The Monthly Agency Reporting Workflow That Drives the Next 30 Days of Decisions
A 3-day monthly reporting sprint structured around Daniel Murphy’s three questions: what we tried, what worked, what we are doing next.
Turn this page into a campaign built for your brand.
Connect Superdirector to ChatGPT and it adds current social-video references, public-video analysis, Brand DNA, and a ready-to-send campaign brief.
By Bell Chen, founder of Superdirector — how the analysis works.
The reporting shape leadership actually reads comes down to three questions that have to anchor any decision-driving social report: what we tried, what worked, what we’re doing next, in that order. That single sequence is the agency-reporting workflow restated.
Every monthly report a social agency sends to a client either answers those three questions or it is a metric dump. From the founder side, where I have been the client receiving these reports, the structural difference between the reports the client paid attention to and the ones the client filed without opening was whether the three questions sat above the metrics.
In my experience watching agency relationships succeed and fail through the first 90 days at Backlinker AI and during the Superpencil launch window in February 2026, the monthly report is where renewal is won or lost. The retainer-pricing math works only if the client believes the agency is making the brand better month over month. Belief is not built by the dashboard. Belief is built by a report that names three decisions the client will make this month because of what last month showed.
What this workflow actually solves
The 2026 problem with agency reporting is not data access. Every agency I have audited can export views, engagement rate, saves, shares, follower growth, and post-level performance from Meta Business Suite, TikTok Creative Center, and a connected analytics dashboard inside 20 minutes a month. The problem is that the export is the report.
Most monthly agency reports I have seen in 2026 are 18 to 26 pages of metric grids with a one-paragraph executive summary on the cover. The client opens the PDF, skims the cover, and closes the file. The report cost the agency 6 to 9 hours to produce and changed exactly nothing about how the client thinks about the next 30 days.
Most social teams report their work on a regular weekly or monthly cadence, but far fewer produce reports that drive specific next-month decisions. That gap is the entire reporting problem. The cadence is being honored. The reporting job is not.
The second cost is renewal. In my experience watching three agency relationships die in 2026 between days 75 and 110, the death sequence was the same. The agency shipped monthly metric dumps. The client read the executive summary. The client did not act on anything in the dump.
By the fourth monthly report the client had no specific evidence the agency was learning, and the renewal conversation became a justification exercise. The agencies that retained past 12 months wrote reports the client CMO pasted slides from into their own quarterly board update. That paste is the moment the report becomes valuable inside the client organization independent of the agency being in the room.
The third cost is the agency’s own learning loop. A reporting workflow that ends in a PDF the agency itself never re-reads is a workflow that does not compound. The agencies I have seen learn fastest are the ones whose principal account leads keep a single rolling document per client across all 12 months of reports, with the three-question shape repeating, so by month 9 the agency can quote month 2’s recommendation back at the client and demonstrate the iteration.
The named-operator playbook
Daniel Murphy, B2B social operator
Three-question shape: what we tried, what worked, what we are doing next
The three-question shape (what we tried, what worked, what we’re doing next) is the load-bearing structure of the entire workflow. Every section of the monthly report belongs to one of the three questions, in order. What we tried is the experiment log: the formats we shipped, the hooks we tested, the new platforms we touched, the production choices we made differently than last month.
What worked is the evidence section: format-level performance benchmarked against the client’s last-90-days baseline, not a generic industry baseline. What we are doing next is the decision section: three specific next-month commitments the client is approving by signing off on this report.
Rachel Karten, formerly social media lead at Bon Appétit, now Link in Bio
100,000 subscribers, Tuesday read for in-house and agency SMMs
Karten was the social media lead at Bon Appétit before she built Link in Bio into the canonical Tuesday read. A useful reporting principle is that the reader does not need to know every line of your spreadsheet, only the broad strokes. The agency-reporting application is the visible architecture of the report. The executive summary on page one is the broad strokes the client CMO can quote in their own board update. The detail-level appendix is the line-by-line spreadsheet the client social manager will use to operationalize the agency’s next-month commitments.
Kendall Hope Tucker, Ramp head of creative experimentation
Glass-box livestream, the brand-twist filter
Ramp's glass-box livestream campaign became a widely cited example of one conceit executed many ways. The transferable rule is to borrow the recognizable frame but always carry a brand-specific twist. The agency-reporting application is the recommendations section. Every next-month commitment the agency makes in the report has to be framed as the brand-twisted version of a category insight, not as a generic best-practice suggestion.
