What social reports are for
Reporting is not the same as measuring. Measuring is collecting numbers; reporting is choosing which of them to show, to whom, with what comparison, so that a decision can be made. A weekly report to the social team decides what to post next week. A monthly report to the marketing lead decides where to put effort and budget. A quarterly report to leadership decides whether social is doing what the strategy asked of it and whether the strategy should change. Each has a different reader, a different question and a different set of numbers, and a single report that tries to serve all three serves none.
The goal comes first. The social media strategy hub gives each platform a role: one is for awareness, another for consideration, a third for community and retention, and a report that does not know the role cannot say whether the platform is succeeding. The measurement plan framework formalises this: for each goal, a question, a metric that answers it, a source for the metric and a target or a comparison. Without the plan the report is a list; with it, every number on the page is there because it answers a question somebody asked. The KPI trees guide shows how to break a business goal into the metrics that lead to it, and the same decomposition is what turns "do well on social" into something a report can check.
Metrics by goal
The platforms report dozens of metrics, and most of them matter for one goal and not for others. The table maps the common goals to the metrics that measure progress and the ones that are diagnostics rather than results.
| Goal for the platform | Primary metric (the KPIA metric tied directly to an objective.) | Supporting metrics | Diagnostics, not results |
|---|---|---|---|
| Awareness | ReachUnique people who saw an ad. among the intended audience, where the platform reports audience breakdown | Reach growth over time; share of reach from non-followers; video views to a meaningful point | Impressions; follower count; likes |
| Consideration | Actions that show intent: profile visits, saves, shares, tracked link clicks, click-through (CTRShare of impressions that result in a click.) on link posts | Visits to the site from social with time or pages viewed; enquiries and sign-ups with a social source | Comments; reach |
| Traffic and conversion | Tracked sessions and conversions attributed to social, with the attribution rule stated | Conversion rate of social sessions; assisted conversions | Clicks without landing confirmation |
| Community and retention | Conversation: comments and replies per post, returning commenters, questions answered | Response time against standard; sentiment on a sample; member-generated content | Follower count; total interactions |
| Reputation and category presence | Share of VoiceYour share of market visibility.: the brand's share of mentions against named alternatives | Sentiment of mentions; mentions by source | Brand hashtag volume |
| Content learning | Grouped performance by pillar, format and slot against the brand's own median | Audience retention on video; saves per reach | Any single post's result |
Two cautions about the table. First, definitions differ: reach on one platform is not reach on another, views are counted after different durations, and engagement bundles different actions. A report that adds these across platforms is adding apples to pears, so report per platform and compare trends, not totals. Meta's Business Help Centre documents how its metrics are defined and when definitions change; other platforms do the same, and the definitions should be noted in the report's footer. Second, exposure measures need a FrequencyAverage times a person saw an ad. check: reach that is growing because the same people see the brand more often tells a different story from reach that is growing because new people are arriving, and the platforms' audience breakdowns are where the difference shows. Mentions and their sentiment come from outside the brand's own accounts, and the guide to setting up social listening covers the queries, the hand-tagged sample and the monthly note that supply them.
The comparison of reach and impressions explains why awareness reports should lead with people reached rather than ads served.
Native analytics vs reporting tools
Every platform provides its own analytics, and they are the source of truth for that platform: nothing a third-party tool shows is more accurate than what the platform reports, because the tool reads from the same interface. For a brand on one or two platforms with a small team, native analytics plus a spreadsheet is a complete reporting system, and starting there teaches the team what each metric means before a tool abstracts it.
Native analytics have real limits. Each platform shows only itself, so comparing platforms means switching tabs and reconciling definitions. History is kept for a window the platform decides and does not document consistently, so a brand that wants to compare this year with last has to have exported the numbers at the time. Posts cannot be grouped by the brand's own categories, such as pillar, format or campaign, unless the brand tags them somewhere else. And none of them join to the website, the CRM or sales, which is where the consideration and conversion goals are actually measured. Social is often an early touch in those journeys, so the conversion row reports Assisted ConversionsConversions in which a channel appeared earlier on the path but was not the final interaction. beside last-click results rather than adding them together. A reporting layer, whether a spreadsheet the team maintains, a general-purpose dashboard tool connected to the platforms' data, or a dedicated social tool, earns its cost when those limits bite: multiple platforms, a need for history, grouping by the brand's categories, or joining to downstream data.
Whatever holds the report, keep the brand's own export as the record. Tools change their connectors, platforms change their interfaces, contracts end, and a brand whose history lives only inside a vendor's dashboard has no history. A monthly export of the core metrics per platform, in a file the brand owns, is cheap insurance. How the dashboard itself should be laid out, which charts suit which comparison and how to avoid the wall of tiles, is the subject of the dashboard design guide; this page stays with what the report should contain.
Cadence: weekly, monthly, quarterly
Cadence is the schedule on which reports are produced, and the reason to fix it is that each cadence answers a different question with a different amount of data. A week is enough data to see what went out and what travelled, and not enough to judge a pillar; a month is enough to compare pillars and formats against the brand's medians; a quarter is enough to see whether the platform is moving the goal it was given. Reporting a quarterly question weekly produces noise and panic; reporting a weekly question quarterly produces a team that learns slowly.
