The KPIs that matter, and the vanity ones.
Most social dashboards drown you in numbers that feel good and predict nothing. This guide breaks down the metrics worth tracking, the exact formulas, and where each one quietly lies to you.
A KPI is a decision you've already agreed to make.
A key performance indicator is not just a number you report. It is a number you have committed to act on. If a metric goes up and your behavior would not change, and it goes down and your behavior still would not change, it is not a KPI. It is decoration. That single test eliminates most of what fills a typical social media report.
The metrics below are ordered roughly the way value flows: from how many people saw you, to how many cared, to how many clicked, to how many did the thing that pays the bills, plus one metric that tells you whether you are winning the category. For each, you get what it means, the formula, and the honest version of how it can mislead you. The benchmark ranges here are general industry guidance to orient you, not guarantees and not figures pulled from any single platform.
The five KPIs worth building a report around.
Track these consistently and you can answer the only question that matters: is the work moving the business, or just the dashboard?
1. Reach & impressions
Reach is unique accounts that saw a post; impressions count every view, including repeats. They size your top of funnel. Watch the ratio: impressions far above reach means a small audience is seeing you many times, which is reach saturation, not growth.
2. Engagement rate
The share of the audience that did something: like, comment, share, save, click. The single best early read on whether content resonates, because it normalizes for audience size and is hard to fake at scale.
3. Click-through rate (CTR)
Of the people who saw a post with a link, how many clicked. This is where attention converts to intent. A strong CTR with weak downstream conversions points at a landing-page problem, not a content problem.
4. Conversions
The action that maps to revenue: a signup, a lead, a purchase, a booked call. The hardest to attribute and the only one a CFO truly cares about. Everything above it is a leading indicator of this.
5. Share of voice (SOV)
Your slice of the total conversation in your category versus competitors. A relative metric: it tells you whether you are growing faster than the people you compete with, which raw follower counts never will.
Reach and impressions: the size of the room.
Reach answers how many distinct people you got in front of. Impressions answers how many times your content was rendered, so one person who sees a post three times is one reach and three impressions. You want both, because the gap between them is itself a signal.
Frequency = impressions ÷ reach. A frequency near 1 means you are reaching mostly new eyes. A frequency of 4 or 5 on organic content usually means the algorithm is recycling you to the same loyal core, which caps growth no matter how high impressions climb. The honest read: rising impressions with flat reach is not a win, it is the same people scrolling past you more often.
Use reach and impressions to size opportunity, never to claim success on their own. They are the denominator for the metrics that actually matter, not the headline.
Engagement rate: the resonance test.
Engagement rate is the most useful single number in social, because it normalizes for audience size. A post that gets 100 interactions from 1,000 followers (10%) is working far harder than one that gets 500 from 100,000 (0.5%). But there is no one formula, and which denominator you choose changes the story:
Three ways to calculate it (pick one and stay consistent)
- By reach: (total engagements ÷ reach) × 100. The fairest measure of content quality, because it counts only people who actually saw the post.
- By impressions: (total engagements ÷ impressions) × 100. Slightly more conservative, useful when reach is not exposed by the platform.
- By followers: (total engagements ÷ followers) × 100. The most common public figure and the most gameable, because it ignores how many followers the post never reached.
General orientation: an engagement rate around 1 to 3 percent is typical for many established accounts, anything above 3 to 5 percent is strong, and sub-1 percent on a large account often signals stale content or an audience that has quietly tuned out. Weight saves and shares above likes when you can, those are the interactions that signal real intent and that platforms reward in distribution. The dishonest move is comparing your follower-based rate to a competitor's reach-based rate. Compare like with like or do not compare at all.
CTR: from attention to intent.
Click-through rate is (clicks ÷ impressions) × 100 (some teams use reach in the denominator; again, be consistent). It is the first metric where a vague good feeling turns into measurable intent, because a click is a person choosing to leave the feed for you.
Social CTRs are humbling. Link clicks in-feed often land well under 1 percent, and that is normal, because most platforms actively suppress off-platform links to keep users in the app. The useful work is diagnostic: a high engagement rate paired with a low CTR usually means your content entertains but never asks for the click, or your call to action is buried. A healthy CTR paired with weak conversions points the finger downstream, at the landing page, the offer, or the load time, not at the post.
Conversions: the only metric finance trusts.
A conversion is the action you actually wanted: a trial start, a qualified lead, a sale, a booked demo. Conversion rate = (conversions ÷ clicks) × 100 for a specific post or campaign. This is the metric that lets you argue for budget, because it is the one that connects to money.
It is also the hardest to attribute honestly. Someone discovers you on social, forgets you, searches your name a week later, and converts through organic search, last-click attribution will hand that win to Google and rob social of credit it earned. Do not let imperfect attribution push you into pretending engagement is the goal. Track conversions with UTM parameters, watch assisted conversions, and accept that social is often a strong first-touch channel whose value shows up one step removed.
Share of voice: are you winning the category?
Share of voice is your portion of the total conversation in your space:your mentions ÷ (your mentions + competitors' mentions) × 100, or the engagement-weighted version of the same idea. It is the one KPI on this list that is inherently relative, and that is exactly why it is valuable. Your engagement can rise 20 percent and still represent a shrinking slice if three competitors grew 40 percent. Absolute numbers feel like progress; share of voice tells you whether you are actually pulling ahead.
