Rank tracking is simple to start and easy to set up in a way you regret three months later. The choices below are the ones that are annoying to change once history exists.
Add the keywords you are judged on first
Not the ones you would like to rank for. The ones that appear in the report someone reads.
History starts the day you add a keyword. There is no backfill — not here, not anywhere. A term added today has one data point today and a trend in a fortnight.
The settings that matter
Location. Rankings differ by country and city. Track where your customers are, not where you are. Getting this wrong makes every number plausible and useless.
Device. Mobile and desktop results genuinely differ. If your traffic is mostly mobile, track mobile. Tracking both doubles the cost, so do it deliberately.
Frequency. Daily for the terms you actively work on. Weekly for the long tail. Most keywords do not move enough day to day to be worth a daily check, and frequency is the main driver of what rank tracking costs you.
How many to track
Enough to see the shape, few enough to read. A hundred well-chosen keywords tell you more than a thousand you never look at — and the thousand costs ten times as much.
Group them with tags as you add them. Tagging by service line, funnel stage, or client is the difference between "rankings went down" and "rankings went down on the commercial pages while the blog held".
Reading the chart
- One check is not a trend. Two data points is a line, not a direction.
- Small movements are noise. Positions 8 to 11 and back is normal weather. Positions 4 to 19 is an event.
- A drop with no traffic change is often a SERP layout change rather than a ranking loss. Something took the click before the result did.
The Insights section connects these movements to your audit, your speed data, and your Search Console clicks, which is where "it moved" becomes "here is why".
Next
- From keyword idea to content task
- My rank tracker shows no positions yet