Takes the raw metrics from your Meta ads (spend, ROAS, CPA, CPM, video play rates at 25/50/75/100%, funnel cost cascade, frequency, ad age) and runs a structured strategic analysis that ends with two specific answers:
"Where should the next best dollar go?" — the highest-ROI allocation of incremental spend, with specific ads to scale, kill, or monitor and exact percentage budget increases.
"Where should the next best week of focus go?" — the primary constraint blocking scale (Creative / Efficiency / Scale / Budget Pacing / Offer-Market) and the top 3 actions for the next 7 days.
This is the prompt I reach for when I'm looking at a spreadsheet of ad performance and I'm not sure what to do next. Or when I'm coaching an operator who hasn't built the pattern recognition yet for reading Meta data and needs the strategic interpretation done for them.
The framing Brandon flagged when he sent it: "helps if you aren't the best at understanding how Meta works." Built specifically to bridge the gap between "I have the numbers in front of me" and "I know exactly what to do next."
Note: the prompt is built around video ad metrics specifically — it uses Hook Rate, video retention curves at 25/50/75/100%, and drop-off analysis as its creative diagnostic layer. The strategic framework (spend distribution, efficiency analysis, funnel diagnostics, constraint identification) translates directly to static image ads, but you'll skip the video retention sections when you run it on a static campaign.
What it analyzes
Tier 1: Spend Signals — Meta's capital allocation vote
Reads what Meta's already telling you through its budget allocation behavior. Amount Spent, Amount Spent %, Spend vs Budget utilization, Current Delivery status (Active / Learning / Learning Limited / Inactive), and ad age fatigue buckets. Built on the post-Andromeda reality — Meta's algorithm has already processed more data than any operator can. When it allocates spend, it's voting. Read the vote before fighting it.
Tier 2: Efficiency — ROAS and CPA against targets
Segments every ad into Exceeds / Meets / Below / Failing buckets against your target ROAS and CPA, then calculates what percentage of total spend is flowing to above-target ads. The healthy target is >70% of spend on above-target performers. Anything less is a signal you have a leak.
Tier 3: Full funnel diagnostic — where the leak actually is
Maps the cost cascade from CPM → Cost Per Click → Cost Per Landing Page View → Cost Per Add to Cart → Cost Per Checkout → Cost Per Purchase. Calculates conversion rates between each stage. Identifies the specific weakest funnel step for each ad and diagnoses the root cause (creative resonance / page speed / page-ad congruence / cart friction / checkout abandonment).
Tier 4: Creative performance — attention and engagement curves
Hook Rate (25% / Impressions), 25%→50% retention, 50%→75% retention, 75%→100% retention, and completion rate for every video ad. Identifies whether the drop-off is a hook problem, a middle-content problem, a pitch/offer problem, or a CTA problem — each diagnosis points to a completely different fix.
Tier 5: Constraint identification — the one thing blocking scale
Five constraint diagnoses, with specific symptoms for each:
- Creative Constraint — spend concentrated in 1-2 ads, new ads not getting spend, winners fatiguing. Fix: new creative concepts, not more budget.
- Efficiency Constraint — spend allocated but ROAS below target, funnel dropoff at a specific stage. Fix: fix the broken funnel stage.
- Scale Constraint — efficient ads but spend limited, high frequency, small audience. Fix: audience expansion / new placements / adjacent persona angles.
- Budget Pacing Constraint — ads underspending budget, bid/cost cap limiting delivery. Fix: review bid strategy, loosen caps.
- Offer/Market Constraint — multiple concepts tested, none working, high CPMs across the board. Fix: offer development, not more creative.
The prompt identifies the ONE primary constraint. Everything else is secondary. This is the most important output — because if you fix the wrong constraint, nothing improves.
Final deliverables
- Executive summary (1 page)
- Ad-by-ad scorecards with SCALE / OPTIMIZE / MONITOR / KILL recommendations
- Specific dollar allocation recommendations (which ads to increase, by what %, freeing how much budget from kills)
- Specific next-week focus areas ranked by impact
- Pattern analysis (what's working across winners, what's not across losers)
- Benchmark reference tables for context
How to use it
Open the GDrive link. Copy the prompt. Plug it into Claude (or your model of choice).
