We are trying to find where press release value shows up

This is not a complaint document. We recommend Search Atlas, we run it across our whole book of business, and press releases are part of what we sell and report on.

That last part is why we are writing. When a client asks what a press release did for them, we want to answer with a mechanism and a number, not with a general statement about brand exposure. Right now we cannot do that, and we would rather find out we are measuring it wrong than quietly stop using a product that works.

So everything below is laid out so your team can check it. Every figure names the tool and the call that produced it, and the last section is a reproduction guide. If we have the methodology wrong anywhere, that is genuinely the most useful outcome of this document.

49 press releases, 48 of them published successfully. One failed to publish. Distribution runs through Signal Genesys, and the status endpoint returns a distribution report ID for each.

Sample distribution receipts: Mayday "Texas Pest Control Provider Mayday Expands Roach Services" returns Signal Genesys ID 426179, status report_available. WW Air "Trusted HVAC Maintenance and Repair Services" returns 420020. Both report as approved and published.

Stating our assumptions plainly, because if the model is wrong then the measurement was always going to come back empty.

1. 1A release is written and sent to the wireSignal Genesys accepts and approves it for distribution.
2. 2It syndicates to publisher sitesEach publisher hosts a copy at its own URL on its own domain.
3. 3Those copies link back to the client siteWe assumed some share of these would be followed links.
4. 4The client gains authority and referral visitsLink equity plus readers clicking through from the coverage.
5. 5Coverage feeds brand and AI visibilityPublished mentions become a source that search and AI systems can cite.

Steps 1 and 2 hold up completely. Our measurement problem starts at step 3.

We want to be clear about this before anything else. The syndicated pages are not placeholders and they are not thin mirrors on a shared host. They are live pages on independent domains with real authority.

We fetched two of them directly. Both returned HTTP 200 with substantial page weight, a self referencing canonical, and an indexable robots directive.

And the receiving domains carry genuine authority scores, per Search Atlas Site Explorer:

So the delivery half of the product is doing its job. Our questions are all about what happens after delivery.

On both pages we fetched, every link pointing back to the client site is tagged nofollow. Not some of them. All of them, in an identical three link pattern.

capitolhilltimes.com  →  3 links to maydaypest.com          →  all rel="nofollow"
xpr.media             →  3 links to wwairspecialistllc.com  →  all rel="nofollow"

We want to be fair about what this means. Nofollow on wire distributed press releases is the compliant, correct choice. Google's link scheme guidance has asked for exactly this for years, and a wire that applied followed links at scale would be creating a risk for our clients rather than a benefit. We are not suggesting this is a defect.

What it does mean is that the mechanism we assumed at step 3 of our model is not available. A nofollow link does not pass PageRank. So if press releases build authority, they must build it through a path other than link equity, and that is the path we would like your team to describe for us.

This is the single item we would most like clarified, because it changes how we report to clients.

Search Atlas Site Explorer returns those same syndicated URLs with a link type of dofollow, while the live pages serve nofollow:

Site Explorer, maydaypest.com backlinks:
  https://lifestyle.capitolhilltimes.com/story/630244/...   DA 2   "dofollow"

Site Explorer, wwairspecialistllc.com backlinks:
  https://xpr.media/story/723639/...                        DA 1   "dofollow"

Live page fetch, same two URLs, same day:
  every link to the client domain                                  rel="nofollow"

There are a few explanations we can think of, and we genuinely do not know which is right. The index may be recording the state at first crawl and the publisher may have changed the template since. The label may mean something different from the raw HTML attribute. Our fetch may be seeing a variant of the page that the crawler does not. Or we may simply be misreading the field.

Setting link equity aside entirely, a syndicated article should be able to send a reader. We looked for that reader across every client account that has published releases.

We pulled every traffic source from GA4 for eight client properties covering 1 August 2025 to today, then matched all 47 syndication domains we had harvested from the two backlink profiles against the full source list.

Across 1.5 million sessions and 1,639 distinct traffic sources, we did not record a single visit from a syndication domain. We are open to the possibility that this is a measurement artifact rather than a real zero, which is question 4 below.

