Discussion
  • SEO Analytics

Which SEO metrics have you removed from executive dashboards because they drove the wrong decisions?

Published 1 reply

Discussion starter: an SEO measurement question, posted with an answer from Niraj Raut to open the thread. If you have dealt with this on a site, reply with what you saw, especially where it differs.

Disclosure: Niraj Raut, who posted this answer, runs the SEO consultancy linked at the end of it.

1 reply

  • admin
    Admin
    Original poster

    The metrics I have removed from executive dashboards share one flaw: they can move without the business changing, so leaders end up making decisions about the metric instead of the business. Sitewide average position, total impressions, indexed page counts, third-party authority scores, tool-reported keyword counts and unsegmented organic sessions are the usual candidates. Most of them are still useful to the SEO team for diagnosis. They just do not belong in front of people who will act on them without the context.

    The metrics and the wrong decisions they produced

    Metric Why it misleads at executive level Typical wrong decision
    Sitewide average position It is an average over whatever queries you appear for. Gaining visibility on new, lower-ranked queries makes it worse. Google’s removal of the &num=100 parameter in September 2025 changed reported impressions and average position for many sites without any real ranking change. Celebrating an “improvement” caused by lost long-tail impressions, or panicking over a “decline” caused by growth
    Total impressions Moves with query mix, SERP layout, AI features counted in the totals, and reporting changes. It is not demand. Treating a reporting change as a market shift
    Indexed pages More indexed pages is not better. Low-value URLs inflate it. Launching thin pages to “grow the index”
    Domain Rating, Authority Score and similar Vendor metrics based on the vendor’s own link crawl. They are gameable and are not used by Google. Buying links to hit an authority target
    Keywords in the top 10 (from a rank tracking tool) Depends on the tool’s keyword database, not your market. Irrelevant keywords count the same as revenue terms. Chasing easy keywords that never convert
    Total organic sessions Brand demand, driven by advertising and PR, swamps non-brand performance. Crediting SEO for a TV campaign, or blaming SEO when brand demand falls
    Content pieces published per month An output, not an outcome. Publishing volume at the expense of quality
    Core Web Vitals pass rate as a target on its own A useful quality threshold, but small changes around the pass line rarely move traffic in isolation. Spending engineering months on marginal gains while bigger technical issues wait

    What I replaced them with

    An executive dashboard should answer three questions: is organic search contributing more or less to the business, why, and what are we doing about it. That usually needs no more than six to eight numbers.

    1. Non-brand organic clicks to commercial pages, from Search Console, split by the product lines the business reports on. This is the closest measure of SEO performance that is not driven by brand demand.
    2. Organic revenue, leads or pipeline, split into brand and non-brand where possible, using the same attribution rules as other channels.
    3. Brand search demand, using Search Console’s branded queries filter or share of search. It shows whether the brand is growing and explains swings in brand traffic that SEO does not control.
    4. Coverage of priority pages: the share of the pages that matter commercially that are indexed and receiving impressions. It turns “indexed pages” into a question with a right answer.
    5. A small number of leading indicators tied to current projects, each with the decision it supports. For example, impressions for the new category pages launched this quarter.
    6. Incidents and risks: releases that caused SEO regressions, time to fix, and major risks such as a planned migration.

    A test for any metric on an executive dashboard:

    If this number moved 20% tomorrow, would leadership know what to do? If the honest answer is “ask the SEO team what it means”, it belongs in the team’s diagnostic dashboard, with the executive view showing the outcome it explains.

    How to remove a metric without looking evasive

    Removing a number people are used to seeing can look like hiding bad news, especially if it is removed while it is falling. What has worked for me:

    • Remove metrics when they are going up, not down, and explain why in a short note.
    • Show the old and new metric side by side for one or two reporting periods, with an example of where they diverged and what the new one got right.
    • Keep the removed metric available in the team dashboard and say so.
    • Annotate reporting changes directly on charts: the &num=100 change, the addition of AI Mode data to Search Console totals, analytics migrations, consent banner changes.

    Hypothetical example:

    An executive dashboard shows average position improving sharply in autumn 2025. The board asks why traffic has not risen with it. The SEO lead explains the reporting change behind it, replaces average position with non-brand clicks to commercial pages by product line, and adds brand search demand alongside. Six months later, when a price increase reduces brand demand, the board can see that non-brand performance held up and does not ask SEO to fix a pricing decision.

    Where the removed metrics still earn their keep

    None of the metrics in the first table are useless. They moved to the team’s working dashboard, where the people reading them know how they break:

    • Average position is useful per query or per tight query group, where the mix does not shift under you. It is a good diagnostic for a specific page or topic.
    • Impressions are useful for spotting coverage loss, especially filtered by page type or directory, and for early signs that a new section is being picked up.
    • Indexed page counts become useful when broken down into “pages we want indexed” and “pages we do not”, using the Page indexing report and your own URL inventory.
    • Authority scores help compare competitors’ link acquisition over time, as long as nobody sets a target on them.
    • Core Web Vitals belong in engineering quality dashboards alongside other performance budgets, not as a standalone SEO target.

    A good split is that the executive view has one row per business question and the team view has one tab per diagnostic area. When an executive metric moves, the team view should explain why within a day.

    AI visibility metrics: handle with the same care

    The newest candidates for executive dashboards are AI visibility scores and AI impressions. Search Console’s generative AI report gives impressions without clicks, and prompt-tracking tools report sampled appearance rates. Both are useful context. Neither should sit on an executive dashboard as a target without a clear note on method and a paired outcome metric, or you will repeat the average-position problem with a new number.

    The principle behind all of this: executive dashboards show outcomes and a few leading indicators with an obvious action attached; everything that needs interpretation lives with the team that can interpret it.

    Need help with this on your own site?

    Niraj Raut works with businesses in Nepal, Australia, the UK, Europe and the US on technical SEO, ecommerce SEO, local SEO and AI search.

    See SEO services

Add a reply

Replies are open to approved members. Log in or apply to join.