The guides weren't failing. Google had never read them.

In late August I reviewed the guides on woodfiredsaunas.co.uk, the directory I run. Sixteen guides had drawn two clicks between them in 28 days, and thirteen had drawn no impressions at all. The question I wrote down for the next monthly review was whether the guide programme was earning its time.

That question rested on a premise I hadn't checked. Inspecting each guide in Search Console showed that twelve of eighteen had never been crawled: Google had the addresses from the sitemap and had never fetched the pages. Nor had it fetched the hub page that links to all of them. Stopping the programme on those numbers would have been a verdict on writing no search engine had read.

On 1 September I asked Google to index the hub and the guides one at a time, until its daily limit cut me off. Nine days later, sixteen of nineteen guide pages were indexed, and twelve showed their last crawl on the day of the request.

Comparing the 22 days to 31 August with the 22 days to 9 October, Google only:

	Before	After
Clicks, whole site	137	496
Pages shown in results	197	368
Clicks on guides	7	58
Guides shown in results	4	18

Not all of that rise is the indexing request. Four guides went live from 28 August onwards, so they had little or no time in the first window, and young sites tend to pick up traffic with age anyway. What the request explains directly is the crawl dates.

The numbers also don't say the guides are good. In August they could only show whether Google had looked at the pages. Whether the writing works is being tested now, for the first time.