How to report social results to clients

HubSpot's 2026 survey of 1,100+ marketers found only 37% say it is easy to tie social activity to business outcomes. A report that hides that behind impressions is not a report.

8 minute read

Most social reports describe activity, not results

Open a typical monthly social report and you will find the number of posts published, total impressions, follower growth, and a grid of screenshots of the best-performing posts. Every one of those describes what you did. None of them describes what happened to the client's business as a result. This is an easy trap to fall into because activity metrics are the ones that always exist, always move, and never require an uncomfortable conversation about attribution. It is also why so many reports get skimmed and filed. A client who cannot tell from your report whether the engagement is working will eventually decide it is not, and the deciding usually happens quietly, several months before they tell you.

Agree the numbers before the engagement starts

The moment to choose what you report is before any work happens, while both sides still have the option to disagree. Ask what the client would need to see in six months to renew without hesitating, then turn the answer into two or three named metrics with a definition, a source, and a starting value written into the statement of work. Two or three, not twelve: a report with twelve numbers is a report with no argument in it, and it lets everyone pick the one that flatters their position. Doing this early also surfaces the engagements that were never going to work, such as the client who wants attributed revenue from an organic brand account with no tracked destination. Better to find that out in week one.

One outcome, one leading indicator, one diagnostic

A structure that holds up across most engagements is to carry exactly three kinds of number. The outcome is the thing the client's business cares about, such as qualified enquiries, trials started, bookings, or applications. The leading indicator is the thing that moves first and predicts the outcome, usually a traffic or click number, or saves and shares on the content that drives them. The diagnostic is the thing that explains a change, typically retention, watch time, or performance split by format. The outcome is what you are accountable for, the leading indicator is what you steer by month to month, and the diagnostic is what you use to explain a bad month without sounding defensive.

Say plainly what you cannot attribute

HubSpot's 2026 report, drawn from more than 1,100 marketers, found that only 37% say it is easy to tie social activity to business outcomes, and that 69% of social teams are under increasing pressure to prove ROI. The barriers respondents named were mostly structural rather than skill-related: platform limits on linking out at nearly 36%, lack of tools at 27%, and the complexity of multi-touch attribution at 26%. A report that quietly implies a clean causal line from a Reel to a sale is claiming something the measurement layer cannot support, and a client with an analytics team will eventually notice. State the attribution model, state its known blind spots, and where a number is directional rather than measured, label it. Being explicit about uncertainty reads as competence, not weakness.

Cadence: monthly reports, fortnightly contact

AgencyAnalytics surveyed 494 agency professionals between February and April 2026 and found 69% send client reports monthly, up from 65% the year before, with 11% weekly, 8% fortnightly and 3% quarterly. Monthly is the right default for a report, because shorter windows produce too much noise on organic social to distinguish a trend from a quiet fortnight. But monthly reporting is not the same as monthly contact, and the gap between reports is where doubt accumulates. A short note every couple of weeks, saying what shipped and what you are watching, costs ten minutes and removes the sense that nothing is happening. The report is a conversation with a document attached, not a document delivered in place of a conversation.

Benchmarks from vendor blogs are a trap in a client report

Dropping an industry benchmark into a client report feels like context and is usually a liability, because engagement rate is not one calculation. Quid, publisher of the report many people still know as Rival IQ, computes it as total interactions divided by follower count, and on that basis reports all-industry medians of 2.01% on TikTok, 0.30% on Instagram, 0.21% on YouTube and 0.03% on X. Socialinsider notes that engagement measured against reach is typically higher than the same content measured against followers, since only a fraction of followers see any given post. So a client whose dashboard reports engagement by reach can look like a runaway success or a disaster against the same benchmark depending only on which denominator each side used. If you quote a benchmark, quote the formula with it.

TikTok2.01%Instagram0.30%YouTube0.21%X0.03%
All-industry median engagement rates from Quid's 2026 Social Media Industry Benchmark Report, based on 150 companies sampled at random per industry from a database of 200,000+. Quid computes engagement rate as total interactions divided by follower count; a tool that divides by reach instead will report a substantially higher figure for identical content, so these are not comparable to a client dashboard without checking the formula.

What to do when the month was bad

The instinct is to lead with whatever went up, and clients read that instinct accurately. A bad month is better handled in a fixed order: state the result against the agreed number first, give the explanation if you have one, say so directly if you do not, and finish with a specific change and the date it takes effect. Specific means naming the format, the surface and the week, not promising to optimise. If the cause is genuinely unknown, commit to a diagnostic with a deadline rather than inventing a story about the algorithm, which is the most common way agencies lose credibility they cannot rebuild. One honestly reported bad month costs far less than a good-looking report the client later realises was arranged.

Write it so your contact can forward it

Your report is rarely read only by the person you send it to. The CMO Survey's 34th edition, covering 281 marketing leaders at US companies, found 63% reporting increased pressure from their CFO to prove marketing's value, 61% facing greater scrutiny from the CEO, and 50% from the board. Your contact is inside that, and the most useful thing you can give them is a document that survives being forwarded without them having to rewrite it. In practice that means a first page that states the agreed numbers, the result against each, and the decision you are recommending, with the platform detail behind it for anyone who wants it. Making your client look prepared to their own leadership is the most durable form of retention there is.

FAQ

What metrics should I report to a social media client?

Two or three, agreed before the work starts. A workable structure is one business outcome such as qualified enquiries or trials, one leading indicator such as clicks or saves that moves earlier, and one diagnostic such as retention or performance by format that explains changes. Impressions and follower counts belong in an appendix, not on the first page.

How often should agencies send social media reports?

Monthly is the standard. AgencyAnalytics' 2026 survey of 494 agency professionals found 69% report monthly, 11% weekly, 8% fortnightly and 3% quarterly. Organic social is too noisy week to week for shorter cycles to be meaningful, but pair the monthly report with a short fortnightly note so there is no long silence between them.

Should I include industry benchmarks in a client report?

Only with the formula attached. Quid's 2026 benchmarks calculate engagement rate as interactions divided by followers, while many dashboards divide by reach, which produces a much higher number for identical content. Comparing the two without saying so makes your results look either far better or far worse than they are.

How do you report a bad month to a client?

State the result against the agreed metric first, explain the cause if you know it, say clearly that you do not if you do not, and end with a specific change and a date. Avoid vague algorithm explanations and avoid leading with whatever number happened to rise. A client will forgive a bad month far more readily than a report that felt arranged.

Sources

Related pages