When you’re writing a marketing plan from scratch, one of the most important aspects to consider is your market / industry. What is your target audience? Who are your competitors? What influencers are most important to reach out to? And what keywords should you be targeting in your advertising?
All these questions are difficult to answer and require research, but are critical components of crafting an effective marketing plan.
Luckily, we’ve made it easy for you to do this yourself.
We’re sharing our entire 2017 market research document with you, starting with our target audience research summary. This is the audience we’ll be targeting with our prospecting ad campaigns.
Because the best way to learn how to do market research is to watch someone else do it first.
Follow along with us as we show you exactly how we conducted market research for Ladder.
Let’s get started:
Need help understanding and reaching your target audience?
Every self-respecting digital marketer has done keyword research, but how many research their potential audiences on other channels? Unfortunately this practice isn’t widespread yet. Most customer personas are based off subjective judgements, not data.
Chances are your gut assessment of who your customers are is probably wrong. Even if that segment does exist, you might have chosen an extraordinarily narrow group that’s too small to sustain your product. It may not even be possible to target the persona you’ve defined. Your tunnel vision might cause you to miss another persona that would have performed better.
Luckily, just like Google’s keyword tool, Facebook has a free tool that helps us explore the data around our marketing personas. You can find it in your Facebook account in the main menu under Plan > Audience Insights.
If you read our Marketing Strategy post, you’ll remember we created a persona called ‘Steve Seed’ based on what we thought our audience was. Of course most companies don’t even get this far (the persona just exists in the CEO’s head), so writing it down was a good start. Now let’s see if it actually translates into a real audience we can target in Facebook.
Let’s start by filtering for demographics. Rather than just choose 28 year olds, let’s choose a range +/- 2 years to make sure we don’t narrow down the audience too soon.
With just these basic demographic filters, we’re already down to 100k – 150k people from the 250m people on Facebook in the U.S.
Let’s try and pull in a few of the likes / interests we mapped out, like the magazines, books and gurus they would be interested in.
We just dropped our audience size by another order of magnitude, but we’re getting pretty targeted here. Let’s add in income and education level.
Wow – another order of magnitude drop down to 1.5k – 2k. That’s without even narrowing further for job title. This might seem like a lot, but remember that when you’re running a marketing campaign you’re very unlikely to hit your entire target audience.
Say for example we managed to serve impressions to 30% of this audience over 30 days, and had a clickthrough rate of 1%. That’d only give us something like 5 clicks per month!
Not only that, but the smaller and more narrow the audience, the more expensive it becomes.
This hopefully illustrates how important it is not to just arbitrarily define a persona without looking at the data. You’ll be left with a tiny audience that can never support your growth goals.
So how do we do this the right way? First let’s start with data on what we already have; our website visitors.
That’s right; if you have the website pixel set up, you’ll actually be able to plug this into the Audience Insights tool and see information on who your website visitors are.
You can see in the top right corner I’ve chosen the custom audience that all our website and blog visitors fall into*. Straight away I can see that 68% of our audience is men, and most are between 25 and 34.
*Note: you can do this for any custom audience, so long as you have > 1,000 people. So for example you could upload a list of your customer’s emails and see insights about them.
If I click on the ‘selected audience’ column, I can sort to see which lifestyle group most of my audience associates with, and a hover over description of what that means. It also shows us how much our audience is over-indexed vs the general Facebook population.
You can also take a look at the relationship status and education level. In this instance you can see most of our audience isn’t married, and does have at least a college degree.
We can also get an idea of Job Titles that match our website audience. In our case it’s a very technical crowd because we over-index on IT job roles.
Just like with the lifestyle groups, you can sort this by volume and see what makes up the bulk of the audience. In our case it’s management, which you’d expect (and is great for us!).
Sometimes you can see what likes / interests an audience has affinity for, but for smaller audiences like ours it’s not always available.
Audience location is really interesting; looks like we were correct about our New York assumption, but we’re also really over-indexing for San Francisco.
Facebook also gives you some information on activities they take on the platform, and what device they access it from.
For us I was surprised to see we under-index for iPhone and over-index for Android – my assumption would have been the opposite.
