Your complete Guide to Facebook Advertising for Local Businesses in 2017: Local Facebook advertising for small business will typically differ from the ‘best practice’ taught by the thought leaders of digital marketing in 2017; read this blog from Ed Challinor to learn more.
For our guest blog, we invited Ed Challinor, the co-founder of Smileworks, Liverpool’s most popular dental practice, to write about Facebook advertising for local businesses.
Smileworks was founded on seed capital provided by MSIF Investment Director Paul Humphray.
Ed is a self-taught local digital marketer who’s not only built his own business on Facebook but also helped a number of other small businesses in his network succeed. He was a Beta member of the “Certificate in Digital Marketing Practice” and has spent more than £50,000 on local paid traffic in the last 3 years, run more tests than Apeture Science, amassed thousands of beautiful cosmetic dentistry patients and grown Smileworks from zero to over 1M turnover.
Today Ed has some interesting distinctions to share with any budding entrepreneurs who want to follow suit and grow a local digital marketing suite with Facebook as the focus.
Why listen to me?
This one’s easy. Most of the content you’ll read online is written by multinational businesses who want to sell you their products. I’m not trying to sell you anything (unless you want to come and have Dental Implants that is) but instead genuinely want to help small businesses in Liverpool grow. I don’t want you to make the same mistakes we made over the years and genuinely care about your success – especially if you’re sponsored by MSIF!
We’ve also got a fair bit of experience and our Facebook ads have reached over 830,000 people in the local Liverpool area and we’ve had more than 250,000 clicks and grown our lists to well beyond the 10,000 subscriber mark.
Our CPA (Cost Per Acquisition) is now so low that we’ve attracted the interest of media giants like Virgin.com, Disruptive Advertising and Eventbrite to talk on subjects close to our hearts like digital marketing and talent acquisition.
Problems with Local Facebook Advertising
The first problem is a knowledge problem. All the marketing gurus are writing about information products. Rarely will you find anyone who really knows what they are doing who actually sells physical products or services. There’s a massive difference between selling ebooks or training and getting people to come to your geographical location and buy things or have treatment.
Facebook marketing was brought to the masses by the posterchildren of information sales like Jeff Walker and Ryan Deiss. And sure they’ll have sections in their inventory relating to ‘bricks and mortar’ or ‘retail’ but it’s not their primary focus and more often than not their teachings will work for local businesses more by fluke than design.
Local advertising is an art. And what we’ve done at Smileworks is consumed the huge amount of learning and data, tested it to death and teased out the elements that work and binned those that don’t.
Problems with data integrity in small audiences
OK so those are some big words but stay tuned because this is super important.
Facebook has come up with some wonderful ways to help us with targeting. Unfortunately you are severely limited in your choices when it comes to local campaigns. The reason for this is in the numbers… There simply aren’t enough people in a local Facebook audience to maintain the required data integrity for a successful campaign.
Facebook are great at Big Data and a customers interests on Facebook are derived from a whole host of sources, not just what you write on your timeline. They also have classy algorithms like the ‘lookalike audience’ feature that uses a multitude of data to produce new and exciting audiences from your existing customers or email lists.
But wait! This all works great on a national or international campaign with millions of users, but when you get down to a few thousand the wheels start to come off. There just aren’t enough data points in a small user base for the advanced Facebook targeting methods to yield any useable results for most small businesses. This isn’t Facebook’s fault, it’s just what happens once you take a smaller audience sample.
To show you where I believe these data problems originate, let’s take the lookalike feature as an example. What we’ve found is that Lookalike audiences simply don’t perform in any of our tests at a sufficiently positive ROI in local populations.
I’ve tested this by creating lookalikes out of our customer lists, high value customers, our favorite customers, our email signup lists, and other custom audiences – just about every way possible. I’ve tested all of it and the CPA never beats the targeting options we will discuss below. Simply, lookalike audiences are lot effective for Facebook targeting.
Local Audiences by the numbers
So what do we know about our local audience on Facebook? Let’s try to surface some of the issues by looking at our audiences in detail. Here are the numbers:
New Zealand 3,000,000
So here are some big and small countries and also some big and small towns. Once you start adding interest targeting (let’s say “Dentist” into the mix, the results look very different:
New Zealand 39,000
So if you’re a dentist (or any service provider) in Stafford – a moderately sized market town in the Midlands (where I grew up), have you got to make do with a target group of 710 people? Any marketing 101 will tell you that 710 is a pretty poor reach by anyone’s standards – especially when you think that many of that 710 will actually be dentists, therapists, hygienists and not actually patients! But more on this later. You can knock about another 10% of users off these figures if you only want to target English speakers – which is probably a good idea for an English campaign.
So what’s a digital marketer to do when the leading sources of information tell you that the targeting ‘sweet spot’ or perfect number is between 500,000 and 1,000,000 users? And that’s their idea of ‘laser targeting’ by the way. Be sure I’m not criticizing the numbers – they are correct, but it should be pretty clear why this isn’t going to work for local campaigns.
But wait? Can’t you just figure out what their ‘laser targeting’ number is as a percentage of their audience and do some maths to get to your ‘sweet spot’ number? Good thinking young padawan. But that doesn’t work particularly well either – since you may well end up with 710 users again. But it’s the margins for error in Facebook’s ‘interests’ that is the crux here – the users with dodgy interest profiles and the users who are outside the curve become an unacceptable percentage of the audience once you get down to the smaller numbers.
