Why impressions per post is the first number to learn
If you post on LinkedIn for yourself or for a company, you already have more numbers than you can use. Every post shows impressions, reactions, comments and reposts, the profile shows views and followers, and the company page has an analytics section of its own. What most people lack is a simple, repeatable way to turn those counts into a judgement: was that a normal post, a weak one or a hit, and is the profile as a whole doing better than last quarter?
This free lesson builds that foundation in a spreadsheet. You will take twelve posts, add up their engagements, calculate the engagement rate of each post, find the average impressions per post and the overall engagement rate of the profile, and measure how far the best post stands above the average. These five calculations take about fifteen minutes and are the base for everything else in organic LinkedIn analysis, from comparing formats to estimating how many profile views a lead target requires.
The example is Quillhaven Payroll, a fictional company that sells payroll software to small manufacturers. Its founder, Dara Lindqvist, posts twice a week from her personal profile. The company, the people, the posts and every number in this article are invented for learning, and any figure that looks like a benchmark is illustrative only. The data are the same twelve posts you see in the lab above.
The concepts: impressions, engagements and two kinds of rates
An impression is one showing of a post on a screen. If the same person scrolls past the post twice, that is usually two impressions, so impressions are not people. They measure how often the post was put in front of someone, which is the raw supply of attention that everything else depends on. Impressions per post is simply the average: all impressions divided by the number of posts.
An engagement is a visible action on the post. In this course the definition is deliberately narrow: engagements = reactions + comments + reposts. Platform dashboards sometimes add clicks, follows or other actions to their own engagement figure, and they change these definitions from time to time. That is exactly why you should calculate the number yourself from the counts you can see. A definition you wrote down stays comparable from quarter to quarter.
The engagement rate of a post is engagements divided by impressions. It answers the question: of all the times this post was shown, how often did someone do something? A rate makes a small post and a large post comparable. A post with 36 engagements may sound weak next to one with 299, but if the first was shown 1,760 times and the second 12,480 times, their rates are closer than the raw counts suggest.
For the whole profile there is a second rate, and it is easy to get wrong. The overall engagement rate is the sum of all engagements divided by the sum of all impressions. It is not the average of the twelve single rates. Averaging rates gives a post that was shown 1,760 times the same weight as a post that was shown 12,480 times. Dividing the sums weights every impression equally, which is what the question “how does this profile perform?” actually asks.
A fully worked example: twelve posts of the Quillhaven Payroll founder
Here are the twelve posts from September and October 2026 with their format, the four counts taken from the post statistics, and the two calculated columns from the lab: engagements and engagement rate.
| Post | Date | Format | Impressions | Reactions | Comments | Reposts | Engagements | Engagement rate |
|---|---|---|---|---|---|---|---|---|
| P01 | 1 Sep | Text | 2,140 | 38 | 9 | 2 | 49 | 2.29 % |
| P02 | 3 Sep | Image | 3,380 | 61 | 14 | 4 | 79 | 2.34 % |
| P03 | 8 Sep | Document | 5,920 | 118 | 31 | 12 | 161 | 2.72 % |
| P04 | 10 Sep | Text | 1,760 | 29 | 6 | 1 | 36 | 2.05 % |
| P05 | 15 Sep | Video | 2,650 | 44 | 8 | 3 | 55 | 2.08 % |
| P06 | 17 Sep | Poll | 4,410 | 35 | 12 | 2 | 49 | 1.11 % |
| P07 | 22 Sep | Document | 7,240 | 152 | 47 | 19 | 218 | 3.01 % |
| P08 | 24 Sep | Image | 2,980 | 57 | 11 | 3 | 71 | 2.38 % |
| P09 | 29 Sep | Text | 3,120 | 66 | 22 | 5 | 93 | 2.98 % |
| P10 | 1 Oct | Video | 1,930 | 31 | 5 | 2 | 38 | 1.97 % |
| P11 | 6 Oct | Text | 12,480 | 214 | 58 | 27 | 299 | 2.40 % |
| P12 | 8 Oct | Image | 2,590 | 49 | 10 | 2 | 61 | 2.36 % |
| Total | 50,600 | 894 | 233 | 82 | 1,209 | 2.39 % |
The table already shows the shape that almost every LinkedIn profile has. Ten of the twelve posts sit between roughly 1,800 and 6,000 impressions. One document post, P07, reaches 7,240. And one text post, P11, travels to 12,480 impressions, far beyond everything else. The total is 50,600 impressions and 1,209 engagements.
