ShareChat and Moj for India, Cut by Language Not by State

Fifteen Indic languages are available across ShareChat and Moj, and in the accounts we run three or four of them take nearly all of the working budget. The others fail on cost per acquisition inside a test of roughly EUR 2,000 each and get closed. Running that test is the first thing this engagement does and the most valuable thing in it, because the alternative, spreading spend across a dozen languages because the platform offers a dozen languages, is how India budgets get consumed without producing a decision anyone can act on. The language test: Each candidate language gets its own creative, its own landing experience and its own MMP-attributed number, run to a sample size agreed before anyone looks at a chart. Subtitled Hindi is not a Telugu test and never has been. It loses on completion rate and takes the language down with it, and then somebody concludes Telugu does not work. Creative is commissioned natively per language, usually from platform creators rather than a studio, because the feed's own visual grammar is what performs and polished brand film reads as an interruption in it. Measurement: Android is effectively the whole audience here, so there is no SKAN puzzle and MMP attribution is clean, through AppsFlyer, Adjust, Branch or Singular, whichever you already run. Platform-reported and MMP numbers will diverge and we reconcile them weekly rather than quoting whichever is kinder that week. Where the outcome is offline, a showroom visit, a donation, a first consultation, we build a geo holdout by state and accept a slower read in exchange for a number that survives a finance review. Nonprofits in particular should not scale on a platform-reported conversion figure. The buy: In-feed video on ShareChat and full-screen placement on Moj, branded effects only where the creative idea genuinely requires one, and creator collaborations where the message needs a face the audience already trusts. Programmatic access through DV360 suits brand campaigns with existing frequency governance. The ads manager suits anything carrying a cost-per-action target. We tell you which one you are and buy that way, instead of running both so the plan looks thorough. Brand safety: This is user-generated video at scale in languages your team cannot read. We set exclusion lists, sample delivered placements weekly with a native reviewer, and report what the sample found including the parts that embarrass the plan. Anyone selling you Indian social inventory without naming their sampling method is not sampling. Not included: Regional-language localisation of your site. App store listing work. Performance creative for Meta or Google alongside this, since we are a single-channel practice by choice. We also do not contract or pay creators. We write the brief and set the standard; you or your India agency handle the paperwork, because being both the buyer of the media and the payer of the creators is a conflict we would rather not hold. Who this is not for: Brands whose product, price or delivery does not work outside metro India. This audience is largely not in Mumbai or Bengaluru, and reaching it before the business can serve it produces complaints rather than customers. Also not for teams who want a monthly report. This engagement produces a language and region decision you act on, after which it becomes a smaller ongoing brief or it ends, and we are content with either. How it runs: A defined test window, then a decision document naming which languages and regions carry the spend and which are closed, with the numbers that closed them printed next to each. Then a scaling phase against the surviving cuts. DPDP consent wording on any lead form is reviewed before launch rather than after the first complaint.