Why do search rankings jump in steps instead of climbing gradually?

IEO Engine Research · Published 2026-07-11 · Measured from production server logs
Because evaluation is not continuous. Ranking movement on a new or re-assessed site arrives in discrete steps separated by flat intervals — the site sits still for days, then moves several positions at once with no corresponding change on your end. The flat period is not failure. It is the interval between evaluations, and judging a content push during one will tell you the opposite of the truth.

What the pattern looks like

Plotted from Google Search Console data across multiple deployments, average position does not describe a smooth curve. It describes a staircase: long flat treads, then a sharp riser. Impressions frequently jump on a day when nothing was published, which is the signature of a scheduled re-assessment landing rather than of anything you did that day.

Why this breaks most measurement

The standard practice — publish, wait seven days, evaluate — has a high chance of landing entirely inside a flat tread. The conclusion drawn is 'that content did nothing,' and the content is abandoned days before the riser arrives. This produces a systematic bias against exactly the slow-compounding work that eventually moves.

What to measure instead

Step timing, not daily position. Record when risers occur and how far apart they are. The interval is the useful number, because it tells you the minimum honest evaluation window for your own site — which is almost certainly longer than seven days.

Primary source: this answer is drawn from production access logs across four live deployments, not from vendor documentation or third-party tooling.
Full data and method: read the underlying field note →
All field notes: IEO Engine Research →