M MurmTools

go-mpulse.net

go-mpulse.net is the host that receives the beacon from mPulse, Akamai's real-user monitoring product. A page that includes it reports back how quickly it loaded for the person who actually loaded it — timings, not content — so the operator can see real performance rather than lab performance. It shows up in request logs on sites large enough to care about that. This index recorded it on 167 of the 5,294 sites measured, 3.2%, ranking it 24th by reach among the 88 companies indexed here.

Almost entirely enterprise infrastructure and large newsrooms

The examples are unusually consistent for this index. Networking and enterprise software fills most of the list: azure.com at rank 27, cisco.com at 249, meraki.com at 260, salesforce.com at 295, webex.com at 182, dell.com at 375, intuit.com at 136. Consumer hardware appears through xiaomi.com at 147 and mi.com at 337, security through avast.com at 258, and US news through washingtonpost.com at 357 and npr.org at 478.

That is the profile of a paid performance-monitoring product rather than a widely embedded free one, and it explains the median site's shape: 9 companies contacted over 45 requests, 1 of them classified as tracking. Among the 88 companies here that is 40th by median company count — the middle of the field.

A late-starting group: 879 ms at the median

Rank the 88 companies in this index from the latest median first third-party contact to the earliest, and the sites carrying go-mpulse.net come ninth. Their median first contact is 879 ms; the earliest recorded anywhere in the group is 145 ms.

One thing that figure is not: a measurement of this host's own request. It records when each site first reached any third party at all, and the crawl does not break that timing down by destination. What can be said is that pages carrying a performance beacon start their outside contact later than most groups here, and that the spread within the group is enormous — 2,834 ms on xiaomi.com and 2,374 ms on mi.com against 436 ms on washingtonpost.com and 514 ms on npr.org.

What travels with it

Google was present on 96 of the 167 sites, 57.5%. Then comes a measurement cluster: akstat.io on 64 sites, 38.3%, and Akamai's own hosts on 47, 28.1%. OneTrust was on 59, 35.3%.

An Adobe stack sits underneath a quarter of these pages — adobedtm.com on 42 sites, 25.1%, and demdex.net on 32, 19.2%. Google Tag Manager reached 21, 12.6%, tiqcdn.com 14, 8.4%, and Amazon CloudFront and Meta 16 each, 9.6%.

The reading is that this is a corporate stack rather than an advertising one. The hosts that keep it company are tag containers, CDN and monitoring endpoints, and a consent platform — not exchanges.

Early contact on 158 of 167 sites

On 158 of the 167 sites, 94.6%, some company was contacted before the consent banner could have been answered; the median site had 6 companies inside that window and added 2 more after it closed.

The individual pages fall into two camps. intuit.com had all 9 of its companies inside the window at 795 ms, avast.com 9 of 10 at 632 ms, washingtonpost.com 10 of 12 at 436 ms. Against them, xiaomi.com and mi.com each recorded zero companies inside the window, with first contact after two seconds in both cases. go-mpulse.net was present across both camps, which is the point: the count belongs to the page, not to any one host on it.

Consent tooling on these sites

A consent tool was detected on 100 of the 167 sites, so roughly two in five carried no detectable banner at the time of the crawl. OneTrust was seen alongside on 66 sites, an IAB TCF consent tool on 28, Usercentrics on 3, Osano on 2 and Cookiebot on 1. Those counts overlap rather than divide the 100, because a page can run more than one banner product.

In the examples the pattern tracks the sector rather than the vendor. The enterprise software sites — intuit.com, cisco.com, webex.com, salesforce.com, dell.com — were recorded with OneTrust, as were the two news sites. azure.com, xiaomi.com, mi.com, meraki.com and avast.com had no consent tool detected.

Highest-ranked sites where this was seen
SiteCompaniesBefore consentFirst contact
azure.com951275 ms
intuit.com99795 ms
xiaomi.com902834 ms
webex.com1261185 ms
cisco.com107744 ms
avast.com109632 ms
meraki.com33887 ms
salesforce.com941742 ms
mi.com1002374 ms
washingtonpost.com1210436 ms
dell.com1541352 ms
npr.org156514 ms
Consent tools seen on these sites
ToolSites
OneTrust66
IAB TCF consent tool28
Usercentrics3
Osano2
Cookiebot1

Seen on the same pages

Companies most often contacted by the same sites. Co-presence only — the crawl records what arrived, not who called whom.

Questions

What does go-mpulse.net do?

It receives real-user monitoring data: how long the page took to load for the visitor who loaded it, broken into timings the site's operations team can act on. It is the reporting endpoint for Akamai's mPulse.

Why do Akamai hosts keep appearing on the same pages?

Because they belong to the same measurement family. Akamai hosts were present on 47 of the 167 sites, 28.1%, and akstat.io on 64, 38.3% — the most common co-presence in this group after Google.

Why do sites carrying it start so late?

The group's median first third-party contact is 879 ms, ninth-latest of the 88 companies here. This index measures that per site, not per host, so it cannot attribute the delay to any single destination — but the sites in question are heavy enterprise pages rather than lightweight ones.

What else is usually on the same page?

Google on 57.5% of them, akstat.io on 38.3%, OneTrust on 35.3%, Akamai hosts on 28.1% and adobedtm.com on 25.1%. It is a corporate measurement stack, and it clusters.

Check any page yourself MurmStack is the tool that produced these figures. Free, two permissions, no host access.
Add to Chrome — free