Step 1: engagements per post
In the lab the counts are in columns E (reactions), F (comments) and G (reposts), and the posts are in rows 4 to 15. The engagements of the first post go into H4: =E4+F4+G4, or shorter =SUM(E4:G4). Copy the formula down to H15 by dragging the fill handle; the row numbers move along, so row 5 adds E5, F5 and G5, and so on.
For P04 that is 29 + 6 + 1 = 36. For P11 it is 214 + 58 + 27 = 299. A typical slip at this step is to include the impressions column in the sum, because it sits right next to the reactions. Impressions are views, not actions. If your engagements come out in the thousands, check where the sum starts.
Step 2: the engagement rate of each post
The rate goes into I4: =H4/D4, engagements divided by impressions, copied down to I15. Format the column as a percentage. For P04 the result is 36 ÷ 1,760 = 2.05 %. For P11 it is 299 ÷ 12,480 = 2.40 %.
Look at what the rate does to the ranking. By impressions, P11 is the clear winner. By engagement rate, the best post is P07 (document) with 3.01 %, and P11 is only a little above the middle. The weakest rate belongs to P06, the poll, with 1.11 %: polls collect votes, which are a different kind of action and are not counted as engagements here, so their rate looks low by construction. Reach and response are two separate qualities of a post, and you need both columns to see them.
Step 3: average impressions per post
The average goes into D17: =AVERAGE(D4:D15). AVERAGE adds the twelve values and divides by twelve: 50,600 ÷ 12 = 4,217 impressions per post. This is the number to quote when someone asks how far a post of this profile usually travels, with one caveat that the next figure makes visible.
Only 4 of the twelve posts beat the average. That is not a sign of eight failures. It is what an average does when one value is very large. Without P11 the other eleven posts average 3,465 impressions, and the middle value of the list, the median, is 3,050. A later lab of the course is devoted to this difference between average and median and to a sensible rule for outliers. For now it is enough to remember that “average post” and “typical post” are not the same thing on LinkedIn.
Step 4: the overall engagement rate
The overall rate goes into D18: =SUM(E4:G15)/SUM(D4:D15). The first SUM covers the whole block of reactions, comments and reposts in one go, the second adds all impressions. With the example data that is 1,209 ÷ 50,600 = 2.39 %. If you have already filled column H, =SUM(H4:H15)/SUM(D4:D15) gives the same result.
Now compare this with the tempting shortcut =AVERAGE(I4:I15), the average of the twelve single rates. That formula returns 2.31 %. In this data set the two numbers are close, because the largest post happens to have a rate near the middle. That is luck, not a rule. Had P11 earned a rate of only 1 %, the average of rates would have slipped by about 0.12 percentage points, while the overall rate would have fallen by about 0.34 points, almost three times as much, because P11 alone accounts for about a quarter of all impressions. Use the ratio of sums whenever you describe a profile, a month or a format.
Step 5: how far above the average is the best post?
The last cell, D19, divides the best post by the average: =MAX(D4:D15)/D17. MAX returns the largest value of a range, here 12,480. Divided by the average of 4,217 the result is 2.96. The best post was shown about three times as often as the average post.
This multiple is a quick health check. A value near 1.5 describes a steady profile whose posts all land in the same range. A value of 3 or more says that the period's reach depended heavily on one post. Neither is good or bad in itself, but they call for different commentary. In a report for Quillhaven Payroll the honest sentence is: “Average impressions per post were 4,217, lifted by one text post with 12,480 impressions; a normal post reached about 3,050.”
The formulas at a glance
- Engagements (H4):
=E4+F4+G4or=SUM(E4:G4), copied down to H15. - Engagement rate of a post (I4):
=H4/D4, copied down to I15 and formatted as a percentage. - Average impressions per post (D17):
=AVERAGE(D4:D15), here 4,217. - Overall engagement rate (D18):
=SUM(E4:G15)/SUM(D4:D15), here 2.39 %. - Best post ÷ average post (D19):
=MAX(D4:D15)/D17, here 2.96.
