I recently re-read the excellent book, The Business of Expertise, by David C Baker. One of the things he talks about is the ability to demonstrate your expertise in a particular subject through being able to come up with 20 facts about your specialist subject matter that even people engaged in the field might find they didn’t know. (He describes this as the equivalent of a military instructor or sports coach getting someone to ‘drop and give me 20’).
Once I’d done the exercise for myself, it occurred to me to turn it into a blog post re: the kind of insights I’ve picked up from my experience promoting clinical trials using Facebook Ads.
1) Facebook doesn’t care whether you spend money with it or not. This concept is one of the hardest things to grasp for ‘old school’ businesspeople and marketers – who believe that spending advertising dollars with Facebook should give them some kind of authority over the organisation. In reality, Facebook’s whole ethos is built around keeping its users happy, not keeping its advertisers happy. And with 10 million advertisers using their service, the people at Facebook really don’t care if you spend money with them or not.
2) The ability to share adverts is one of the key drivers of registrations for clinical trials. Even if someone in your target audience isn’t on Facebook when you’re showing your ads, they are likely to have a friend or family member who is. That person can share the ad with others, thus helping get your message in front of the right people. As well as this, if some members of your target audience are not on Facebook at all, they will almost certainly know someone who is. That person can see your ad and alert their connection regarding the existence of the trial. Which means you can reach both people on Facebook and people not on Facebook with a Facebook ad.
3) Targeting within Facebook is not just based on demographic information and the location of the user. Interest targeting was pretty much Facebook’s original ‘killer app’ – enabling you to target people based on their behaviour online. (ie Facebook tracks what people are doing online and can show them relevant adverts accordingly – something you can tap into for promoting your trials).
4) Point 3 notwithstanding, Interest targeting is only half the story when it comes to targeting people on Facebook. The actual content of the ad itself is a form of targeting as well. For example, when you’re browsing magazines on a newsstand, you will likely ignore the headlines for topics you’re not interested in and be drawn to the magazine headlines and images that relate to topics you are interested in. The same applies with Facebook ads, where if you or someone you know is living with a particular condition, the mention of that condition in the ad headline will likely draw your attention.
5) As well as demographics, Interest, and content targeting, Facebook allows you to develop Lookalike audiences – based on people who visit your website or an existing contact list. Which means that, if you have a database of people who have been in touch with you to express their interest in a particular clinical trial – either through visiting a relevant page on your website or through having submitted a contact form – you can use this audience to develop a new audience of people that ‘look like’ your ideal target audience – allowing Facebook to show your ads to people who are more likely to be interested in your trial.
6) Another of Facebook’s ‘killer apps’ is the ability to retarget people with your ads. We’ve all seen adverts that seem to follow us around the internet. For example, if you look at a particular garden tool on a shopping website, you may see lots of adverts for the same tool – or a different brand – when you’re next online. This is an example of Retargeting in action. For the purpose of patient recruitment, there are many reasons why someone might not go through with submitting the registration form – the phone or doorbell ringing, becoming distracted by something else they’re viewing online etc. Which is where showing them an ad for the trial when they’re next online might be just the reminder they need to go through with the registration.
7) The ‘killer apps’ and targeting methods mentioned in the previous points actually forced Google to ‘up its game’ when it comes to targeting, such that their system is now much more similar in functionality to what Facebook can provide.
8) You don’t have to send people away from Facebook to a landing page in order to capture their contact details in a registration form. Facebook Lead Ads allow people to express their interest in your trials with a simple button click. The form can be pre-populated with their contact details, so all the person has to do is click the Submit button. This way they don’t have to leave Facebook, which leads to a higher volume of leads being generated. The quality of these leads isn’t generally as high as for leads that come from a landing page that features trial exclusion criteria. But the quantity can make up for this – and ultimately you’re targeting the same people anyway.
9) Swift follow up of leads is the key to success for converting people from initial enquiry through to randomized patients. Within 24 hours is what you should be aiming for – with ideally a follow up phone call being made within an hour of the person submitting the registration form. (Note: if your leads have come from Facebook Lead Ads, as in point 8 above, you will need to do all the screening questions in your follow up call, as Facebook does not allow health-related questions in Lead Ads).
