![\"ironSource](\"https:\/\/alienroad.com\/wp-content\/uploads\/blog-ici\/ar-176-1-5f56f178.png\")
<\/figure>\n\n\n\nStart with the cohort, not the campaign<\/h2>\n\n\n\n
Mobile revenue arrives late and unevenly. A user acquired on the third of the month may generate nothing for eleven days and then convert on a subscription or a large in app purchase. If you read performance by calendar week of spend, you are comparing a cohort with three days of life against one with twenty four. It will always look like performance is collapsing, because the newest cohort has had the least time to pay you back.<\/p>\n\n\n\n
So group by install date and hold the group fixed. Then report the same milestones every time: day zero, day seven, day fourteen, day thirty, and whatever longer horizon your payback model uses. The useful question is never whether a campaign is profitable today. It is whether this cohort is tracking above or below the day seven mark that historically preceded a profitable day thirty. That comparison is stable enough to act on and honest enough to publish internally.<\/p>\n\n\n\n
The short list of ironSource Ads metrics worth reporting<\/h2>\n\n\n\n
We keep the weekly report deliberately short. A report with forty columns gets skimmed. A report with six gets read, questioned, and improved. The list below is what we have settled on for most in app purchase and ad monetised titles, and it survives contact with people who did not attend the campaign build.<\/p>\n\n\n\n Three failure modes account for most of the discrepancies we investigate. The first is attribution window mismatch. Your measurement partner, the network, and your internal analytics may each use a different lookback, and the totals will never reconcile until you write one window down and force everything to it. The second is creative level data read at low volume. A creative with sixty installs and a spectacular return on ad spend is noise, and treating it as signal leads to shifting budget onto something that cannot repeat.<\/p>\n\n\n\n The third is the long tail. Rewarded and interstitial inventory arrives from a very large number of individual apps. A handful of them will show excellent surface metrics and terrible downstream behaviour, and when they are each one percent of spend nobody notices. Reviewing ironSource Ads metrics at source level on a fixed cadence, with a minimum volume threshold before any source is judged, catches this early. Blocking twenty weak sources rarely moves the headline cost per install much, but it reliably improves what happens after the install.<\/p>\n\n\n\n The uncomfortable truth is that most of the work is done before any budget moves. Events need to fire reliably on both platforms. Revenue events need to carry a value, in a single currency, with a stated rule about whether it is gross or net of store fees. Someone needs to own the definition of a qualified user, and that definition needs to be the same one used in the payback model. We have seen teams argue about network quality for a quarter when the actual problem was a purchase event that never fired on one operating system version.<\/p>\n\n\n\n Once that foundation exists, the ironSource Ads metrics start agreeing with each other and optimisation becomes ordinary work. You compare cohorts at equal age, you move budget toward sources that hold users past the first week, you cut the tail, and you refresh creative before fatigue shows up in the frequency and click data. None of it is dramatic. But it means that when someone asks why spend went up forty percent last month, the answer takes ninety seconds and comes with a number nobody disputes. That, more than any individual tactic, is what makes a mobile acquisition programme survive its second year.<\/p>\n\n\n\n Keep reading:<\/strong> Ironsource Ads<\/a> \u00b7 Advertising<\/a><\/p>\n\n\n","protected":false},"excerpt":{"rendered":" Which ironSource Ads metrics actually predict profit, which ones mislead, and how to build a reporting layer your finance team will accept.<\/p>\n","protected":false},"author":0,"featured_media":31029,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[29],"tags":[],"class_list":["post-31028","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-advertising"],"_links":{"self":[{"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/posts\/31028","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/comments?post=31028"}],"version-history":[{"count":0,"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/posts\/31028\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/media\/31029"}],"wp:attachment":[{"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/media?parent=31028"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/categories?post=31028"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/alienroad.com\/wp-json\/wp\/v2\/tags?post=31028"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"ironSource](\"https:\/\/alienroad.com\/wp-content\/uploads\/blog-ici\/ar-176-2-5f56f178.png\")
<\/figure>\n\n\n\n\n
Where the numbers quietly lie<\/h2>\n\n\n\n
Build the measurement before you build the campaign<\/h2>\n\n\n\n