Agencies urged to forecast SEO fixes before prioritizing work

The fastest way to lose a client’s patience is to let an SEO backlog grow without a clear answer on what each fix is worth. Adam Heitzman’s argument is straightforward: estimate the traffic lift before you prioritize the work, and the roadmap stops sounding like a wish list. That matters for agencies because the real job is not just shipping fixes, it is proving why one recommendation deserves budget now while another can wait.

Why forecasting beats instinct

Heitzman, the co-founder and managing partner of HigherVisibility, frames the problem the way agency teams actually feel it: developers want specificity, managers want a reason the roadmap is moving slowly, and clients want to know why traffic has not budged yet. The answer is not to defend every recommendation with gut feel. It is to attach a directional traffic forecast to the fix so the agency can separate a small, tidy task from the change that could move thousands of clicks.

That is a different way of running SEO operations. Instead of asking whether a task is technically interesting, you ask whether it changes the business enough to matter. Search Engine Land has been pushing that same idea in other technical SEO coverage too, especially around internal linking, where a high-value fix on important pages can affect millions in revenue while a dramatic-looking 404 problem may not justify the same urgency.

The five-step model agencies can actually use

The framework Heitzman lays out is built for prioritization, not theater. It gives teams a repeatable way to estimate likely impact before a request gets promoted from backlog to sprint. In practice, that means the agency can walk into a client meeting with a forecast that is conservative enough to be believable and specific enough to influence the roadmap.

The five steps are simple enough to run inside most agency workflows:

That model is especially useful when the backlog is crowded. It turns prioritization into a business conversation instead of a technical argument, which is exactly where agencies win or lose trust.

Use the data you already have

This kind of forecasting works because the basic inputs already exist in Google Search Console. Google’s Performance report centers on clicks, impressions, CTR, and average position, and those four metrics are enough to show whether a page is getting exposure but failing to convert that exposure into visits. Agencies do not need a perfect econometric model to start making better calls. They need enough signal to decide whether a fix is likely to recover meaningful traffic or simply make the site cleaner.

Structured data fits into the same logic. Google says it can make a page eligible for rich results, but it does not guarantee that rich result appearance will happen. That means a schema project should be forecast like any other SEO task: measure the pages involved, estimate the likely upside, and avoid treating eligibility as a promise. For agencies, that distinction matters because it keeps clients from assuming that every technical improvement will immediately turn into more clicks.

AI search has changed what “lift” even means

The warning buried inside this framework is that the SERP is no longer a stable target. Google’s AI features documentation now tells site owners to think about inclusion and performance in AI search experiences, which means forecasts need to account for more than traditional blue links. If visibility shifts inside AI Overviews or other AI-driven experiences, the old assumption that every improvement maps neatly to an organic click is too simple.

The click problem is already visible in the data. Pew Research Center found that users were less likely to click links when AI summaries appeared in Google search results, based on browsing data from 900 U.S. adults and 68,879 Google queries collected in March 2025. That does not make forecasting impossible, but it does mean agencies need to be more honest about the range of outcomes. The floor may be lower than it was, and the client conversation needs to reflect that reality.

There is also a regulatory pressure point. Reuters reported that the United Kingdom’s Competition and Markets Authority ordered Google to let publishers opt out of having their content used for AI search features. That is a reminder that AI search is not just a product issue anymore. It is shaping policy, publisher leverage, and the assumptions agencies make when they promise future traffic.

How the backlog changes when you forecast first

Here is where the playbook gets practical. Before forecasting, a small schema issue affecting a handful of pages might get priority because it looks neat, low risk, and easy to close. After forecasting, a broken title template that touches a much larger URL set can jump ahead because the traffic recovery case is simply stronger. That is the kind of swap agencies need when they are defending a roadmap to a client who wants to see business impact, not just a cleaner crawl report.

A simple agency-side version of that conversation sounds like this: the first fix is easy, but it only touches a few pages. The second fix is messier, but it affects thousands of clicks and creates a much better story for retained value. Once the forecast is attached, the agency can explain why the harder work belongs earlier in the queue, because the expected payoff is bigger and the recommendation is rooted in exposure, not instinct.

Why this is a retainer conversation, not just a ranking conversation

This is where forecasting stops being a technical SEO trick and starts functioning like agency operations. When you can explain which fixes are worth doing and how much traffic they might recover, it becomes easier to justify retainers, defend a roadmap in budget meetings, and keep clients from judging SEO by short-term noise. It also gives managers a cleaner way to talk to developers, because the work is now tied to impact rather than urgency alone.

The larger shift is obvious across the field. Agencies are being pushed to prove that technical SEO is not a pile of chores, but a set of revenue-relevant decisions. Forecast first, prioritize second, and the whole operation gets sharper: fewer wasted sprints, better client expectations, and a stronger case that SEO still earns its seat at the table.