Common mistakes and how to avoid them
The arithmetic is simple, and the same handful of errors still turns up in real reports again and again. Most of them come from mixing definitions or from letting one unusual post speak for the whole profile.
- Averaging the rates. The average of single engagement rates treats every post as equal, whatever its reach. For a profile, a month or a format, divide the sum of engagements by the sum of impressions.
- Comparing rates with different definitions. One report counts clicks as engagements, another does not; one divides by impressions, another by followers. Rates built in different ways cannot be compared. Fix one definition and keep it.
- Reading impressions as people. Impressions count showings. The number of different people who saw a post is lower, and for personal profiles it is not always available. Do not write “12,480 people saw the post”.
- Judging a post by reactions alone. Reactions are the cheapest action. A post with few reactions and many thoughtful comments from the right people can be worth more than a popular photo. The course later weights the interactions and looks at who comments.
- Letting one viral post set the target. If the plan for next quarter assumes that every post will perform like P11, the plan will fail. Plan with the typical post and treat the hit as a bonus.
- Comparing with outside benchmarks. Published engagement-rate benchmarks mix industries, audience sizes and definitions. Treat them as illustrative. Your own last twelve posts are the benchmark that fits your audience.
How to read the result
Put the five numbers together and Quillhaven Payroll has a clear first picture of the founder profile. A post is shown about 4,217 times on average and about 3,050 times in a typical case. Roughly 2.4 % of impressions lead to a reaction, comment or repost. Document posts (P03 and P07) combine high reach with high rates, which is a hint worth testing with more data. And the reach of the period leaned on one text post that travelled three times as far as the average.
Just as important is what the numbers do not say. They do not say whether the right people saw the posts: 12,480 impressions among students and competitors help less than 2,000 among payroll managers of manufacturers. They do not say whether anyone visited the profile, started a conversation or asked for a demo. And twelve posts are a small sample, so a difference of a few tenths of a percentage point between two posts is noise. Reach and engagement are the top of a chain, and the rest of the course follows that chain down to inbound leads.
Using the sheet for your own profile
To repeat the exercise with real data, open the statistics of your last ten to twenty posts and type the date, format, impressions, reactions, comments and reposts into six columns. For a company page the same counts can usually be exported from the page analytics; for a personal profile they are often copied by hand, which takes a few minutes per month. Then add the two calculated columns and the three summary cells exactly as in the lab.
Update the sheet at a fixed rhythm, for example on the first working day of each month, and wait at least a week after a post went out before you record it, because posts keep collecting impressions for several days. Keep one row per post and never overwrite old rows. After three months you will have a post log that supports every later analysis in this course: by format, by weekday, by topic and by month.
What the rest of the course covers
This lesson is the first of twenty labs. Each lab uses a new data set of Quillhaven Payroll and adds one layer to the analysis, with every cell checked as you type. The next labs build directly on what you have just done:
Engagement by format. You summarise posts by format with COUNTIF and SUMIFS and compare impressions per post and engagement rate for text, image, document, video and poll.
A weighted engagement score. You give comments and reposts more weight than reactions, calculate points per 1,000 impressions and highlight the three best posts.
Follower growth and follower fit. You turn new followers and unfollows into net growth, then check which share of the audience matches the target profile and whether recent followers fit better than the existing base.
Connection cadence and inbound leads. You track weekly connection requests and their acceptance rate against a self-set cap, follow profile views through conversations to leads and work out the views a lead target requires.
A social selling score sheet and beyond. You build a four-pillar score sheet for a team, then go deeper: posting day and time, outliers, comment quality, company page against personal profiles, content pillars, a follower forecast, a test of connection notes, leads by first touch, the cost of time, a what-if grid, data quality checks and a final quarterly dashboard.
Summary
Impressions per post, engagement rate and the overall rate are the base of organic LinkedIn analysis. Engagements are reactions plus comments plus reposts: =E4+F4+G4. The engagement rate of a post is =H4/D4. The average post comes from =AVERAGE(D4:D15), the overall rate from =SUM(E4:G15)/SUM(D4:D15) and the weight of the best post from =MAX(D4:D15)/D17. For the fictional Quillhaven Payroll founder the twelve posts average 4,217 impressions, the overall engagement rate is 2.39 %, and the best post reached 2.96 times the average.
