// Guide — Choosing a partner

How to choose a mobile ad network

Every ad network claims quality traffic and transparent reporting. The claims are identical; the operations behind them are not. This is the evaluation framework we would use if we were the advertiser — five criteria that separate networks that perform from networks that invoice.

Who owns the inventory?

Ask where impressions actually come from. Networks that own inventory — their own placements, apps or publisher relationships — can answer placement-level questions and enforce placement-level quality. Networks that resell exchange traffic cannot: they see averages, not placements, and averages hide everything that matters. Neither model is inherently wrong, but you should know which one you are buying, because it determines what reporting is even possible. A network that cannot show you placement-level data is asking you to trust a black box with your budget.

Where does measurement live?

The referee should be yours. Serious networks integrate with your MMP — AppsFlyer, Adjust, Branch, Kochava, Singular — and accept its verdict, including rejection rules written into the agreement before spend starts. Red flags: pressure to use the network's own dashboard as the source of truth, resistance to documented rejection rules, or attribution windows that quietly favor the seller. If a network's numbers and your MMP's numbers cannot be reconciled line by line, you will spend every month-end arguing instead of optimizing.

What is the fraud policy — in writing?

Fraud exists on every source; the question is what the network does about it and who pays for it. Ask four questions: What is filtered before my tracker sees the traffic? What is scored after? What happens commercially to conversions my MMP rejects? Can I see the filter categories? A network with real answers names its methods — emulators, click injection, SDK spoofing, install farms — and puts the rejection logic in the contract. A network without them answers with the word 'quality'.

Which pricing model are they selling you — and why?

CPI, CPA, CPM, CPL, hybrid and revenue share each shift risk between buyer and seller. CPI puts funnel risk on you; CPA puts it on the network; CPM puts everything on you. A trustworthy partner recommends the model that fits your funnel math and tells you why — including when the honest answer is that your tracking is not ready to buy at all. Be suspicious of anyone who sells you CPM on unproven funnels or CPI without asking what happens after the install.

What does reporting actually look like?

Ask for a sample report before you sign. The minimum bar: spend, installs, billable events and downstream quality markers broken down per source or placement, on a cadence you can act on weekly. Bonus signals: the report is the same table the buyers optimize from, not a curated summary; your contact can explain any line in it; and the network proactively shows you what it cut, not just what it scaled. Transparency is easy to claim and easy to test — ask for the raw view.

A word about us

Mobadly is built to this framework because we work both sides: we buy media for advertisers and we own an in-app ad network. Buyers who control inventory see placement quality directly, and we report at that same level — source scoring, MMP-verified events, rejection rules in writing. Whether or not you work with us, hold every network to these five criteria. The ones worth your budget will pass them without flinching.

Start advertising