Running Facebook ad campaigns is so damn confusing. There are so many things to keep track of: make sure to include certain demographic and exclude others, target this interest and not another. How do you even get going if you’re a newbie and are on a tight budget?
A couple of months ago I connected with Jason Dea from TooCoo Media who is a pure master at Facebook ads, the guy could write novels about the topic. He lives and breathes this stuff. He builds and runs Facebook ads campaigns to help B2C companies to get more customers. I asked him to share the steps startups should follow to advertise on Facebook on a tight budget.
|Free Exclusive Download: Click here for an exclusive step by step guide to Facebook advertising|
Without further ado here is Jason…
The one question I get asked the most by folks trying to get started in paid advertising and Facebook advertising in particular is, “How would you spend your first $100?”
Well, the answer to that is a bit of a hard one.
There really isn’t a nice way to say this, but spending $100 is not enough to do anything really meaningful on the Facebook ad platform. Worse, it could even hurt you. While some might consider that statement controversial, having run thousands of Facebook campaigns myself, and having made thousands of mistakes, I do have a reason for this position.
In the world of online advertising (whether on Facebook or any other platform), data is far and away the most important asset when it comes to buying ad space. However, there is one caveat to keep in mind – you need enough data to be able to make insight-driven decisions.
If you get lucky using a small ad budget, like the $100 proposed, that could end up creating a big risk for you: the risk of making a large decision based off of a small sample size of data.
In online marketing, that initial success is often not repeatable, and using a small sample size of data to make decisions may lead to a catastrophic (read: expensive) end.
Instead what you’ll want to do is spend an appropriate amount of money to get the data and insights you’re looking for.
What to Do with a Bigger Budget?
So: what if you are able to scrounge up, say, $2000 for that same Facebook advertising experiment. What do you do now?
Well, with that kind of budget you should have more than enough money to get a much more substantial pool of data (assuming the cost per click of your audience isn’t astronomical due to being so niche) and much better insights.
Here is the 6 step process I use for getting my campaigns up and running on the road to meaningful data and results.
1. Design Your Ad Creatives
Which ads and landing pages are you planning to use?
Don’t just use what you see your competitors using. Aside from a significant laziness factor, and an entire discussion on unique value propositions, there’s a bigger problem. Everything online has a fixed half-life, advertising especially. If you’ve noticed your competition using a particular type of ad (image, copy, etc’¦) for a while, it probably works. Actually it probably did work and is now close to its saturation point of effectiveness.
So instead let’s start with a brainstorming exercise. If you’ve already got a focused value proposition or “elevator pitch”, this should be the starting point and the rest should be an easy exercise in word-smithing.
a) Brainstorm angles. What is the motivating theme of the ad – fear, pride, price, emotion? Think of headlines and ad copy that work with that theme.
Newsfeed ads don’t have any character limits, and there are 3 areas you have to insert ad copy into:
1. Post Text (above the photo)
2. Link Headline (written in blue just underneath the photo)
3. Description (under the link headline and the text is slightly greyed out)
For the right hand side ads you have 2 areas to insert ad copy:
1. Headline (25 character limit)
2. Body (90 character limit)
For the new generation right hand side you still have 2 areas for ad copy with different character limits:
1. Headline (35 character limit)
2. Body (90 character limit)
b) Find (or better yet, create) pictures to use in ads that support these ideas.
Newsfeed ad images should be 400px by 209px
Mobile ad images should be 560px by 292px
Sidebar ad images should be 100px by 72px
c) Mix and match the headlines, body copy and images and filter out the combinations that don’t quite make sense
So if you had 3 headlines, 3 bodies and 3 images for a right hand side ad, you now have 27 different ads you can try.
d) Create basic wireframes for landing pages to support each category of ad campaign
Designing a winning Ad creative
There are no right and wrong answers, but here’s some advice. The Internet (as we all know) is flooded with advertising. Getting a potential customer’s attention is far easier said than done. On Facebook, this is particularly challenging since you are not only competing with other ads, you are competing with the newsfeed which has content that users are far more interested in.
So how do you create your own ads that stand out from the newsfeed?
One thing I like to do is collect amazing ads that I see when I’m using Facebook. I’m usually numb to ads, having seen and created tens of thousands of them, so if there’s a creative that I notice, it has to be good.
I keep a folder with hundreds of screen grabs of these that I use for inspiration. Not to copy, but to learn new or unusual techniques and image ideas. For example, there was one company that used a lens flare effect on a pretty mundane photo. But guess what? That lens flare caught my attention even though it was buried in a right hand side ad.
That showed me that simple Photoshop effects make a huge difference in making images that pop (and proved to me that you can do that without a lame technique like putting a fat red border around your ad – Facebook hates that btw).
|Free Exclusive Download: Click here for an exclusive step by step guide to Facebook advertising|
2. Campaign Setup – Tracking
This is all the “technical stuff” you have to do.
Never launch campaigns without proper tracking. Depending on your level of sophistication and budget, there are a number of options here. On the low end of the spectrum, you should be able to get by with the native reporting from Facebook and Google Analytics. Just take all those stats and collate them all together in a spreadsheet.
If you want to be more sophisticated (and have the budget) you can use simple tools like CPVLab to do this for you in an automated fashion. The campaign tracking market is a big one and you can even go as premium as tools like Marin software. But use your own needs, budget, and available resources as a guide to which direction to go.
Personally I like to go the free route.
Now you obviously should have a separate landing page dedicated to the campaign itself, usually that’s the best practice. There is a slew of services out there which can help you make a landing page (LeadPages.net and Unbounce.com are just some of them).