Adam Mosseri, head of Instagram
The three top ranking signals: watch time, likes per reach, sends per reach
In a January 21, 2025 Reel, Mosseri named Instagram’s three top ranking signals as watch time, likes per reach, and sends per reach. The agency-reporting application is the format-level analysis section. Reporting on raw views alone is reporting on a top-of-funnel signal Instagram has explicitly downranked relative to the watch-time and sends signals. The monthly report should benchmark each top-five performing post against the client’s own last-90-days baseline on all three Mosseri signals, not against a generic industry engagement-rate benchmark.
Format half-life
TikTok formats decay fast; treat first-appearance dates as the clock
Format half-life across Instagram, TikTok, LinkedIn, Threads, and Bluesky is the data the agency reporting workflow has to absorb. TikTok formats in particular decay fast. The agency-reporting application is the time horizon on next-month commitments. A recommendation to invest the next 30 days in a format that peaked two weeks before the report is a recommendation to ship a post that is already past its half-life. The reporting workflow has to date-stamp every recommendation with the format first-appearance window.
The monthly reporting sprint (audit-grade time blocks)
The sprint assumes a calendar-month reporting cadence, with the report delivered to the client by the third business day of the following month. Total agency-side time investment: 4 to 5 hours of principal account lead time, plus 90 minutes of analyst or junior strategist time for the data pull. The investment is roughly the same as the metric-dump workflow most agencies run. The reorganization is the lift, not the total hours.
Day 1 morning: data collection and platform pull (20 minutes, analyst time). Export the previous month performance data from each platform. Pull views, watch time, likes, sends, saves, shares, follower growth, and post-level performance for every shipped post. Focus on the metrics the client agreed to during onboarding. Do not export everything. The export goes into a single tab of the rolling client analytics spreadsheet the agency maintains across the full retainer.
Day 1 afternoon: competitive benchmark refresh (45 minutes, analyst time). Re-run the competitive scan from the agency-client onboarding sprint against the three brands the client wants to take share from and the three the client admires. The benchmark refresh produces one number per competitor: did the competitor category-leading format ship this month, and what was its observable performance versus their own 90-day baseline.
Day 1 to Day 2: format performance analysis (60 minutes, principal account lead). Rank every shipped post for the month against the client’s own 90-day baseline on Mosseri’s three signals. The output is a one-page table: top 5 posts, bottom 3 posts, the format pattern across both groups, and the one observation about the client audience the month surfaced. The pattern is the load-bearing analytic insight of the report.
Day 2 morning: client-specific narrative draft (90 minutes, principal account lead). Write the report against Daniel Murphy’s three-question shape. The what-we-tried section names the four to six format and hook decisions the agency made the previous month. The what-worked section presents the format performance analysis with the named pattern and the competitive context. The what-we-are-doing-next section commits the agency to three specific decisions for the next 30 days. Each commitment names the format, the production estimate, and the success threshold. Three commitments, not seven.
Day 2 afternoon: strategic recommendations and brand-twist filter (45 minutes, principal account lead). Every recommendation in the what-we-are-doing-next section gets pushed through Tucker’s brand-twist filter. If the recommendation could be lifted into any other client report without changing a word, the recommendation is generic and gets rewritten. The brand-twist version cites the specific client signal from the discovery call or a recent client conversation that motivates the recommendation.
Day 3 morning: report assembly and review (45 minutes, principal account lead). Assemble the report in a single shared document with the executive summary on page one, the format performance analysis on page two, the what-we-are-doing-next commitments on page three, and the appendix tables for the client social manager on pages four through six. The broad-strokes rule governs the page-one executive summary. Six bullet points at the top of the page. No more.
Day 3 afternoon: client review call and decision capture (60 minutes, principal account lead). Schedule a 45 to 60-minute review call with the client. Run the call against Murphy’s three-question shape. Spend 10 minutes on what-we-tried, 20 minutes on what-worked, and 30 minutes on what-we-are-doing-next. Document each decision with a one-line acceptance criterion, an owner, and a due date that lands inside the next 30 days. Send the decision capture sheet to the client by end of Day 3. The decision capture sheet, not the PDF, is the artifact the agency operates against for the rest of the month.
| Row | What it shows | Decision it forces |
|---|---|---|
| What we tried | The post-level record against the account baseline | What to keep producing |
| What worked | The format pattern across the top and bottom performers | What to double down on |
| What we are doing next | The named next-period decision and its owner | The commit that makes the report worth opening |
A worked example (Vespera Skin, fictional)
The brand is Vespera Skin, a hypothetical direct-to-consumer skincare label. Contract signed June 1, 2026. Monthly report cadence: third business day of the following month. The principal account lead is the agency senior strategist; the analyst is the agency junior production-coordinator.
