SEO Strategy & Stakeholder Reporting

An audit tells you what is wrong. A strategy tells you what to do about it, in what order, and why it is worth the money. Reporting closes the loop: it shows what happened, explains why, and asks for the next decision. This guide owns strategy and communication; the measurement mechanics (KPI tiers, Search Console caveats, the forecasting model) are in Analytics & Forecasting.

Strategy vs tactics#

A tactic is "add internal links to the pricing page". A strategy is a short argument for where you will win and why:

1. Goal: the business outcome, in the business's units ("grow self-serve signups from organic by 30% in 12 months").
2. Audience and demand: who searches, for what, at which stage. Your keyword research and search intent work go here.
3. Advantage: what you have that competitors cannot easily copy: original data, product expertise, a community, a brand people already search for. In 2026 this matters more, because Google's guidance for AI features stresses "non-commodity" content, and generic pages are the ones AI answers replace.
4. Bets: 3 to 5 themes that follow from the above, such as "own comparison queries for our category" or "fix crawl waste on 2M filter URLs".
5. Constraints and risks: team size, platform limits, AI answers reducing clicks for informational queries.

If a proposed task does not serve one of the bets, it goes in the backlog, not the roadmap.

The roadmap: impact, confidence, effort#

Turn audit findings, content gaps and ideas into a scored list. Rate each 1 to 10:

• Impact: how much it moves the goal if it works (traffic × conversion value, or pages affected × severity).
• Confidence: how sure you are it will work: your own test results and Google's documentation score high, a blog post's correlation study scores low.
• Effort: person-days across every team involved, including review and QA.

score = impact × confidence ÷ effort
 
item                                  I   C   E   score
Remove noindex from product template  9  10   1    90.0
Fix CDN block on AI search bots       6   8   1    48.0
Comparison pages for top 10 rivals    8   6   6     8.0
Rebuild faceted navigation            9   7  12     5.3
"Chunk" every article for AI          3   2   8     0.8

The last row is there on purpose. Google's AI optimization guide (May 2026) says there is no need to break content into tiny pieces for AI, so confidence is low and the score sinks it. Scoring makes that conversation about evidence rather than opinion.

Then group the top items into quarterly themes with owners and a definition of done. A roadmap people follow fits on one screen: themes, items, owner, quarter, status.

The business case and forecast caveats#

Use the bottom-up model in Analytics & Forecasting and present three scenarios. When you present it, say out loud what can make it wrong:

• AI answers change the click curve. Pew Research found in March 2025 data that users clicked a traditional result on 8% of visits to a results page with an AI summary vs 15% without (published 2025-07-22). Your CTR assumptions for informational queries should be lower and wider than they were in 2023.
• Core updates are not schedulable. In 2026 alone Google confirmed core updates in March and May. A forecast is a direction, not a promise on a date.
• Tool volumes are estimates. Haircut them and say so.
• Delivery risk is the biggest variable. A forecast assumes the work ships. Tie each scenario to what gets built.

Never present a single number. Present a range, the assumptions, and the date you will compare forecast with actuals.

Reporting by audience#

A useful executive structure is what happened, why, what next:

## Organic search, Q3 2026
 
**What happened:** Non-branded organic signups 1,240 (target 1,100, +13%).
Branded clicks flat. AI Overview/AI Mode impressions up 22% (Search Console).
 
**Why:** Comparison pages shipped in July now drive 31% of non-branded signups.
Informational traffic -9% as AI answers absorb simple queries (expected, in forecast).
 
**Next:** Ship 6 more comparison pages (Q4). Decision needed: approve
2 dev days to fix the product template canonical bug (est. +5-8% product clicks).

Branded vs non-branded: always split#

Branded demand follows advertising, PR and product launches. Non-branded demand is where SEO work shows. Reporting them together lets a TV campaign take credit for SEO, or hides an SEO decline behind brand growth.

Use Search Console's branded queries filter (Performance report, Query filter, Branded or Non-branded; all eligible sites since 2026-03-11). It uses AI classification, catches misspellings and brand products, and has an Insights card for the branded share. It is not available for sub-properties or low-volume sites, so keep a regex fallback for those. Report both lines, and lead with non-branded when you talk about SEO's contribution.

Handling data breaks#

Every long chart has breaks. Name them before someone else misreads them:

• &num=100 (mid-September 2025). Google stopped honouring the 100-results parameter around 2025-09-12. Rank-tracker impressions vanished from Search Console: Locomotive's study of 319 properties found 87.7% saw impressions decline. Impressions fell and average position "improved" overnight. In year-over-year reports through September 2026, compare clicks, not impressions or average position, and footnote the break.
• Google updates. Mark confirmed core and spam updates from the Search Status Dashboard. See Algorithm Updates.
• Your own changes. Migrations, tracking changes, consent banners and new GA4 channels (the AI Assistant channel started in May 2026 and is not retroactive, so AI traffic "appears" that was previously in Referral).

Put annotations in Search Console (right-click the Performance chart, 120 characters, visible to all property users since November 2025) so the explanation travels with the data.

AI visibility in reports#

Stakeholders will ask "are we in ChatGPT and AI Overviews?" Answer with first-party data first, estimates second:

Frame AI visibility as reach, alongside clicks, not as a replacement for them. "We appeared in 18,000 AI Overview and AI Mode impressions and were cited in 9 of 25 tracked ChatGPT prompts" is honest. "Our AI ranking is #1" is not a thing. Measuring AI Visibility covers the methods.

Common reporting lies (and the honest version)#

• Cherry-picked date ranges. Pick the comparison window before you look, and keep it consistent month to month.
• Totals without segments. "Organic up 20%" when branded is up 60% and non-branded down 5%.
• Average position as a headline. It moves with your query mix and data breaks, not just rankings.
• Taking credit for seasonality. Compare year over year, not just month over month.
• Silence about misses. Report what did not work and what you learned. It is what makes the wins believable.

Templates to reuse#

• Strategy page: goal, audience, advantage, 3 to 5 bets, constraints (one page).
• Roadmap sheet: item, theme, impact, confidence, effort, score, owner, quarter, status.
• Exec report: what happened, why, next, decision needed (one page).
• Team dashboard: branded and non-branded clicks, clicks by template, indexation, CWV, AI impressions, AI referrals, annotations.

Practise all of this on a real site with the Capstone: Audit Your Own Site: its deliverable is the starting point for your first roadmap.