This section is about paid traffic: what you pay, what comes back, what broke this week, and which numbers you can trust at all. All eight templates run on one demo database: 26 weeks and 14 campaigns across search, social, and display advertising. This section covers 8 templates.
Every template is a live walkthrough on demo data: open it in the Templates section of Analytics Agent and press “Use this template” to run it on your own data.
Campaign Metrics Passport — Where We Stand on Every Basic Metric
Shows: a summary table for every campaign: the cost per thousand impressions, click, view, lead, and purchase, plus the revenue-to-spend ratio; flags on unusual values with a hint at what may be behind them; it also looks for campaigns that keep buying the same audience.
You get: all your advertising comes together in one picture, and you immediately see which campaigns waste money and which duplicate each other.
Channel Benchmarks — Are Our CTR/CPA/ROAS Good or Bad
Shows: your channels’ metrics compared with market ranges that you can replace with your own; each result shows how much headroom it has inside the norm, or how far it has gone beyond it.
You get: the answer to “is this good or bad” rests on numbers instead of debate, and the channels standing right on the edge show up in advance.
Ad Analytics Maturity — Which Step Is Next
Shows: a short audit across four stages — from manual spreadsheets to running on a forecast; your stage comes from answers about how the team works, and the conditions for moving up are measured against your own spend; every unchecked item is named with a concrete fix.
You get: you see which cleanup step will pay off at your current spend — and that it needs no budget.
Metrics Watchdog
Shows: a weekly check for metrics that disagree with each other — clicks are growing while sales are falling; impression cost is rising at the same reach; random noise is cut off by rules, and every divergence found comes with an action for Monday.
You get: small failures — junk placements, a swapped audience — surface the same week, not a month later in a summary report.
Optimization Speed
Shows: for every change in the ad account — a bid, a creative, an audience, a budget — whether enough data has already piled up to judge the result, and how much longer to wait if it has not.
You get: decisions are not made on three days of data, and at the same time the account does not go untouched where the data is already enough.
Creative Fatigue
Shows: how the response changes over the weeks of each ad’s life, and tells two different causes of decline apart: the audience is tired of the frequency, or the creative itself has stopped working; the decline comes with a creative replacement queue.
You get: budget does not stay in an ad that has stopped working, and money is not spent on replacement where the creative is not the problem.
Cookieless Tracking – Three Streams Reconciled
Shows: three sources of sales data reconciled — the ad account, server-side event transfer, and your own order log — at the level of individual events, showing how far each one diverges and what that gap costs in money.
You get: it is clear which number to trust and how much money each fix in data collection will bring back.
Incrementality – the Geo Lift Test
Shows: regions with ads and without them compared to see how many sales the ads really brought, not how many were credited to them; the ad account’s figure and an honest caveat about variance sit next to it; a separate table calculates whether an experiment can answer the question at all.
You get: it is clear whether the ads add sales or intercept people who would have bought anyway — and which decisions cannot yet be made on the data you have.