Weekly reports go to the people producing the content and take minutes to read. They cover what was published against the calendar, which posts stood out in either direction and a one-line hypothesis about why, the inbox against the response standard, and anything that needs a decision this week. No charts are needed; a short note and a link to the calendar suffice. Reading the inbox against the response standardA written commitment about which customer and audience messages get answered, how fast, by whom and how, set per message type and channel. means counting, by message type, how many replies arrived within the promised time.
Monthly reports go to whoever owns marketing and compare the month with the brand's own previous months. They cover each platform against its goal metric, performance grouped by pillar, format and slot against the brand's medians, community health against the standards, a short section on what was learned, and a list of changes for next month. This is the report where the calendar skeleton is adjusted and budget between platforms is reconsidered. For a business that sells to other businesses, the tutorial on building a monthly LinkedIn report shows how to join page and post figures to Campaign Manager and the CRM so the month ends in pipeline rather than engagement.
Quarterly reports go to leadership and answer the strategy's question: is each platform doing what it was given to do, at what cost, and should the allocation change. They include the trend of the goal metric per platform over the past several quarters, the contribution of social to the measures the business cares about, the share of voice trend where it is tracked, and recommendations about platforms to add, reduce or leave. This is the report that feeds the strategy hub's decisions.
The planner below lays out which metrics, readers and decisions belong to each cadence, so a team can see its whole reporting calendar on one screen and notice where a question is being asked at the wrong frequency.
Reporting cadence planner
Comparing against yourself
Published benchmarks are the most common comparison in social reports and the least useful. They average brands of different sizes, categories, budgets and formats, they rarely state how the metric was defined or which platform's definition was used, and they are usually produced by companies that sell tools. A brand cannot learn from them whether this month was good. What it can learn from is its own history, broken down finely enough that like is compared with like: this month's short videos on this platform against the median of the past six months' short videos on the same platform; this pillar against that pillar in the same slot; this campaign against the last campaign of the same kind.
Build the baseline first. For each platform, keep a rolling median and a spread for the goal metric and the main supporting metrics, by format and pillar, so that a new post or a new month can be placed against it: typical, better than typical, or an outlier. Medians rather than averages, because social results are skewed by occasional posts that travel, and an average that includes one viral post flatters every month it is in. Note the context that explains the outliers, such as a campaign, a news event, a platform change or a holiday, in the record, so that next year's comparison can account for them. The tutorial on building a social media performance baseline walks through that record post by post: the tags, the groups worth a figure, the typical level and spread of each, and how the latest posts are placed against them.
Comparing against yourself also means changing one thing at a time where you can. A month in which the cadence, the formats and the pillars all changed cannot be read; a month in which only the format mix changed can. The content performance guide takes the same approach for the wider content programme, separating engagement from outcomes and grouping content so that comparisons mean something. Where the audience itself has changed, say through a follower campaign or a viral post that brought in people outside the intended audience, note it, because a baseline built on one audience will mislead about the next.
Reports that end in decisions
The structure of a decisive report is the same at every cadence. It opens with the goal and the question being asked. It shows the goal metric against the brand's own baseline, with the context that explains any outlier. It shows the supporting metrics only where they explain the goal metric's movement. It states what was learned, in a sentence each, with the evidence. And it ends with decisions: what to do more of, what to stop, what to test, who owns each, and when the result will be checked. A report without that final section is a measurement, and the reader is left to work out the decision themselves, which most will not do. Writing each finding as what, so what and now what produces that final section almost by default.
Two habits make the decision section honest. The first is to write the hypothesis before the data: what did the team expect this month, and did it happen. A report that only explains results after the fact can explain anything. The second is to close the loop: each decision from last month's report appears in this month's, with whether it was done and what happened. Decisions that are never followed up train everyone to ignore the section. Keep the report short enough that the decision section is read; a page for the monthly, a few lines for the weekly, a short deck for the quarterly. Everything else lives in the dashboard or the calendar, linked from the report, for anyone who wants to look. The social media management hub places reporting in the routine alongside pages and community; the discipline of the decision section is what makes the routine worth running.
A furniture retailer's monthly report
A furniture retailer runs 3 platforms with distinct goals: Instagram for consideration, Pinterest for traffic and Facebook for community around a loyalty group. The October report shows Instagram saves per post at 48 against a 6-month median of 31, driven by a new room-tour carousel format that accounts for most of the lift; Pinterest sessions down 14% on September, which the context note attributes to a seasonal drop seen in each of the past 3 years; and 220 comments in the loyalty group with a median response time of 3 hours against a 4-hour standard. Decisions: move 2 of the 4 weekly Instagram slots to the carousel format and re-check in December; leave Pinterest unchanged until the November comparison; feature the 5 most active group members in a post. November's report opens by confirming the carousel change happened and reports saves per post at 52.
Common mistakes
- Reporting every metric the platform offers. A list is not a report; the reader cannot find the decision in it.
- Adding metrics across platforms. Reach on one platform is not reach on another; report per platform and compare trends.
- Benchmarks instead of a baseline. They cannot say whether this month was good; the brand's own history can.
- Averages where medians belong. One post that travelled flatters the whole month.
- Judging a quarterly question weekly. Noise is read as signal and the plan changes every week.
- No history of your own. When the tool or the platform changes, the record goes with it.
- No decision section, or decisions nobody follows up. The report becomes a ritual.
- Ignoring audience change. A baseline built on one audience misleads about the next.
