Social Media KPIsNumbers that change decisions

Metrics are not KPIs#

Social platforms offer hundreds of metrics. A KPI is a metric you have chosen to steer by because it reflects progress towards the outcome you care about. Most social reports confuse the two, presenting every available number and leaving the reader to guess which ones matter.

The test for a KPI is the decision test: if this number doubled or halved next month, would we do something differently? If the answer is no, it is a metric, perhaps interesting, but not a KPI. A good scorecard has five lines or fewer.

If a number cannot change a decision, it belongs in an appendix, not a dashboard.

The KPI hierarchy#

Social metrics form a chain. At the top are business outcomes that a chief executive recognises. Below them are the social outcomes that drive those results, then the engagement signals that predict the social outcomes, and at the base the activity metrics that describe what the team did. Reporting should start at the top and work down.

Choose KPIs by the job social is doing#

The right KPIs depend on the outcome your social media strategy has given to social. A brand-building programme and a lead generation programme should not share a scorecard. Mixing them is how memory-building work is cancelled for failing to generate leads it was never designed to produce.

Calculate rates the right way#

Engagement rate is the most common social metric and the most inconsistently calculated. Some divide by followers, some by reach, some by impressions. Each gives a different answer. Dividing by followers is misleading because many followers never see a given post, while non-followers might. Dividing by reach is usually the most honest measure of how a post performed with the people who saw it.

Whatever you choose, be consistent and write the definition on the report. Comparisons across agencies, tools or time periods using different definitions are meaningless. The calculator below shows the reach-based approach and the commercial ratios that sit above it.

Leading and lagging indicators#

Lagging indicators, such as revenue and retention, tell you whether you succeeded but arrive too late to steer by. Leading indicators, such as saves, shares and enquiries, move earlier and predict the lagging ones. A useful scorecard contains both, and the team should understand which leading indicators are believed to drive which lagging ones.

That belief should be tested. If share rates rise for three months and enquiries do not move, either the link is weaker than assumed or the content is earning shares from the wrong audience. This is where social reporting becomes analysis rather than recitation.

The attribution problem#

Social's influence is routinely undercounted by last-click attribution. People see a post on their phone, then search for the brand days later on a laptop and buy. Search gets the credit. Equally, platform-reported conversions can overcount, because platforms claim credit for people who would have bought anyway.

Use several lenses together: platform data, your own analytics with consistent UTM parameters, self-reported attribution on forms and calls, and, for larger budgets, incrementality testing. No single source is the truth; triangulation is.

Benchmarks: use with care#

Teams often ask what a good engagement rate or click-through rate is. Published benchmarks exist, but they mix industries, account sizes, platforms and definitions, and they are often produced by tool vendors with their own methodology. Comparing your numbers with someone else's average rarely leads to a useful decision.

Your most reliable benchmark is your own history. Compare this month with the previous three, this format with your other formats, this campaign with your last campaign. Where external comparison helps, compare like for like: the same platform, a similar audience size and the same definition. Treat any single external figure as a rough reference, not a target.

Include the full cost#

Social reports often present cost per result using media spend alone. That flatters organic activity, which appears free, and makes paid look expensive by comparison. A fair cost figure includes people's time, production, tools and agency fees as well as media. Only then can you compare social honestly with other channels or decide where the next rupee should go.

This also changes how you view organic. If a month of organic posting costs a meaningful amount in team time and produces few outcomes, it deserves the same scrutiny as an underperforming ad campaign. Read CAC and LTV for the wider economics.

Vanity metrics and when they matter#

Follower counts, likes and impressions are often dismissed as vanity metrics. They are not useless; they are simply weak evidence on their own. Follower growth in a target audience can indicate rising interest. Impressions matter when frequency is the goal. The problem is when they are reported without context or as the headline result.

Above all, end every report with decisions: what you will do more of, less of and differently next month, and what you will test. A report that ends with numbers invites admiration or anxiety; a report that ends with decisions invites action.

Reporting rhythm#

Look at activity and engagement weekly for operational adjustments. Review the scorecard monthly with recommended actions. Review the link between social and business outcomes quarterly, alongside the strategy. Present leadership with the top of the hierarchy only; see board reporting and marketing dashboards.

Published by Fabulous.Media, a network of specialist marketing agencies. Updated 9 October 2026. Platform features change often; check current official documentation before acting on platform-specific detail.