Related Posts
Need 11 Hearts for Dm
What is the salary structure in EXL?
Not a fan of the Christmas decorations
Additional Posts in Creatives
What do we think of mekanism ny?
Can I say no to working the weekend?
New to Fishbowl?
Download the Fishbowl app to
unlock all discussions on Fishbowl.
unlock all discussions on Fishbowl.
Definitely care, but would be more interested to know why? Like are we to assume serif is always better across the board just because? That’s the thing I always struggled most with data analytics, heavy on the stats, but light on the insights, rationale and conclusions.
Lol no comic sans
We were able to isolate each variable through full factorial multivariate testing i.e. we tested every possible combination of layouts and fonts to figure out exactly which individual changes had an impact.
So we know for sure it was just the shift between sans-serif and serif that accounted for the click & CTR changes
Coach
At this point just have machines make banners and social posts if you want the logo upfront and to get your metrics over concept.
Then use creatives for the bigger ad buys.
I personally don’t believe any of the metrics being sold to you by Facebook. It’s in their best interest to keep you buying ad space there and they’ve been caught lying before.
https://www.wsj.com/articles/facebook-reaches-proposed-settlement-in-video-measurement-lawsuit-11570482031
Coach
Wait what? ...sorry fell asleep multiple times when I got to the second sentence.
From a designers perspective what were the products in question being advertised? I’m surprised sans serif was doing better since it’s so played out that’s why so many companies (like Mailchimp and Policygenius) in the tech sphere are reintroducing serifs to break the sea of sameness.
For sure! I don’t think one can exist without the other and create a properly executed system.
Any way we can see this experiment?
We broke all the images when we switched our Instagram to a new account (we pivoted from full stack agency to testing service)
You can see the data here though: http://doescreativematter.com/archive/2
You can see the ads in this article: https://www.marpipe.com/blog/how-to-improve-your-top-performing-creative-in-3-easy-steps
Subject Expert
Yeah is this banner ads? Social ads? Pre-roll? Was there a specific target?
It’s very interesting, but without the context it’s hard to learn anything from that.
On the complexity, it's all about tagging your ads properly. On expensiveness, you can run click campaigns with 50+ ads, then do small scale tests for conversions to see if the data is true.
As far as marketers as a generalizable target, they're definitely more scalable than trying to take, say, sports equipment data and using it for furniture sales lol
This kind of testing works for measuring the effectiveness of DM (direct marketing) but it’s near impossible to use it to measure the effectiveness of big brand building.
No. It isn’t.
The thing about marketing data is it matters in silo context.
Was your ad dramatic and serious?
Were you selling jeans or servers?
Was your budget impactful or did the client skimp?
Is the brand well known?
Who is your audience?
Your data is interesting per campaign but I find it struggles to make generalized design insights.
Oh god, not more data in creative 😭
I don't see why not
I mean, they're more likely to test big ideas rather than small changes, but the principles still apply, right?
I do!!!
Ooh ooh, would you tell me how A&F decides on ad creative?!
Is it mostly around the clothing or the copy?
I've seen your FB ad library and you're doing some small scale tests rn, right?
Lol thanks coach. Advice on editing?
Yes, I care.
What was it advertising? Fonts have personality that work well or don't work well given the context.
So no, I don't care about this data in a vacuum. But given context, I'd be absolutely fascinated. Just need more info.
I guess that's all I really care about- if it would be fascinating.
Like, if it was data for your company or your client, you would care, right?
cool! are we able to see the results in any way? a report?
I did a short write up of it and we actually have the data live on a microsite linked in the article, but we broke the images on one page by deleting the IG account the ads were run on lol
Feedback welcome :D
https://www.marpipe.com/blog/does-creative-matter-the-meta-experiment
I mean do you care about this kind of data or is this something only marketing managers, budgeteers, and founders care about?
True. We're partnering with market research firms and academics to try closing the loop, but you really need a lot of data to find the why.
On the 'across the board' part, data is definitely industry, subcategory, and company specific, but it generalizes pretty well the closer to your niche the data comes from.
Plus, if it scales well, does it really matter why it works?
We need to know why it worked to draw objective conclusions and make them applicable to future work.