10) Long forms on landing pages are generally thought to put people off. However, in the field of patient recruitment for clinical trials, people have shown they are prepared to answer a large number of quite in-depth questions. I’ve seen as many as 25-30 questions on a web screening form work well, without seeming to put people off by the number of questions they have to fill in.
11) Despite what web developers and techie people might try to tell you, I’ve found that when tracking conversions online – ie the number of successful registrations or contact form submissions – it’s always best to have a separate Thankyou/Success page. Identifying that the tracking code or pixel has been fired when someone visits this page is always more accurate than any other method of tracking conversions.
12) Single image ads are still the most widely used and effective form of advert on Facebook. Though video ads are certainly valuable and are gaining in popularity all the time, ads featuring a single image have consistently outperformed any other type of ad for patient recruitment campaigns.
13) You can enhance the results of your Facebook Ads campaign by complementing it with a Google Ads campaign. There are always going to be people who will see your ad on Facebook and try to check you out on Google for credibility. Having a ‘brand name’ campaign within Google Ads – based on the name of the Facebook Page you’re showing your Facebook Ads through – is an inexpensive and potentially valuable way to capitalise on this kind of ‘credibility search’. (You may want to link these Google Ads to a landing page relevant for the specific trial you’re advertising on Facebook, in order to maximise the potential results).
14) Your Facebook ads for clinical trials will attract comments from people on Facebook. In particular, they will attract the type of negative comments you might more readily associate with trolls on Twitter. References to ‘guinea pigs’ or ‘Bill Gates’ have always been around, but have certainly increased during the Covid pandemic. Somewhat unbelievably, the most common type of comment I’ve seen on ads for Alzheimer’s trials has been of a supposedly humorous nature – such as “I can’t remember if I applied or not…”. Hiding these comments can work to stop them putting other people off. Or, and perhaps more usefully, responding to them in a professional manner can also be useful for boosting credibility.
15) As well as negative comments, your ads will also attract positive, and sometimes even uplifting ones. It’s well worth responding to these comments, too, as the people making them may be suitable for becoming advocates for your trials, or be able to provide a testimonial.
16) You can use Facebook Ads for building a database of potential trial participants. If you’re not promoting a specific trial, you can avoid the necessity for IRB/EC approval for your ad content. Thus enabling you to target people with a specific condition, get them to register their interest, then contacting them in the future when a suitable trial comes available.
17) Talking of IRB/EC approvals, Facebook’s own rules on content are equally as strict when it comes to healthcare ads. For instance, you have to be quite generic in what you say and not highlight an individual’s personal characteristics. Such that ‘We’re looking for people with psoriasis’ should be approved rather than making a direct assumption about something personal related to the user, eg ‘Do you have psoriasis?’ Further to this, you’re not allowed to mention prescription drugs by name in your ads or on your landing pages.
18) The majority of web traffic – and in particular Facebook traffic – now comes through people using smartphones. For this reason, you should ensure your landing pages are mobile-friendly. (Which is pretty much the standard nowadays anyway). And if you have a screening form with multiple questions, you might decide to check the option to only show Ads to people browsing their phones while connected to wi-fi. This allows you to target people who are more likely to be stationary in one place and thus have sufficient time to fill in your forms, rather than while they’re on the go.
19) Despite the obvious sophistication of Facebook and other digital advertising platforms, the type of A/B or Split Testing that would be familiar to olden day advertisers such as Claud C Hopkins still forms the basis for every successful digital ad campaign. ‘Beat the control’ is still the most useful tool for improving your Ads’ performance – a tool with its roots in the principles of tracking coupon advertising from over a hundred years ago.
20) Building on these principles of Split Testing and ‘Beat the Control’, Facebook’s machine learning capabilities will always be able to far outstrip those of any human being. With billions of data points to work with, Facebook’s internal algorithms are an extraordinarily valuable tool that you should tap into with your patient recruitment ads. Giving Facebook’s machine learning a head start by setting things up well in the first place, then managing them in the right way ongoing, is how best to make use of the power of this tool. It’s by adding the human traits of creativity and experience to Facebook’s machine learning functionality that you’ll achieve the best possible results overall.
