesson 6 covers email marketing.

Questions

Do I need to be on every platform?

No. Choose one or two where your customers are and that you can keep up with, and do them well.

Are the scores and rates real?

No. They are made-up samples to show the method. Use your own weights and posts.

What counts as engagement?

Anything where people act, such as likes, comments, shares and saves. Comments and saves usually show stronger interest than likes.

Progress is saved in this browser.