Replacement Level Is Rising: what happens to businesses that don't adapt

Because the measurement is relative, and that is deliberate. An absolute ranking tells you where you sit; it cannot tell you whether sitting there is good. Comparing against an average business at the same distance can, and the cost of that honesty is that the baseline moves.

Concretely: a soil yard four miles out ranks sixth for the terms that matter. Nothing about it changes for two years. Over those two years, three of its competitors start asking every customer for a review, two fix the category on their Google Business Profile, and one publishes a product catalog an AI agent can read. The yard still ranks sixth on a good day, but an average business at four miles now ranks fifth. Its Rank Above Replacement has gone from positive to negative without a single thing happening on its own side of the ledger.

This is why the metric is named after baseball's wins above replacement rather than after a batting average. A player whose numbers hold steady for a decade while the league improves around him is getting worse at his job, and the statistic that says so is the useful one.

The dashed line is the replacement level, rising as competitors adopt. The solid line is you. The shaded gap between them is your Rank Above Replacement: teal while you are ahead of an average competitor, orange once you are behind.

Set your business to do nothing and watch where the lines cross. Then set it to adopt, and move the month. What the chart makes visible is the thing an owner cannot see from inside the decision: the curve bends upward with a lag, so the cost of waiting is not the months you waited but the months you waited plus the ramp.

Nothing on this list is clever. That is the point: the businesses pulling the replacement level up are not doing anything you could not do, they are doing it on a system instead of when they remember.

The third row is the one that has changed most recently and is least understood. Two years ago no local business published agent-readable surfaces, so it cost nothing to skip. That is no longer true, and it is covered on AI Agent Visibility.

A simulator that only argued for us would not be worth showing. Here is where it is weak, in the order the weaknesses matter.

1. It cannot predict your business. It projects one measured starting point forward under assumptions you set, and says so on the chart.
2. It assumes the field improves roughly evenly. In a market where one competitor does everything and the rest do nothing, the median moves more slowly than this.
3. It does not model a market shrinking or growing. Only relative position moves here, which is the whole point of a replacement-level metric.
4. The dollar figures inherit the market assumptions in the drawer. Change the search volume or cost per click and every dollar on the page changes with it.
5. Adopting is treated as one decision with one ramp. In practice it is a dozen pieces of work with different lead times, and some of them you will do and some you will not.

What the model is good for is the shape of the problem rather than the size of it: relative position decays without maintenance, recovery lags the decision, and both effects are measurable once a grid exists. If you want the size of it for your own business, the honest route is a measurement, not a slider. That is what Rank Above Replacement describes and what the diagnostic produces.