Reporting is where an agency relationship becomes visible. A good report helps you decide what to do next. A poor one exists to justify the invoice. Learn to distinguish them before you sign, by asking for a sample report.
Three layers of reporting
Strong reports separate three things that are often blurred together.
1. Activity. What the agency did: pages audited, content published, fixes shipped, outreach conducted. Necessary for accountability, but not evidence of results.
2. Leading indicators. Signals that the work is taking effect: pages newly indexed, crawl errors resolved, impressions and visibility for target queries, improved internal link coverage, page experience measurements.
3. Outcomes. What the business cares about: organic leads, orders, revenue, sign-ups or calls attributable to search.
A report that stops at layer one tells you nothing about value. One that jumps straight to layer three without explaining attribution may overclaim.
Metrics that matter
- Organic conversions or revenue, with tracking method explained.
- Non-branded organic clicks and impressions, separated from branded queries.
- Visibility for a defined set of priority queries or page groups, not a hand-picked list that changes monthly.
- Indexation and crawl health for important sections.
- Landing-page performance: which pages bring qualified visits.
- Technical health trends, tied to specific issues and remediation status.
Metrics that can mislead
- Total traffic without segmenting by intent and branding.
- Rankings for a small keyword list chosen for flattering results.
- Proprietary "authority" or "health" scores with no explanation of how they are calculated or why they matter to you.
- Number of backlinks with no regard to quality or relevance.
- Impressions alone, which can rise from irrelevant queries.
- Averages that hide a few high-volume pages driving the entire result.
None of these is useless; each needs context.
What a good report includes
| Section | Purpose |
|---|---|
| Summary in plain language | What happened, what it means, what is next |
| Outcomes vs agreed goals | Progress against measures defined at the start |
| Work completed | Specific deliverables and dates |
| Work planned | Priorities, reasoning, dependencies |
| Issues and risks | Blockers, changes in the market or platform |
| Data notes | Definitions, sources, known tracking gaps |
| Decisions needed from you | Approvals, developer time, content input |
Honest reports also include things that did not work and what was learned.
Attribution caveats
Organic search interacts with other channels. Users may discover you through search, return via email and convert later. Analytics tools attribute conversions by rules that differ between platforms. Ask the agency to explain which attribution model it uses and what it cannot see. Also ask how branded demand, seasonality and offline marketing are separated from SEO effect.
Agree reporting before work begins
Set these in the contract or statement of work:
- The goals and metrics that define success.
- The baseline period and how it is calculated.
- Reporting frequency and format.
- Data access. You should have direct access to your analytics and search console data, so you can verify the numbers.
- Meeting cadence for discussing results and priorities.
Red flags
- Reports delivered only as screenshots, with no access to underlying data.
- Frequent changes in which metrics are shown.
- Success defined after the fact.
- Reports with no mention of problems.
- Explanations that rely on jargon in place of evidence.
- Attribution of all organic growth to the agency, ignoring brand and market effects.
- Reluctance to give you ownership of dashboards and accounts.
Using reports to hold a meeting
A useful routine: review outcomes first, then leading indicators, then activity. For each major deviation from plan, ask for the hypothesis, the evidence and the proposed adjustment. If the same issues recur without change in approach, escalate or review the engagement terms; see Monthly Retainer vs Project Pricing for accountability mechanisms.
Validating claims before you hire
Agencies' own reporting examples are marketing. Ask to see a sample with client data redacted, and compare it with the standards above. On this site, listings show whether claims come from the agency's own site, a published case study or independent corroboration; sponsorship never buys verification. Treat reporting claims the same way: favor evidence you can inspect. For case-level evidence, see Evaluating an SEO Case Study.