This was the result we most expected to be positive, since published coverage is exactly the kind of source an AI system might cite, and AI visibility is a large part of why we run Search Atlas.

Using LLM Visibility on Mayday Pest for the 90 day window ending 27 July 2026:

The AI platforms cite the client's own site, Orkin, Terminix, This Old House, Reddit, Yelp and Google for this account. We could not find a syndicated release among the sources.

We ran this comparison to check ourselves. If press releases were the driver, two accounts running them should move in a similar direction. They do not.

Both accounts published four releases through the same wire in the same period. Here is Search Console organic clicks for each.

Both accounts got the same product. One is flat, one grew almost eightfold. Whatever separates them, the press releases are the constant, not the variable. We read this as evidence that releases are not the driver, but we would welcome a different reading of it.

Answering any of these would change how we use and report on the product. They are ordered by how much they would change our behaviour.

1. 1Which link attribute is authoritativeSite Explorer reports these syndicated links as dofollow and the live pages serve nofollow. Which should we trust, and what exactly does the link type field represent?
2. 2Where does the SEO value accrue if not through PageRankGiven that nofollow is the correct and compliant choice for a wire, what is the intended mechanism? Unlinked brand mentions, entity signals, indexed coverage that ranks on its own, something else?
3. 3How should an agency measure that mechanismIf the value is real but not a referral or a followed link, tell us the metric and the report where it becomes visible, and we will track it there.
4. 4Should we expect referral traffic at allIf the honest answer is that wire syndication does not generate clicks, we are fine with that. We would just like to set client expectations correctly rather than treating zero as a problem.
5. 5How are syndication publishers selectedSeveral receiving domains sit well outside the client's vertical. We saw an HVAC release on parenting, spirituality and film industry domains. Is topical relevance a factor in placement, and does it matter for the intended value?
6. 6Can we get the full syndication list per releaseThe MCP distribution endpoint returns a Signal Genesys report ID and points to the dashboard. An API accessible list of every URL a release landed on would let us verify placement and report it to clients directly.
7. 7Is there benchmark data you can shareAcross your wider customer base, what does a successful press release program look like in numbers? If there is a benchmark we are falling short of, we would like to know what we are doing differently.
8. 8What is the credit cost per releaseWe could not find per release cost exposed through the MCP. Having it there would let us model return properly instead of estimating.

Nothing here is derived or modelled. Each figure comes from a single call, listed below so your team can run the same checks.

Release inventory
  pr_list(page_size=50, slim=false)              → 49 releases, 48 Published, 1 Publish failed

Distribution status
  pr_get_distribution_status(0320cea8-...)       → Signal Genesys 426179, report_available
  pr_get_distribution_status(0c896668-...)       → Signal Genesys 420020, report_available

Backlink profiles and link type labels
  se_analyze_domain("maydaypest.com", ["backlinks"])          → 342 backlinks, 196 referring domains
  se_analyze_domain("wwairspecialistllc.com", ["backlinks"])  → 452 backlinks, 166 referring domains

Live page inspection
  GET lifestyle.capitolhilltimes.com/story/630244/...   → 200, 97,496 bytes, 3 nofollow links
  GET xpr.media/story/723639/...                        → 200, 152,993 bytes, 3 nofollow links

Referral traffic
  GA4 Data API runReport, 8 properties, dimension sessionSource
  Window 2025-08-01 to 2026-07-27, matched against 47 syndication domains

Organic performance
  Search Console API searchAnalytics/query, dimension date, aggregated monthly
  Properties: wwairspecialistllc.com, maydaypest.com

AI citations
  llmv_get_citations_overview("maydaypest.com", 2026-04-27 to 2026-07-27)
  Repeated with search="capitolhill" and search="newsblaze"

We are not looking for a refund or an argument. We are looking for a mechanism we can point at.

Press releases are one of the easier things to explain to a business owner, and clients like seeing their name on a published article. If there is real compounding value underneath that, we want to keep buying it and we want to report it accurately. If the honest positioning is that this is a brand asset rather than an SEO lever, we are equally fine with that, and we would adjust how we sell it.

What we cannot do is keep recommending it while being unable to answer a client asking what it did. Any of the eight questions above would resolve that for us.