Now scroll up to the lab and build the sheet yourself. The Coach checks each cell as you go, and at the end you will have a small template you can fill with the numbers of your own profile today.
Frequently asked questions
How do I calculate the engagement rate of a LinkedIn post in Excel?
Add reactions, comments and reposts to get the engagements, then divide by the impressions of the post, for example =(E4+F4+G4)/D4, and format the cell as a percentage. Write down the definition, because dashboards sometimes include clicks or divide by followers instead.
What is the difference between the overall engagement rate and the average engagement rate?
The overall rate divides the sum of all engagements by the sum of all impressions, so every impression counts equally. The average of single post rates gives a small post the same weight as a large one. For a profile, a month or a format use the ratio of sums.
How do I calculate average impressions per post?
Use =AVERAGE(range) on the impressions column, for example =AVERAGE(D4:D15). In the fictional example twelve posts with 50,600 impressions in total average 4,217 impressions per post.
Why are most of my posts below my own average?
One very large post pulls the average upwards. In the example only four of twelve posts are above the average of 4,217 impressions, while the middle post reached about 3,050. Compare the average with the median before you judge a post as weak.
Are impressions the same as the number of people who saw a post?
No. An impression is one showing of the post on a screen, and the same person can account for several impressions. The number of different people reached is lower.
What is a good engagement rate on LinkedIn?
There is no single figure. Published benchmarks mix industries, audience sizes and definitions and are illustrative at best. The useful comparison is your own profile over time, calculated with one fixed definition. All numbers in this lesson are fictional.
Keep going: Organic LinkedIn for Personal Brands & Company Pages: Impressions per Post, Engagement by Format, Follower Fit, Connection Cadence & Inbound Leads
This free lab is lesson 1 of 20. Unlock the other 19 hands-on labs, the final exam and your certificate for a one-time $9.90.
- Engagement by Format: Posts, Impressions per Post and Engagement Rate for Text, Image, Document, Video and Poll with COUNTIF and SUMIFS
- A Weighted Engagement Score: Comments Count More Than Reactions, Points per 1,000 Impressions and a Top-3 Highlight
- Follower Growth: New Followers, Unfollows, Net New, the Monthly Growth Rate and a Line Chart of a Company Page
- Follower Demographics Fit: Share by Job Function, the Target-Profile Share and a Fit Index for New Followers
- Connection Cadence: Weekly Requests, Acceptance Rate, a Self-Set Cap with IF and a Running Total of New Connections
- From Profile Views to Inbound Leads: Step Rates, Leads per 1,000 Views and the Views Needed for a Lead Target
- A Social Selling Score Sheet: Four Pillars, a Total Score, Bands with IFS and the Weakest Pillar with XLOOKUP
- When to Post: Average Impressions by Weekday and Time Slot with AVERAGEIFS, a Two-Criteria Check and a Column Chart
- Typical Reach vs One Viral Post: Average, Median, an Outlier Flag with IF, LARGE and the Average Without Outliers
- Comment Quality: Target-Profile Share of Comments, the Author's Reply Rate and Comments per 1,000 Impressions
- Company Page vs Personal Profiles: Impressions per Follower, Impressions per Post and Engagement Rate by Account Type
- Content Pillars: Share of Posts vs Share of Engagements, an Index per Pillar, a Do-More Rule and a Clustered Chart
- Follower Forecast: SLOPE, a Trend Line with FORECAST.LINEAR, a Three-Month Outlook and the Month a Target Is Reached
- Testing a Connection Note: Acceptance and Reply Rate per Variant, the Difference in Points, Relative Lift and a Sample Check
- Inbound Leads by First Touch: Leads, Pipeline, Won Deals and Won Revenue with COUNTIFS and SUMIFS on Two Criteria
- What Organic Costs: Hours per Activity, Cost of Time, Cost per Lead, Leads per 10 Hours and a Rank with RANK.EQ
- What-If Grid: Leads per Month for Every Posting Frequency and Views-to-Lead Rate with Mixed References and a Heat Map
- Before You Analyse: Duplicate Rows, Impossible Values and Missing Dates in a Post Export with IF, OR and SUMPRODUCT
- Final Project: A Quarterly LinkedIn Dashboard with Impressions, Engagement Rate, Profile Views per 1,000, Leads, a Target Flag and a Combo Chart