In-App: Global Mobile Marketing Blog

#advices from 📱

Today I am sharing with you some thoughts https://www.linkedin.com/posts/davidvargasmontielbe-careful-with-web-and-web-to-app-campaigns-activity-7442850385255608320-MKSq/?utmsource=share&utmmedium=memberdesktop&rcm=ACoAABOvCk8BhJQttfLkpYTNnhar7CO5aOyIsbMon https://www.linkedin.com/feed/update/urn:li:activity:7490335608884781056/ by https://www.linkedin.com/in/davidvargasmontiel/

I've warned here many times about the risk of leveraging DSPs while using default attribution windows, and I've even shared posts showing how DSPs can quickly steal your organic attributions due to their aggressive approach with forced clicks.

Today, I'm sharing a setup that's worked for one of my accounts, letting me spend more than $80k while maintaining my CPA and ROAS targets, with spend increasing WoW.

🪟 ATTRIBUTION WINDOWS 🪟
As already suggested, I encourage people to really limit VTA when using DSPs. In this case, we started a CPA campaign using a 12h VTA window but the banners stole all the conversions while I did not see any incrementality looking at the blended and also saw an organic decrease. We then switched to 6h, and the same thing happened.

The solution was limiting the VTA window to 3h. With that window, we saw the campaign start limiting view conversions and bringing in more incrementality.

⚙️CAMPAIGN SETTINGS⚙️
As I mentioned above, we first tried a normal CPA campaign, since most of our users were triggering free trials but the results weren't great, either in pure efficiency or in incrementality.

So we created a projected_purchase event that sent the expected value per subscription based on our historical trial conversion rate, and used that event for a ROAS campaign (Moloco has a specific tag for these proxy value events during campaign creation).

That campaign hit its ROAS and CPA goals from day zero.

🖼️CREATIVES🖼️
This is the trickiest part. First, I always recommend starting with:
-Videos and Native Videos(9:16 / 1:1)
-Native Images in 1200x628, 720x1280, 720x720
-Images: 320x50

➡️ But there's an important nuance here:  if you see the algorithm pushing 320x50 while your organic installs and conversions drop, you're likely cannibalizing. If that doesn't happen, I'd still recommend focusing your production on videos, native videos, and native images to avoid any risk of cannibalization altogether.

In my case, the campaign leaned heavily on 320x50 in the beginning, but after introducing more videos and gradually pausing the 320x50s, we ended up with a creative mix where the top 3 spending assets are all videos and 1200x628 banners.

There's one last thing: we scaled really aggressively throughout this whole time, reaching up to +$4k/day in daily budget, but I think we hit the ceiling of this setup last week. I still need to look at some more numbers to confirm that, but if this post gets enough support, I'll share how I found that out too 👀

➡️ Check your "DM" and "link in bio" links in Instagram 📷🍏

Instagram has recently started blocking these links

➡️ Budget allocation benchmarks 2026 for different platforms by category from https://superscale.ai/learn/paid-social-media/ 💰

⚡️Adjust also https://www.adjust.com/blog/adjust-chatgpt#chatgptadsand_adjustan integration with ChatGPT Ads 🎨

Adjust has introduced a native ChatGPT integration, allowing marketers to run advertising campaigns directly within ChatGPT, creating a new channel for user acquisition (UA)

It enables advertisers to attribute installs and in-app events back to ChatGPT ad campaigns

To integrate the ChatGPT Ads in Adjust, advertisers need to:
• Select Partners under Campaign Lab
• Select New partner > ChatGPT Ads
• Choose the relevant app
• Enter their ChatGPT Ads API key (you can generate this from the Settings page in OpenAI’s Ads Manager) and Pixel ID
• Map Adjust events to ChatGPT events
• Use the Adjust URL templates as campaign targets in OpenAI’s Ads Manager

📢 Latest job opportunities in mobile marketing

🔍 Google is searching for https://www.google.com/about/careers/applications/jobs/results/78953016983462598-app-specialist-customer-onboarding-global-customer-solutions-english-russian?location=Dublin,Ireland&skills=mobile

🔍 Adapty is hiring https://adapty.io/careers/user-acquisition-team-lead/?rsid=3f63ba64-f51a-4e3b-8052-fe5cfa96157e and https://adapty.io/careers/senior-user-acquisition-manager/?rsid=3f63ba64-f51a-4e3b-8052-fe5cfa96157e https://hirify.me/jobs/512953-senior-ua-manager-incentivized-growth
🔍 Sporty Group is looking for https://job-boards.eu.greenhouse.io/sportygroup/jobs/4933606101?gh_src=0CV1JG

