How to allocate a backlink budget

Fewer than the competitor count suggests and more than one. Look at the referring domains of the pages that currently rank for the keyword, discount the ones that are not buyable, and the remaining gap is the order of magnitude. A page ten domains behind does not need fifty; a page three hundred behind will not be moved by five. The gap map gives the count with the prices attached.

The price distribution answers this better than opinion does. Across the catalogue, half of all asks fall between $71 and $326 per placement; a DR 70–79 root domain costs a median $653 against $170 for DR 30–39. The strong link costs several times the cheap one, not dozens of times, and a single strong in-topic link is worth more than the arithmetic suggests because it is what the ranking pages have.

The cheap end also carries the risk. Suspect domains ask a median well under their honest neighbours in every band, so a budget that optimises for count drifts toward exactly the inventory that fails verification. Buy the best in-topic links the budget affords; fill with mid-band, never with the bottom decile.

Over weeks, not in one afternoon. A burst of new referring domains to one page is a footprint on its own, and it also throws away information: you learn nothing about what a tranche did if the next one lands before the first has been observed. A standing order places a fixed number of links per fortnight or month inside a per-run budget, each run a held quote you approve, and the weekly brief reports what the last tranche did before the next one goes out.

• A reserve for the second tranche. Around a fifth. The first tranche tells you which publishers delivered and stayed live; the reserve goes to more of those.
• Content. A guest post needs an article. Publisher-written is usually included; a supplied article is due within 14 days; having it written for you is a flat fee. Decide before ordering, because an order waiting on content is a paid order not yet placed.
• Nothing for “insurance” links on suspect domains. They are excluded from plans by default for a reason.

Every order tracks the target’s rankings and the link’s presence at fixed checkpoints out to a year. The outcome is reported as observed movement with a confidence, never as attribution; the honest version of “did it work” is “the page moved from 14 to 9 in the 60 days after these links landed, while the rest of the SERP moved little”. Allocate the next budget on that, not on the invoice.