+4 votes

It seems user acquisition process will likely be fully automated in the future (we all see where Google and Facebook are going in this respect), so ad creative will probably be the main differentiator for advertisers. From what I see in practice, creatives already contribute ~70-80% to the overall campaign success, while the remaining 20-30% is because of UA techniques and approaches.

Due to this, there is a logical question about how to make the creative production process more efficient.

Would appreciate your thoughts on the following points:
 - how to better organize an in-house creatives production process?
 - how to efficiently create variations and refreshes for top-performing creatives?
 - what is a nice source of ideas for new concepts, apart from FB Ad library and other spy tools?
 - what tools may help to organize the workflow for all the creative team (creative manager, illustrator, motion designer, analyst)?

by (360 points)

1 Answer

+2 votes

I'm of the mind that creative production is absolutely the most important aspect of success in modern mobile marketing. I considered merging this question with How do you test new creatives on Facebook properly?, but since this question is specifically about process, I'll leave it as a standalone thread. That said, here are some other good, relevant threads on this topic:

Background aside, here are some broad tips on building an effective creative production process:

1) First, the process part is actually very important. You need to be operating on a regular schedule, with the team being assigned specific roles and tasked with a defined number of deliverables. Ambiguity is bad here: you want very clear requirements with well-defined production volumes, eg. Person A will deliver 10 new creative concept storyboards each week, Person B will deliver 20 new static interstitial variants of an approved concept each week, Person C will deliver 3 new video variants of an approved concept each week, etc. Everyone needs to know exactly what is expected of them;

2) Someone needs to own it. One person needs to be responsible for the process -- they need authority to assign tasks and must have accountability for results. Creative production doesn't work when it is run by committee: it has to be one person's responsibility. Maybe that's a Creative Director, maybe that's a Director of Ad Creative, maybe that's the Head of Marketing, whatever -- someone needs to fully own the process and have the ability to assign work, engage outsourcers, and demand results.

3) It needs to be weekly. Creative is a volume game and the cadence needs to run on a weekly basis in order to produce the number of variants needed to create a competitive advantage on the basis of ad creative. Monthly is too slow -- even bi-weekly is too slow. 

4) Be mindful of the funnel. Creative can't be turned within a week, especially not video creative. The team needs to be mindful of the funnel, which can span an entire month in order to produce one outperforming video ad creative: concept generation goes to storyboarding goes to production goes to testing. Coming up with a great concept for a video ad on January 1 means that ad isn't being exposed to potential users until February 1, so enough concepts need to be pushed through the funnel every week to ensure delivery of a sufficient number of ad creatives on a rolling basis.

5) Don't stop. It's tempting to stop the creative process after a big wave of outperforming creatives has been delivered -- don't. Creative production is a regular, ongoing process, and it's a critical part of doing mobile user acquisition efficiently at scale. Creative production never stops.

6) Invest in the test. Testing new ad creatives should account for 5-10% of your budget at scale. You need to make an investment into testing new creatives; it's just part of running a modern user acquisition function. Don't skimp on testing -- you'll gain far more from an outperforming creative than you'll lose in testing the 99% that perform at par or worse.

by (15.2k points)
One addition to Eric's answer would be to create a shared understanding of KPIs between channel managers and designers. There is nothing worse for a designer to produce a ton of iterations of a single concept, which can be quite boring, without understanding 1. why and 2. what makes a creative a success or not. I would highly recommend to spend enough time with the designers to get them "excited" about the data generated by the creatives test and over-communicate at first about successes and failures.
This is a great point. Hiring artists / creatives for ads is hard enough, keeping them engaged and involved can be impossible unless they are given some agency in the process. Also, the best ideas tend to come from the art team.