🔍 Nexters is searching for https://bolt.eu/en/careers/positions/52a4d0d4-8dd9-4ce7-922d-15f8c001a5a7/ https://www.linkedin.com/feed/update/urn:li:activity:7482810299885142016/

🔍 Lóvi is hiring https://lovi.care/careers/ua

🔍 Hitapps is hiring https://hitapps.peopleforce.io/careers/v/22468-ua-manager

🔍 "Fast-growing game development studio" is searching for https://team-up.zohorecruit.com/jobs/Careers/616278000021903016/User-Acquisition-Manager?source=CareerSite

🔍 Moon Active is hiring https://www.moonactive.com/moonactive-position/?uid=97952ab6-4175-4727-8c16-914895df3726

🔍 Uplive is searching for https://bolt.eu/en/careers/positions/52a4d0d4-8dd9-4ce7-922d-15f8c001a5a7/ https://uplive.zone/careers/job/user-acquisition-marketing-manager-4139388

#advices from 📱

👾 https://www.linkedin.com/in/mariuszgasiewski/?lipi=urn:li:page:dflagship3detail_base;R/EowcqZTfCgZSuX2VCSSw== just https://www.linkedin.com/feed/update/urn:li:activity:7488104095019614208/ his huge new gaming industry report focused on different dimensions of gaming

Topics included in the report:
▪️ Users in Gaming on different platforms
▪️ The Attention Economy: Gaming vs. Other Leisure Spending
▪️ Updates on downloads and revenue voncentration in Mobile Gaming
▪️ Mobile Gaming vs Non-Gaming
▪️ Mobile Gaming Growth (Android vs. iOS)
▪️ Mobile Gaming Trends by sub-Category (iOS & Android)
▪️ Mobile Gaming Trends in Top Countries (iOS & Android)

♾ Meta’s “Incremental Attribution” shows better results than “Standard Attribution” in 2026, research by Haus 👈

A https://www.haus.io/blog/is-metas-incremental-attribution-outperforming-standard-attribution-what-the-data-shows by Haus suggests that Meta’s Incremental Attribution (IA) is now performing better than the platform’s standard attribution model

Across experiments run between July 2025 and June 2026, campaigns optimized with IA achieved a 1.26× higher incremental ROAS, reversing last year’s findings when standard attribution still had the edge

➡️ The biggest improvement was seen among direct-to-consumer (DTC) brands, while omnichannel advertisers saw similar results with both models

#advices from 📱

♾ Today I am sharing with you some thoughts about https://www.linkedin.com/posts/curtishowland_does-meta-always-choose-the-best-ad-and-share-7486359939829972992-r4CK/a by https://www.linkedin.com/feed/update/urn:li:activity:7436417945590784000/ from https://misfitmarketing.co/meta-creative-guide

Does Meta always choose the best ad?

And how do you know when it's wrong?

- I analyzed 1702 ad sets.
- Each ad set had 5+ ads
- Meta predicts the best ad, begins spending, but doesn't always stick with it

Results (from day 1 to day 4)
- No. 1 ad continues to be top ad 64% of the time
- No. 4 ad continues to be fourth ad 60% of the time
- No. 2 and 3. ad have roughly the same chance, X%, of becoming top ad

This helps you make more informed decisions (with of course some level of risk)

Deciding on day 1 data

→ You can turn ads on / off change budgets, but it's low confidence
→ Only act on big outliers, or make small adjustments (10-15% budget max)
→ Fold in pacing from the current day (call it day 1.5) before you touch anything

Holding for day 2 and 3

→ Hold most changes for here unless there's an extreme outlier
→ This is where the eventual winner separates from the pack
→ Let Meta finish sorting before you scale or cut

What counts as an extreme outlier:

1. Flip signal

→ Top ad doing very poorly, a lower-budget ad crushing it
→ A strong sign the ad set is about to flip
→ Force it early instead of waiting on Meta

2. Dead ad set

→ Top ad poor and the whole ad set poor (e.g. 5x CPA, 0 conversions)
→ Nothing in here is going to turn around
→ Start trimming budget now

Most of the time the answer is to wait for more data to come in

♾ Meta has started rolling out an Ads MCP Server ⚡️

Meta has officially opened its Ads MCP Server to all developers, allowing AI assistants to connect directly with Meta Ads accounts through a standardised interface