You cannot measure this from your own analytics alone, by definition it requires watching competitors too. That is the work most teams skip, which is precisely why the teams that do it tend to know things their rivals do not.
The honest part: vanity metrics.
A vanity metric is any number that reliably goes up, looks impressive in a slide, and has no demonstrated link to a business outcome. Follower count is the classic offender: 50,000 followers who never see or act on your posts are worth less than 2,000 who buy. Raw like counts, total impressions in isolation, and cumulative video views fall into the same trap. They are not worthless, they are context for the real KPIs, but reported as headlines they actively mislead.
The fix is not to ban them, it is to demote them. Pair every vanity number with the metric that gives it meaning: followers with engagement rate, impressions with reach and frequency, views with watch-through rate, likes with saves and shares. A number that survives being paired with its honest denominator is a KPI. A number that only looks good alone is decoration, and decoration is how teams stay busy while the business stalls.
One more honest note: benchmarks are orientation, not law. The ranges in this guide help you sanity-check your own numbers, but the only benchmark that truly matters is your own trend line and your direct competitors. Beat last quarter and out-grow your category, and the absolute figures take care of themselves.
How to turn KPIs into a report you'll actually use.
- Pick one denominator and freeze it. Decide engagement-by-reach or engagement-by-followers once, document it, and never quietly switch. A report whose math changes mid-year is worse than no report.
- Track trend, not snapshots. A single week's engagement rate is noise. The four-week direction is signal. Report the slope, not the dot.
- Always show the pair. Never report a vanity number without its honest partner beside it. Followers next to engagement rate, impressions next to reach.
- Benchmark against rivals, not yourself only. Add share of voice so a flat month against a falling category still reads as a win, and a great month against a surging one still reads as a warning.
- Tie the top of the report to the bottom. Reach feeds engagement feeds clicks feeds conversions. Show that chain so anyone reading sees where attention leaks out.
Where BAT fits, briefly and honestly.
You can run all of this in a spreadsheet, and plenty of good marketers do. BAT exists for when the manual export-and-paste loop stops scaling: it computes per-platform engagement and performance scores over one unified index of your posts and the competitors you track, so owned-versus-rival benchmarking and share-of-voice context sit in the same view instead of three browser tabs. First-party analytics are strongest for LinkedIn, Instagram, and Facebook; X, TikTok, Threads, and Pinterest data is gathered via scrapers, and the product labels which is which rather than pretending it is all the same. If you want to see the full picture, the social media analytics tool is the place to start, and the engagement rate calculator will do the one formula by hand for free. Plans start at $100/mo per brand workspace with a 14-day free trial, there is no free tier.
Keep going: related guides.
Metrics are one piece. These cover the work that moves them.
Engagement rate calculator
Drop in your numbers and get engagement rate by reach, impressions, or followers, with the formula shown.
Social media competitor analysis
How to track rivals and build share of voice into your reporting instead of guessing where you stand.
Best times to post
Reach starts with timing. How to find the windows your own audience is actually online.
Social media report template
A ready structure for turning these KPIs into a report a stakeholder will actually read.
Frequently asked questions.
What are the most important social media KPIs to track?
Reach and impressions (audience size), engagement rate (resonance), click-through rate (intent), conversions (revenue), and share of voice (competitive position). Reach and engagement are leading indicators; conversions are the lagging one finance cares about. Anything you would not act on if it moved is not a KPI, it is a vanity metric.
How do you calculate engagement rate?
Divide total engagements (likes, comments, shares, saves, clicks) by your denominator and multiply by 100. By reach: (engagements ÷ reach) × 100, the fairest measure of content quality. By followers: (engagements ÷ followers) × 100, the most common public figure but the most gameable. Pick one method and use it consistently, and never compare a reach-based rate to a follower-based one.
What is a good engagement rate on social media?
As general orientation, 1 to 3 percent is typical for many established accounts, above 3 to 5 percent is strong, and below 1 percent on a large account often signals stale content. These are ranges to sanity-check against, not guarantees; your own trend line and your direct competitors are the benchmarks that actually matter.
Are followers and likes vanity metrics?
On their own, yes. They reliably climb, look impressive, and have no direct link to business outcomes. The fix is to demote rather than ban them: report followers next to engagement rate, impressions next to reach, and views next to watch-through rate. A number that survives being paired with its honest denominator is a real KPI.
What is share of voice and why does it matter?
Share of voice is your portion of the total conversation in your category: your mentions divided by your mentions plus competitors' mentions, times 100. It matters because it is relative. Your engagement can rise while your slice shrinks if rivals grow faster. It is the only KPI that tells you whether you are actually winning the category rather than just getting bigger.
Why don't my social conversions match my engagement?
Attribution. Someone discovers you on social, leaves, and converts days later through search or direct, which last-click models credit elsewhere. Use UTM parameters, watch assisted conversions, and treat social as a strong first-touch channel whose value often shows up one step removed rather than expecting engagement and conversions to line up post-for-post.
See your real KPIs, not vanity ones.
BAT scores engagement per platform, benchmarks you against the competitors you track, and turns the numbers into a strategy report, so you spend time acting on metrics instead of assembling them.