Quick tips for SMB advertisers exploring Facebook Ads...

Not really an article - But I hope this gives insight and clears up some confusion for small business owners and their employees who are not very experienced in general with PPC/Performance Advertising. These tips are based on my experiences consulting for smaller brands spending below roughly about $500 on average per day on FB and the nuances you'll have to take into account at lower budget levels. Not in any particular order of importance:

- You can't stack interest targeting on top of Behaviors/Demo. I am not confident from my experience that a smaller advertiser can even leverage or serve ads accurately to specific FB categories such as "First Time Home Buyer". Try broader targeting first to make sure your account can get up and running with some sort of history before refining it.

- You may need to begin your campaign with CPM bidding (Pay a certain amount for 1,000 impressions). Do not allow Facebook to auto-optimize (there is a checkbox conveniently hidden deep in the ad set targeting/bidding screen) because your account is likely new and has no data to inform the auto-optimizer. In addition once data collection begins it will be so small that the arbitrary changes it makes will not be based on anything of statistical significance.

- You absolutely cannot afford to ignore the 20% text rule.

- Unfortunately I don't think "Multi-Cultural Affinity" targeting is very accurate particularly at lower budget levels. What you can do to get around this is use the data from your market research to inform what interest targeting you should focus on, then build similar audience lists once your volume reaches the threshold required (it may take months depending on your account). If you have not conducted any surveys on your target market or incentivized participation in some sort of information collection, you will ultimately limit your ability to scale your audience efficiently through paid media. If we don't know more characteristics or attributes of our best repeat customers then we won't be able to find more people who share these same traits and are also likely to buy. (At lower volumes you'll have to do manual "look-a-like" modeling by cherry picking the highest volume, highest converting interest categories and expanding on to similar interests one-by-one.)

- As with any performance media, low volume campaigns in general need to run longer before significant efficiency gains can be seen. If you are only reaching 1,000 people per day and your ad frequency is 1.0 (it probably won't be higher than that with low budgets) it will take longer (1-3 months) to see any performance differences in creative or targeting. At the end of the day the goal for the publisher is to make more money, and cost-savings will only be granted to brands that show potential for greater volume.

- Facebook's diagnostic tools still aren't up to par in my opinion vs. AdWords, so it will be more difficult to get an accurate read on what-if scenarios (what if I increase my bid to ___) at lower volumes. You will fly blind for a a few weeks if not 3 months. For some business models it may not make sense to run paid media if your target market is too narrow. It will be very costly to "find them" using paid media alone.

If you are going to test Facebook I recommend committing at least $5-10k (in a given year) and make sure to test out 1-3 theories you are trying to prove regarding your target market's identity or purchase behavior. This will help cut down on the ambiguity a lot. Bottom line you are not going to find out much insight or get much traction by spending $1k over the course of 12 months for any new product launch which doesn't have strong awareness - Or if you don't have a strong CTA (Call-to-Action) with a good competitive value proposition for the consumer or target audience.

Hope this has been informative and has helped someone in the madness that is Social Media Advertising!