Honest SEO KPIs: beyond rankings and traffic
Average position and organic sessions became vanity metrics. These KPIs tie SEO to revenue and survive a CFO review.
Average position climbed from 14 to 9, organic traffic grew 22%, and the board still cut the SEO budget in Q2. That scenario played out three times in 2025 with different clients, and it exposes an uncomfortable truth: rankings and sessions do not pay payroll. When a CFO stares at the GA4 dashboard and cannot connect a rising curve to a line on the P&L, investment becomes expense. Honest SEO starts here, picking indicators that survive a finance grilling, not just a marketing standup.
The first KPI I retire in any kickoff is the famous 'organic sessions'. A session without qualification measures curiosity, not intent. Instead, I instrument 'sessions by search intent', crossing Search Console queries with the taxonomy I described in Search intent: 4 types and how to map them on the SERP. At a B2B SaaS client, 61% of organic traffic came from informational queries with zero correlation to pipeline. We cut focus on those pages, redirected editorial effort to commercial and transactional queries, and MRR attributed to organic rose 38% in 90 days without touching total volume.
The second block is assisted revenue with an honest window. Last-click attribution kills SEO because organic usually opens the journey, it does not close it. I configure data-driven models in GA4 with a 30-day window and cross them with content cohorts, replicating the framework from Cohort analysis applied to organic content. For e-commerce clients, I use organic ROAS calculated as assisted revenue divided by editorial cost + technical cost + link building cost. Ugly number, defensible number. I broke down the full logic in SEO Attribution: Proving ROI Without Last-Click, including how to present the model to a skeptical CFO without losing credibility.
Technical efficiency KPIs enter the scorecard with smaller weight, but they exist. 'Share of indexed pages generating at least one session in 90 days' says more about site health than raw Search Console coverage. In a recent audit, 47% of a marketplace's indexed URLs had zero sessions, eating crawl budget and diluting authority. I solved it via selective noindex and PLP consolidation, following what I wrote in Crawl budget: when to worry and how to measure it. The win did not show up in average position, it showed up in revenue from PDPs that started receiving real Googlebot crawl.
Qualified engagement is the fourth pillar. Bounce rate died with GA4, but 'engaged sessions with scroll above 75% and average time above 90 seconds' remains a decent proxy for dwell time, as I discussed in Dwell time: measuring engagement without official data. I combine that signal with micro-conversion completion rate: PDF downloads, secondary CTA clicks, pricing page views. A fintech client had 8% organic CTR but 0.3% final conversion. The problem was not traffic, it was a title promise misaligned with content, something I covered in Title tags that convert: 7 patterns tested on real SERPs. An honest title cut traffic 12% and doubled conversion.
The fifth KPI is share of voice weighted by potential revenue. Tracking 200 keywords without commercial weight produces pretty reports and bad decisions. Instead, I assign every query an estimated value, multiplying monthly volume by expected CTR at current position (I use the benchmark I maintain in CTR benchmark by position: updated 2026 data) and by segment LTV. Sum those and you have 'potential organic revenue' as a single number for the board. When that KPI rises and traffic drops, you are capturing more expensive keywords, exactly the goal. When it drops and traffic rises, red alert: you are diluting into irrelevant queries.
Closing, the most ignored KPI: cost per incremental organic lead, isolated from brand. I filter branded queries from Search Console via regex, calculate organic CAC over non-brand only, and compare it to paid CAC in the same period. In 2025, clients who ran this math discovered SEO delivered leads 4 to 11 times cheaper than Google Ads, a data point that justified budget increases instead of cuts. Practical takeaway: build a scorecard with five KPIs, sessions by intent, assisted revenue, productive pages index, qualified engagement, and non-brand organic CAC. Present it monthly in dollars, not in positions. Rankings go back to being a means, not an end.