But don’t worry, there are ways around this…
Fixing Local Interest Targeting Problems
I’d always go with demographic targeting and behavioral targeting over interest targeting for a local audience. I say this for two reasons. First, with a smaller number of users you are going to want a broader audience anyway. Second, interest targeting groups get more and more questionable the smaller sample size you look at. In fact any sample size has a whole load of users in there who are certainly not your customers.
Do this yourself in the ads manager – it’s a great exercise.
Let’s assume you want to target people who want to buy golf clubs in Liverpool. So you select ‘golf’ from your interest targeting options and up comes 6,000 users. Pretty good right?
Well now use the ‘narrow audience’ tool to find out how many of those are actual golfers and how many work at golf clubs or went to the open? Or how many of them work for companies who make golf clubs? Then narrow out all the people who like Tiger Woods. Or are interested in his relationship problems? Many of these individuals have no interest at all in buying a set of golf clubs. But they show up in the audience. And with some more smart audience narrowing your 6,000 users soon turns into 710 and you’re back to square one!
Local behavioral and demographic targeting
Demographic and behavior targeting gives you a much wider audience and excludes all those not interested in your products. It also gets your ad shown to more people who might be interested in your actual products.
A fun game (for serious geeks) is to reduce the audience down to about 1/5 of your local users using just the demographic and behavior options.
There is no local sweet spot for Facebook advertising but if you want your ad to remain viable for at least a few weeks at a positive ROI then you’ll need a few thousand users in there at the very least.
This takes some critical thinking. My mentor, Dan Bradbury talks about this a lot. He quotes a source saying that ‘business is an intellectual sport’ and that you need to spend less time running around fighting fires in your business and more time thinking about who your customers are. So I encourage you to sit down and really think about tightening up your demographic targeting and trying to go it alone without Facebook’s supremely dodgy interest targeting options muddying up your results.
At Smileworks we’ve got this down to an art – and if you send me an email I’ll tell you what targeting groups we use and what demographics will get you the best success for a number of different sectors. You can find my details on our website at www.smileworksliverpool.co.uk and you’ll need to email me from your commercial accounts to help me not get spammed.
Also think about what your competition is doing. With huge international audiences it doesn’t really matter what your competition is doing. You may see an increase in CPM over Black Friday or lumps and bumps due to seasonal variations. But it’s not really a biggie if you know the right bidding strategies. We find that our seasonal fluctuations coincide in a strongly negative correlation to the normal retail promotions calendar. So when it’s Easter or Christmas or Black Friday we see a drop in activity. Conversely in the quiet times we see spikes. Could this be because our competition in retail is out bidding us on Facebook and scatter gunning our audiences? Or maybe it’s because customers don’t want dental work at these times of the year when they’re seeing friends? I guess we’ll never know but the important thing is we’ve pegged the trend and can use it to our advantage.
I love animals. And I’ve always said that in Facebook Advertising you’re not a Woolf or a Lion. Unfortunately most dentists and entrepreneurs are sometimes a little arrogant and will want to ‘dominate the platform’ or make a big splash. But on Facebook (on any platform for that matter) that’s a strategy that’s sure to lose you a load of money. I think of myself as a fox or a squirrel. Ferreting around looking for the opportunities and exploiting them to my advantage. You need to make yourself 1% better in 100 different ways rather than try to make yourself 50% better in only one area. From a modelling perspective the first is far more likely to succeed. And I like mathematical models because it’s hard to make them lie.
But if there are two main players in your town then you need to be researching what they do, who they target and what sorts of customers they attract.
Diversity of Messaging
Facebook campaigns must be enormously diverse to succeed. You need to break your audience up into many many demographic and behavior-based segments. I’m not talking about ‘market segmentation’ in the classical marketing sense but a Facebook specific sort of segmentation of your audiences and a tailoring of your message to each.
So to give you a flavour of what I mean, use the demographics you’ve chosen to tailor your messaging to those individuals. So let’s say you’re selling gym equipment. Men and women want different things from the gym and they certainly want to look different in the gym. There’s absolutely no use creating an ad on Facebook for both men and women. Instead you need to create two variations targeting the different sexes.
Facebook is great because it allows us to create ads in minutes. So when I see generic ads served to me it makes me sad. You should be featuring images tailored to a customers age, sex and social status. You have all these amazing data points so use them!
I get retargeted a great deal on Facebook by apps that might help me in my job. So which do you think I’ll click out of one ad that says:
“Discover our amazing widget that helps you run better tests and get more customers"
“Calling dental marketers in the UK: if you want to run better tests to attract more Implant patients then take a look at our amazing widget”
Obviously the second is going to excite me. And that information is not only available on my Facebook account, it’s even information that can be dynamically inserted into the ad with certain technologies.
Here at Smileworks we have some pretty laser targeted copy and creatives. Just like in fitness, men and women want dentistry for different reasons. And we make the most of this. Women want to feel ‘beautiful’ but imagine being a guy and having Facebook ask you if you want to feel ‘beautiful’? I don’t think so! Men want to be successful, confident and handsome – not beautiful.
So here’s what we’ve learned
- Be careful of interest targeting in small local audiences
- Use the ‘narrow audience function to determine where your audience actually is
- Test demographic and behavioural targeting for the best results
- Use tailored and diverse ads and run them to each individual demographic audience
We hope you enjoyed our little round-up and learned something about this remarkable advertising platform. This was a pretty advanced dive into the platform so if you want the basics check out Digital Marketer Labs and John Loomer Digital.
And as always send me an email if you’re a small business starting up in Liverpool wanting to run some Facebook ads effectively and with a positive ROI. I’ll help you the best I can – find my contact details at www.smileworksliverpool.co.uk