Broad Match or Exact Match?

I have set different ad groups up, within one campaign, each using different tactics in an effort to see what is going on. So one only does Broad Match, another only does Exact Match, and another only uses the automatic Search Match with no keywords supplied by me. I'm hoping the data will eventually also indicate what keywords users are actually searching on in the App Store so I can optimise my keyword list, as well as helping decide what to bid more for. Volume is still very low. If you search on the App store for the same keyword you'll see it very quickly stops showing you ads (if it shows you any at all), and when it does it isn't the same ad each time. So every user search isn't leading to an ad impression being displayed. It's worth testing any search terms you're using to see who else pops up. They may all be outbidding you.

With Broad Match, Apple are taking your keyword phrase and matching it against user Search Terms broadly (this might seem a bit stupid to say, but there are unexpected results). This means if you've a keyword "cool game" they'll maybe user search terms like "cool games", "cold game" etc etc. From the early results I'm seeing, broad seems to be incredibly broad in that they're matching searches to keywords that don't even look close, even in cases where the user search term was apparently an exact match for one of my existing keywords. So broad match should match more often than exact match (assuming the same bid amount), as exact match should only match user search terms that exactly match the keyword phrase you're bidding on.

I'd suggest setting up different ad groups, with the same keywords but the matches set differently in each group. Use the broad match (and the automatic Search Match) more as a way to cast a wide net and discover what keywords users are searching on, and catch ones you wouldn't otherwise. I'm thinking it might be worth bidding more for Exact Match keywords, and less for the others.

Good test and analysis. It's worth trying different match types.

My another problem is that the TTR (Tap Through Rate) is really low, less than 10%, meawhile the impression volume is also very low. How is yours? I'm wondering TTR with Broad Match keywords will be lower than Exact Match.

by and large, exact will have higher TTR, CR, CPA than broad; unless the broad match is taking exact match impressions (i.e. to do list matching to [to do list]).

Eventually you want most of your spend to from exact matches (because they are more consistent and you can raise bid on them and know there's only one possible search you're bidding up or down for).

But at first you will want to start mostly on search match and broad match so that you can figure out which search terms do well, and then add those that do as exact match keywords (but negate them from the broad/search match campaign so your exact match campaign will serve for those impressions).

-Gabe