First Monday of July, 9am Pacific. The analyst pulls the June export from Instagram Business Suite and TikTok Analytics. Vespera shipped 19 posts in June: 8 founder-on-camera ingredient explainer Reels, 4 behind-the-scenes formulation Reels, 4 carousels, and 3 customer-reaction stitch Reels. The competitive scan re-run shows Topicals, Glossier, and Ami Colé each shipped a founder-on-camera explainer format with sends-per-reach ranging from 2.4 to 3.8 percent.
First Monday afternoon. The principal account lead runs the format performance analysis. The top three posts of the month for Vespera are all founder-on-camera ingredient explainers, with sends-per-reach between 2.7 and 3.1 percent versus the client’s 90-day baseline of 0.9 percent. The bottom two posts are carousels, with sends-per-reach of 0.3 and 0.4 percent. The named pattern is that Vespera’s audience saves and shares Reels in which the founder narrates a single named ingredient over 45 to 80 seconds, and does not save carousels regardless of design quality.
First Tuesday morning. The principal account lead drafts the report. What-we-tried: 19 posts across 4 formats, with the explicit decision to test carousels for the first time as a save-and-return format for educational content. What-worked: the founder-on-camera ingredient explainer format, which produced 3 of the top 5 posts at 2.7 to 3.1 percent sends-per-reach versus the 0.9 percent baseline.
What-we-are- doing-next: three commitments. Commitment one is to retire the carousel slot for the next 30 days and reinvest the production hours into 6 founder-on-camera ingredient explainer Reels. Commitment two is to test one behind-the-scenes formulation Reel with the Vespera-specific twist. Commitment three is to publish the first of a six-part Brian’s-Office-style sustained format featuring the founder morning routine.
First Wednesday morning. The report assembles in a six-page Google Doc. The executive summary page reads, in full: “June shipped 19 posts across 4 formats. Single-ingredient founder Reels drove 3.1x the 90-day sends-per-reach baseline. Carousels underperformed. July retires carousels, doubles single-ingredient Reels, and launches a sustained founder-morning-routine format. Three commitments below. Production estimate $1,720 inside existing $1,800 monthly working budget.” Six lines.
First Wednesday afternoon at 2pm Pacific. The review call runs 47 minutes. The founder approves all three commitments. The agency captures the decision in a one-page sheet, cross-links it to the creative brief workflow, and Vespera’s part-time editor receives the production calendar for the next 30 days by end of business Wednesday. The agency has spent 4 hours 50 minutes on the report. The client has spent 47 minutes plus the time to read a one-page executive summary. The retainer is structurally healthier at month one than most agency relationships are at month three.
Where this typically breaks
The metric-dump failure. The report is 18 to 26 pages of platform export grids with a one-paragraph executive summary. The client opens the PDF, skims the cover, files the document. The report cost the agency 6 to 9 hours and changed nothing about the next 30 days. The fix is the Murphy three-question shape, per Murphy. Every page of the report belongs to one of what-we-tried, what-worked, or what-we-are-doing-next. The platform export grids belong in the appendix, not the body.
The generic-recommendations failure. The report recommendations could be copy-pasted into any other client report without changing a word. The client reads them, recognizes them as generic, and concludes the agency is not paying attention to their specific brand. The fix is a brand-twist filter, the one Ramp's glass-box livestream illustrates: borrow the recognizable frame, but always with the brand's own angle. Every recommendation has to cite a specific client-only signal from the discovery call or a recent conversation.
The vanity-metric failure. The report leads with raw view counts, follower-growth percentages, and the agency’s favorite three viral posts of the month, with no benchmark against the client’s own 90-day baseline. The client reads the numbers, has no context for whether the numbers are good, and assumes they are average. The fix is the Mosseri three-signal benchmark. Every post-level number in the report sits next to the client’s own 90-day baseline on watch time, likes, and sends-per-reach. Post half-life raises the stakes: a viral spike on a trend format decays within days, while steady output on a sustained format compounds, and the report has to make that distinction explicit.
Where a planning-first tool fits
Most planning-and-feed tools, including Superdirector, are built around the planning side of the monthly cycle, not the reporting side. The monthly report itself is produced primarily from platform exports and the agency’s own rolling spreadsheet, regardless of which planning tool the agency uses.
The relevant question for an agency adding a tool to the reporting stack is whether the tool compresses the data-pull and competitive-benchmark steps from the working 65 minutes down to 30 minutes without degrading the depth of the format-pattern analysis. If the tool does that, it pays for itself across 12 monthly reports a year. If the tool produces a slicker dashboard but the principal account lead still has to write the three-question narrative manually, the math reverses. The report’s load-bearing artifact is the three-question narrative, not the dashboard underneath it.
Drive Decisions With Your Monthly Report
Paste your Instagram or TikTok profile URL to get a niche reference feed, then generate brand-fit scripts and shot plans from the same workflow.