Find out why your posts aren't working

You post regularly, but hardly anyone likes, comments, or gets in touch. This guide helps you look back at your own posts, see which kinds get a response and which get scrolled past, and pick one change to test for the next two weeks.

Before you start

• Your business account on the platform you care about most. Personal profiles usually don’t show post numbers.
• A list or spreadsheet with your last 20 to 30 posts. Step 2 shows you what to note for each one.
• About 30 minutes. Most of it is copying numbers.
• An AI chat tool such as ChatGPT, Claude, or Gemini for step 4. You can also do it with a pen.

Want to do it yourself?

Difficulty

4 out of 10

Estimated completion time

1–2 hours, plus a two-week test

Skills you’ll need: Export or record post metrics, compare like-for-like averages, and track one change without guessing missing numbers.

I’ll do it myselfI’d rather get help

How this helps your brand

Relevant posts can prompt people to explore your profile and visit a matching website page. Looking beyond engagement to enquiries and bookings helps you assess whether the whole customer journey is working.

1. 1

   Step 1: Find the numbers for each post

   Every major platform shows how each post did, but only on a business or professional account. The page is usually called Insights, Analytics, or Performance. Menus move around often, so if you can’t find it, search the platform’s official help pages for “post insights”.

   Where to start looking:

   • Instagram: switch to a professional account, then open a post and look for “View insights”, or open your professional dashboard.
   • Facebook: open your Page’s professional dashboard or Meta Business Suite and look for Insights.
   • TikTok: use a business or creator account and look for Analytics in your profile menu.
   • LinkedIn: open a post and look for its analytics, or use the Analytics tab on your company Page.
   • Google Business Profile: open your profile while signed in and look for Performance. It shows how people found and contacted you, more than how single posts did.

   Tip: If you just switched to a business account, older posts may have no numbers. Use the posts that do, even if that’s fewer than 20.

2. 2

   Step 2: Make a simple list of your last 20 to 30 posts

   One row per post. You don’t need every number the platform offers. These are enough:

   • The date, and roughly what time you posted.
   • What kind of post it was, in a few words: a tip, an offer, a behind-the-scenes photo, a customer story, a before-and-after, a video of you talking.
   • Whether a real person appeared in it: you, your team, or a customer.
   • Reach (how many people saw it), plus likes, comments, shares, and saves.
   • Any messages, calls, or bookings you know came from it.

   Tip: Screenshots of each post’s insights are fine too. Most AI tools can read numbers from pasted images.

3. 3

   Step 3: Look at what people did, not just how many saw it

   A like takes a second. A comment, a share, a save, or a message takes effort, so those tell you more about what people actually valued.

   Reach mostly tells you how many people had the chance to respond. If reach is low across the board, the content may not be the only issue. If reach is fine but nobody responds, the content is where to look.

4. 4

   Step 4: Group your posts by type and compare

   Don’t chase your single best post. One post can do well by luck. What you want is a type of post that keeps doing better than the others, and a type that keeps falling flat.

   Paste your list into an AI tool with the prompt at the end of this guide. It will group the posts by type and compare them for you. Or do it by hand: sort the list by type and look at the comments, shares, and saves for each group.

   • Which type gets the most comments, shares, saves, or messages?
   • Which type gets the least, even when it reached plenty of people?
   • Do posts with a real person in them do better or worse?

   Tip: Compare posts from the same platform only. What works on LinkedIn won’t tell you much about TikTok.

   Illustrated exampleLook for a pattern across several posts

   Example · one business · one platformCompare a group, not one lucky post.

   Four posts of each type, with similar reach. These are average actions per post.

   • Comments
   • Shares
   • Saves

   Plain offers

   4 actions

   1 comment · 1 share · 2 saves

   Customer stories

   12 actions

   5 comments · 3 shares · 4 saves

   Practical tips

   15 actions

   2 comments · 3 shares · 10 saves

   A pattern to test: tips are saved more often; plain offers get fewer actions. This small sample suggests an idea, not a guarantee of more bookings.

5. 5

   Step 5: Pick one change and test it for two weeks

   Choose one change based on what you found. For example: more customer stories and fewer plain offers, or showing your face in videos instead of using text on a photo.

   Change only that one thing. Keep posting about as often, at about the same times, as before. If you change three things at once, you won’t know which one made the difference.

   Write down the change and today’s date, so you remember what you were testing.

   If the change aims to bring enquiries, check the rest of that journey too: the post should give a relevant reason to explore your profile, the profile link should reach a matching website page, and that page’s contact or booking route should work. Record the visits or enquiries you can actually attribute separately from likes and saves. Use a controlled test before concluding that more engagement will produce more customers.

   Illustrated exampleTurn one finding into a two-week test

   One written changeShow my face in the tip videos.

   Keep the tips, posting days, and usual time the same.

   1. 1
      Day 1 · write it downSave the starting numbers

      Use earlier tip videos without your face as the comparison.

   2. 2
      Days 1–14 · keep it steadyPost the new tip videos

      Track reach, comments, shares, saves, and any enquiries for each one.

   3. 3
      Day 15 · reviewCompare the same kind of post

      Compare average responses per tip video before and after, with reach alongside.

   Write your review date now

   Helped? Keep it and test the next idea.
   No improvement? Drop it and try another pattern.

6. 6

   Step 6: Check the numbers again after two weeks

   Add the new posts to your list and compare them with the old ones of the same type. If the change helped, keep it and test the next idea. If it didn’t, that’s still useful: drop it and try the next pattern from step 4.

   A small account will see small numbers. A handful of extra comments or one or two extra enquiries is a real result.

I run [what your business does] for [who your customers are]. My posts on [platform] get little response and I want to understand why.

Here are my last [number] posts. For each one I’ve noted the date, what kind of post it was, whether a real person appeared in it, and its reach, likes, comments, shares, and saves:
[paste your list or screenshots]

Please:
1. Group the posts by type and compare the comments, shares, saves, and reach for each group.
2. Tell me in plain English which kinds of posts work best for me and which don’t, and how sure you are given how few posts there are.
3. Suggest one change to test for the next two weeks, and tell me exactly what to compare afterwards to know if it worked.

Only use the numbers I’ve given you. Don’t guess at numbers I haven’t shared, and don’t make claims about how the [platform] algorithm works.