In short
- What a tracker gives
- A trend and an alert. Useful, and it answers "has it got worse" and not "because of what".
- What a fix list gives
- Per finding an address: the source, the page, the spot where it goes wrong. Something someone can work with.
- When measuring is enough
- If there is someone internally who can act on an address. If not, a dashboard is a subscription to worry.
What can one do that the other cannot?
Five questions, and on the first both give an answer. The divergence starts at the second, and the last row is there because neither product answers it — that remains a choice for someone who knows the business.
| On this question | A tracker | A fix list |
|---|---|---|
| Did I drop? | yes, with a chart and a date | yes, and on exactly which question |
| Because of what? | usually not — the cause disappears in the aggregation | the source that carried the answer, with its address |
| What should I do? | one recommendation per category | one action at a nameable place, with the reason next to it |
| Did it work? | the number moves or it does not | the same questions again, with the difference per finding |
| What neither of them does | decide whether this question is worth wanting to win | neither — that stays a human decision |
Why does the cause disappear during aggregation?
Because summarising by definition removes what was unique. A hundred answers each have their own sources; add them up into a percentage and one number remains while the hundred lists of addresses are gone. That is not sloppiness but a design choice, usually made because one number looks better in an overview.
The same form of loss sits in more places than people think. A list of sources reduced to domain names loses the thread that was actually about you; a list of search terms added into a total loses the term on which your position changed. The pattern is always the same: the field comes in, someone calculates it away, and the repair it could have produced never exists.
When is measuring alone enough for me?
If there is someone who can act on an address. A company with its own web editor and a developer is fine with a good measurement — there the translation from finding to action already exists. The bill for a fix service would then be paying twice for work already in place.
The reverse case is more common: one person doing marketing on the side gets a monthly alert that the figure fell and has no way in to change anything about it. That is the situation where a dashboard costs money every month and changes nothing.
Can I take the two separately?
Yes, and that is how it is set up here: the measurement exists on its own, and executing is a separate route with its own boundary. What does not exist is an in-between form where something gets published on your behalf without you having seen it. That boundary sits in the code and not in the terms.
Frequently asked questions
Can I start with a tracker and expand later?
Yes, and it is usually the sensible order. Measure for a quarter first, see what moves and which findings recur, then decide whether the work fits internally. Starting the other way round means buying execution for problems you have not measured yet.
Do I get an address for every finding, even when the source is not mine?
The address yes, the execution not always. If something wrong about you sits on someone else's site, you get the location and the text to write to them — the correction itself is human work at a third party, and we cannot compel it. That difference belongs in the report and not in a footnote.
Read on
Free, no account and no card.