The integration https://developers.facebook.com/blog/post/2026/07/16/meta-ads-mcp-server/ to create campaigns, manage product catalogues, analyse performance, generate reports, and automate campaign workflows using live advertising data instead of manual exports

♾ Meta launches new app for Facebook Marketplace sellers

Meta has https://about.fb.com/news/2026/07/introducing-seller-app-facebook-marketplace/ Seller, a new app built for people who sell regularly on Facebook Marketplace, giving them dedicated tools, like AI-powered listing creation, a unified inbox, inventory management, and performance insights, in one place

🔍 Google will use UK and EU user IP addresses for ad personalization after August 3, 2026 ⚖️

Interesting fact: Google didn't use IP addresses for this purpose in these regions before

Google receives IP addresses to route traffic and deliver ads, through customer tags, SDKs, HTTP calls and uploads, but it is a novel practice in the UK and EU, where IP addresses are regulated as a personal data under GDPR

🍏 Apple introduced a new subscription model: users can now pay monthly — but with a 12-month commitment

In simple https://developer.apple.com/news/?id=agq42lxe it’s a hybrid between monthly and yearly plans: users don’t need to pay upfront for a yearly subscription, but they’re still locked into the full term.

Cancel early - and you still pay until the end of the period 💳

"AI" is the most-used keyword in 6 App Store categories

According to the latest https://www.mobileaction.co/report/2026-aso-report/ “AI” is now the #1 keyword in multiple App Store categories - including Productivity, Photo & Video, Utilities, and even Entertainment

• Productivity: AI is #1 keyword, above “notes,” “tasks,” and “daily”

• Photo & Video: #1 also, above “photo,” “video,” and “camera”

• Health & Fitness: #2, just below “health” itself

• Education: #2, just below “learning”

• Graphics & Design: #1, above “design” and “create”

• Lifestyle: #1, above “time,” “daily,” and “life”

• Utilities: #1, above “device,”“phone,” and “manage”

• Entertainment: #1, above “TV,” “content,” and “subscription”

📈 Tinuiti Q1 2026 Digital Ads Benchmark Report

The Tinuiti Digital Ads Benchmark Report is based on anonymized performance data from advertising programs under Tinuiti management, with annual digital ad spend under management totaling over $4 billion

Table of Content:
🔴 Overall Trends (Spending Growth Mostly Higher Across Platforms as Pricing Growth Remains Muted)
🔴 Paid Social (Facebook, Instagram, TikTok, Snapchat, and Pinterest benchmarks)
🔴 Paid Search (Google and Microsoft benchmarks)
🔴 Commerce Media (Amazon and Walmart)
🔴 Video and Display (YouTube, Prime Video, Netflix, GDN, and More)

♾ Meta added One-Click Conversions API Setup for Advertisers ✨

Meta is finally https://www.facebook.com/business/news/pixel-conversionsapi-updates one of the biggest barriers in performance marketing - complex CAPI setup

With the new update, advertiser now can enable Conversions API directly in Events Manager with a single click

#advices from 📱

Today I am sharing with you some https://www.linkedin.com/posts/karen-charykovfirst-mistakes-keep-working-ive-been-activity-7447622831049355264-5ru?utmsource=share&utmmedium=member_desktop&rcm=ACoAABOvCk8BhJQttfLkpYTNnhar7CO5aOyIsbM by https://www.linkedin.com/in/karen-charykov/ (Founder of https://apps.apple.com/app/memacy/id6760121647)

🍏 First mistakes. Keep working.

I’ve been running Apple Search Ads experiments across a few iterations, and something became very obvious: what you write in the App Store matters as much as how much you spend.

Small budgets don’t mean you can ignore analytics. It’s actually the opposite.

When you don’t have scale, every signal matters more - not less. You start seeing patterns you’d normally miss, especially around conversion and positioning.

Titles. Subtitles. Categories.

They directly impact how traffic converts - and how relevant your app looks to the algorithm.

Looking back, I made a few classic mistakes. I started changing the app name and categories too early, without enough structured observation. Instead of learning, I was introducing noise.

Now I'm correcting that - slowing down changes, measuring properly, and treating metadata as part of the product, not just marketing.

ps
One more important nuance: the second peak on this chart came with roughly 2x less budget than the first one.

Which means performance didn’t just come from spend - it came from better alignment.

That’s the part many people underestimate.

Less budget doesn’t mean less opportunity. It just forces you to actually understand what’s working.