-
Impact of Budget Changes on Target CPA Campaigns
Posted by on June 30, 2021 at 12:52 pmBeen running multiple Target CPA search campaigns & display campaigns for several years now. Something that’s never sat quite right with me is the recommendation that Target CPA campaigns need (or prefer) uncapped budgets. Every single time I get a campaign to the point of an uncapped budget, performance becomes all over the place.
For example, let’s say I have a campaign with a $50 CPA. As I’m scaling with Target CPA (by increasing budget 10%-20% at a time) performance holds steady and the campaign continues to spend the full budget after the increase. But once I get to the point of “uncapped”, let’s say for this campaign that’s at $800 per day, the spend drops to maybe half of that and only spends $300-$400 per day for the vast majority of the time.
Now let’s say I have an issue with inventory and need to scale back until I can restock product so I start dropping the budget. Guess what happens? Spend actually increases. So for my $800/day campaign that only likes to spend $300-$400 per day, I drop the budget by half to $400/day. Except that now, it not only spends the entire budget every single day… but it often overspends (while actually doing a BETTER job of hitting my CPA targets).
While this is great, it’s obvious counter to my issue of scaling back ads until inventory is restocked. But I’m curious if anyone else has seen this? It hasn’t happened to me just once or twice, but probably 6 or 7 times now over the last several years. Almost seems like the more restrictions I give the campaign (CPA target + budget) the better it performs when compared to an uncapped budget.
-
3 Replies
-
I’ve seen this phenomenom occur several times in some of our accounts as well. We’ve run a few tests in determining if scale really is the breaking factor and came to the conclusion that it actually wasn’t as there are more variables at play that can cause this sudden drop.
One combination that seems to cause this more often than not is when you have limited audience sizes and reach the maximum serving threshhold or exceed marginal utility.
There is only a certain supply each day of people on your audience list or people searching with a certain keyword. The point where your CPA stays stable is all within marginal utility as Google’s algorithm seemingly manages to juggle all advertisers campaigns against their set performance goals and budgets.
But once there is no more surplus in supply the algorithm seems to struggle in balancing objectives and prioritizes spending over performance. But then again the tCPA bid strategy will limit down spent in order to reach CPA with a given confidence. Therefore by increasing the budget over an audience threshhold you create this dilemma of Google wanting your money, but also wanting to reach your goals. This in turn results in the campaign tuning down spent again.
What we found as a potential solution (currently still low confidence but preliminary results are looking good) is to gradually while increasing budget also increasing audience sizes. You gotta be careful though to not overlap audiences since then this can occur as well – for example: broadening the remarketing campaign with an additional customer list. I’ve found this working very well with affinity audiences on higher indexes. Also we’ve had promising results using the audience extension slider within the campaign.
-
I’ve had this EXACT thing happen. Whatever the recommendation they give you, budget should be 1/2 to 1/3
-
I’ve experimented with this as well on search campaigns with a tCPA bid strategy and the findings sound similar. Glad to hear that other’s have experienced it as well. It almost seems like if you set the budget too high on a tCPA campaign and start to hit that spend wall, then you need to scale back your budget and start to incrementally increase again as your spend hits the new thresholds.
Log in to reply.