Then feed it the inputs the prompt asks for. Critical: the prompt requires real context, not just metrics. It explicitly asks for:
- Brand & product context — what you're selling, AOV, target audience, mechanism
- Campaign context — campaign objective, bid strategy, optimization event, total budget, start date
- Performance targets — target ROAS, target CPA, break-even, current business priority (scale vs efficiency)
- Creative context for each ad — hook, message angle, format (UGC / talking head / B-roll), talent, CTA, length
- Ad metrics data — paste a CSV, screenshot a spreadsheet, or paste a table. Required columns documented in the prompt.
Without the brand/campaign/target context, the analysis will be generic. With it, the recommendations are sharp.
Recommended workflow: export your Meta Ads Manager view to CSV → paste the CSV → answer the 4 context sections → run the prompt → get the full analysis with executive summary, ad-by-ad scorecards, and Next Best Dollar / Next Best Week recommendations.
When to use it
Use this prompt when:
- You're staring at a Meta Ads Manager view and you're not sure which ads to scale, which to kill, and which to leave alone
- You're spending real money ($5K+/month) and need a structured second opinion on allocation decisions before you commit a budget shift
- You're coaching a team member or junior media buyer who has the data but doesn't yet have the pattern recognition to interpret it strategically
- A campaign that was working is starting to slip and you need to diagnose what changed before throwing more budget at it
- You're hitting a scaling ceiling and need to identify whether the constraint is creative / funnel / audience / offer before you spend the next week working on the wrong thing
- You're handing off campaign management and need a documented analysis the next operator can pick up
Especially useful for:
- Operators who run their own brand and don't have a senior media buyer to call when the data gets confusing
- Agencies who need to produce structured client reports without spending 4 hours per account on manual analysis
- Teams running 3+ brands where deep weekly analysis on every account isn't sustainable manually
- Anyone who knows the metrics matter but hasn't built the strategic framework for translating them into specific decisions yet
Not for:
- Static image ads (you can use the framework but skip the video retention sections — gatekept has a dedicated image ad version)
- Sub-$1K/month spend levels where you don't have enough data volume for the analysis to be statistically meaningful (run for a longer time period or use the brief generators instead)
- Pre-launch decisions — this is post-launch analysis. For pre-launch test sequencing use the Brief & Concept Tracking System in Module 5
- Real-time intraday decisions — this is built for weekly cadence reviews, not minute-by-minute optimization
The prompt
Open in GDrive → docs.google.com
Copy → paste into your model → fill in the brand/campaign/target context sections → paste your ad metrics CSV or table → run.
Cross-references
- Module 4 → Different Ad Account Strategies — once the prompt identifies which ads to scale, this module covers the multi-account architecture for actually deploying that scale without triggering Meta's duplicate detection
- Module 4 → Blended NCPA — the efficiency metric that sits one layer above the prompt's ROAS/CPA analysis. The prompt diagnoses individual ad performance. Blended NCPA tells you whether the portfolio is profitable
- Module 4 → Split Tests — the prompt's "Next Best Week" recommendations frequently surface as split test hypotheses. This module covers how to structure those tests so the data is interpretable downstream
- Module 6 → Reading & Interpreting Data — the strategic frame for data interpretation as a discipline. This prompt automates a specific slice of that frame. The module is the underlying philosophy
- Module 6 → Sales Campaigns in Meta — the campaign architecture the prompt's analysis is built on. The prompt assumes you understand the architecture; this module is where you learn it
- Module 8 → Microsoft Clarity — complementary funnel diagnostic tool for the landing page layer where Meta's data ends. When the prompt diagnoses an Efficiency Constraint at the LPV→ATC stage, Clarity is where you go to actually see what's breaking on the page
- Plug & Play → Dialed Native Ad Prompts / Dialed Image Ad Prompts / Emotional Valence Prompt — the creative production side. When this analysis prompt identifies a Creative Constraint as the bottleneck, those are the prompts you reach for next to build the new creative
Want the full data analysis prompt library + Loom walkthroughs of real campaigns + live updates as Meta evolves? All inside the RunItUp Masterclass → Join here
Join now to lock in your rate, prices are increasing quickly.