Upload your landing pages and confirm you have added the Google Analytics script correctly as well as the native Facebook tracking pixel script. Go through the steps in this knowledge base to create and add native Facebook tracking pixels.
3. Campaign Setup – Ad Creatives
Now it’s time to upload your ads. This can be a time-consuming process both as a user (Facebook power editor is not the most intuitive tool) and also because it can take anywhere from 1 to 48 hours for the Facebook team to approve all your ads. This is a reality of Facebook advertising though, so just plan for this in your schedule.
You’ll know it’s working if you see this happen.
4. Launch and Collect Data
I always launch a new campaign with a really small test (say the $100 you originally wanted to spend). This testing isn’t for ad testing, but for testing that all your tracking pixels and tracking scripts are firing correctly, your pages are loading quickly, Google Analytics is working etc’¦
If you’ve correctly inserted the Facebook tracking into your page, when you look at your ad performance you should be able to see more detailed user behavior data like this.
It’s better to take the time to make sure everything works, before you start really spending your hard-earned cash.
After your first day of spending, you should first check that you’re seeing numbers at all — clicks, click-through-rate, etc. It’s not unheard of for something to be missed, or even an ad being declined without you noticing causing all your impressions to be paused.
If you are getting traffic make sure that the Cost per Click (CPC) is something you’re comfortable with. Most importantly you should check that your traffic numbers from Facebook are matching what you’re seeing in Google Analytics.
Once everything has passed your tests, then you are good to go.
Cross your fingers, you are ready to spend. Remember the goal is to collect data. You need to get statistically significant data before you start making decisions.
It’s more than likely your campaign will not start off with a bang. That’s ok, and frankly it’s to be expected. Every good campaign takes a bit of optimization to get there.
The way to turn a campaign profitable is by running tests, over and over, and over again. You should consistently, and constantly run experiments into order to improve your campaign performance.
Your objective with each test is to incrementally improve the performance and profitability of your campaign.
Here’s a simplified example of how to iterate through a few variables to do this:
Let’s assume that you’ve launched your first campaign. Initially you might be at a completely negative 100% ROI and losing money, what do you do next?
- First, within the group of ads you’ve launched your campaign with, there must be one that is performing better than the others. The interesting thing about the Facebook Ad platform is Facebook will often do this for you by giving a disproportionate amount of ad impressions to one ad over all your others. Find that one, and we’ll use that to start optimizing.
- Test #1: Test that winning ad using 3 different images, but use the same headline and ad copy. Run these ad variations until you get a number of clicks on each one. Keep the best one based on click-through performance (usually better click-through will translate to lower cost per click as well). Improving your call to action is the first round of optimization and might bring you up to -50% ROI.
- Test #2: Test 3 variations of your call to action. Use the winning image + headline + body combination from Test #1. Again, let them run and keep the best performing ad. Now you’ve improved your ad performance over 2 test iterations, and perhaps brought your campaign close to break even at -10% ROI.
- Test #3: Test 5 landing pages that your winning ad from Test #2 sends visitors to. Continue the process, and keep the best one. Now you’re in the black at +25% ROI. This is the ongoing process required to optimize your ad campaigns. There’s not really much magic, just time and effort.
The biggest mistake is not optimizing at all. Some people will try to set it and forget it, assuming that traffic will normalize, or that it’s a bad time of year.
You don’t find winning campaigns, you create them.
When you’re running your experiments, you will want to split test (or A/B test) individual variables at a time. Here are some of the variables to optimize against:
Ads — Don’t focus on only the highest click-through rate. An ad could have an amazing CTR, but still lose money because it doesn’t convert. Sure, people will click on pictures of scantily clad women, but I’m pretty sure that has nothing to do with your business. Some images are going to get people to click, but won’t convince them to buy your stuff. That’s why it’s important to track.
Landing Pages — Different styles of pages, with images, headlines, buttons, etc.
Ad Bids — There’s no established protocol on the best way to bid. Experiment. Low to high, high to low, bid a certain percentage above or below the suggestions Facebook gives you. Try them all. To be honest, it’s more superstition than science from what I’ve seen.
Day & Week Parting— Day parting is running during the best times of the day. If you’re only profitable from 5pm-12am EST, then try running only during those times. Week parting is running only on the days of the week that you’re profitable. If you’re making money on weekends only, then try running only during those days.
Jason Dea has some sound advice around putting split testing into action. “Set a budget for yourself to test ad formats and audience segments. Once you find a recipe that fits, scale it.”
6. Analyze the data!
Now that you have collected real data, you can start making real decisions to make the campaign better.
If you’re tracking everything then you will have very detailed daily stats. You should also have been calculating your own metrics such as Cost Per Acquisition (CPA) to summarize all of the ad, page and click-through numbers. (Your CPA is calculated as Total Spend/Total Conversions)
If not already tracking, do it now. Spend as much time as you need to log data into a spreadsheet with exact spend, clicks, leads, conversions, etc. If you’re a visual learner like I am, you can convert this data into charts, to try to identify patterns that are emerging.
Which campaigns do you keep, which ones to you trash? Just because a campaign is making money doesn’t always mean you should keep running it. Does it have future potential? By this point you should have other campaign variations as well and you have to decide which ones are worth your time.
How can I start systemizing this campaign? Is there something here that can be scaled?
Just Do It
Hopefully you now have a rough idea of how and when you’ll want to dip your toes into advertising on Facebook. Spend the time to prepare, and then go for it.
Jason is Chief Operating Officer at Toocoo, a Toronto-based digital agency specializing in eCommerce, offering services ranging from design, integration and media buying, to their referral marketing platform, Forewards. Whether Jason is, in fact, “too coo” is yet to be determined. You can follow him on Twitter @threadyblock