Nothing too unexpected from a household perspective. Of course this is partially skewed, for example our audience is mostly New York and San Francisco, and people from those places tend to rent vs buy. You should also be careful with the match rate; for household size we only matched 2% of our data so it’s probably not to be trusted.
Finally you can see a little bit about their purchase behavior. This is more useful on the B2C side, but still can help you build up a good picture of what type of people our users are.
So how do we do this the right way?
First, let’s throw in our initial guesses at likes/interests, and see what Facebook spits out as related interests to make sure we’ve got it all covered.
So we have a movie we didn’t think about (Steve Jobs) a public figure, unrelated to tech that we wouldn’t have though of (Bill Nye), a number of new publications (TED, VentureBeat, Harvard Business Review, etc) and even a car (Tesla).
We can also see what Facebook pages our audience tends to like:
Here we have a number of likes and interests that were not immediately obvious, and therefore are audiences that could be potentially cheaper than the obvious ones your competitors are bidding on. FYI – this is essentially how Facebook Lookalike audiences work.*
*Note: With B2B we have to be a little more careful with targeting as our customers are more niche. So for us, we might use these unrelated but relevant ‘science’ audiences to narrow audiences that are too broad, but never as standalone audiences.
We can keep going with this; plugging in different likes and interests and seeing what’s related. For example if I plug in the entrepreneur-focused magazines I get the following likes/interests.
This has surfaced not just a couple more magazines I hadn’t thought about (Young Entrepreneur, Fortune), but new influencers (Gary Vaynerchuk, Barbara Corcoran) and other organizations (Udemy, General Assembly).
There are also more general likes / interests that Facebook constructs for you. For example there’s one for people who show interest in ‘Entrepreneurship’.
Though in my experience these audiences tend to work worse because they’re filled with groups that aren’t necessarily related to the theme you’re targeting.
You can obviously keep going with this for a while, adding different likes / interests you know are relevant, then finding related likes / interests that you hadn’t thought of / didn’t know about.
The next step would be to take the list of likes / interests we have now, and split them into different thematic segments, ready for us to test. Here are a few examples:
You can save these audiences directly from the audience insights tool.
The ideal audience size depends on your budget, but as a rule of thumb you want to be aiming for 1m-5m to really allow Facebook to optimize and drive performance efficiently. Anything below 100k is probably going to be very expensive, so try broaden it if possible.
The audience has to be bigger than you think, because you don’t show to the whole audience. Within the audience you pick, Facebook pulls out the people that are highly likely to convert (based on what it’s seen on your conversion pixel) and are also relatively cheap to reach.
A narrow audience gives Facebook less to work with, and performance suffers. Additionally, you’re more likely to burn out a small audience. Eventually they get tired of seeing your ads and stop clicking, which Facebook reads as decreased relevance, increasing your CPM accordingly.
Now you might be curious; how can we be sure we’re not just targeting the same people in different audiences? That would kind of invalidate our testing, skew our results and burn out our audience quicker than expected. The good news is that Facebook has an audience overlap tool!
If we compare the general Entrepreneurship audience with the Publisher audience we put together, it looks like there’s a 22% overlap.
Let’s add in the Book audience and see how that compares.
Wow – only 3% overlap. Wait a second; this means that only 3% of the Entrepreneurship audience is in the Book audience; let’s flip this around and see it the other way.
Ah ok – the vast majority of the book audience is already hit by Entrepreneurship, and quite a few of them are covered in the Publisher audience too. Probably not worth running what was already a small audience in the first place.
We can also see overlap with our custom audiences; for example our website visitors.
20% of our website audience fall into our Entrepreneurship targeting, whereas only 1% and 7% are in the Book and Publisher audience respectively. This might be a sign they aren’t as relevant as we originally thought.
Now what if you see your audience has irrelevant likes / interests and you want to narrow it and make it more relevant? For example we probably don’t want some of these self-improvement pages included in our Entrepreneurship targeting.
We can actually remove them, but not in this tool. We have to move over to the audience section of the ads manager.
If we click ‘exclude people’ we can add in whatever likes interests we want as negatives.
Now this audience is still really big, at 19m – probably too broad to perform. So let’s narrow it further based on some of the trends we spotted in our website visitor audience. We just click ‘narrow audience’ to filter only for the likes / interests we want.
Now we’re down to 1m of that larger audience that we’re much more confident would convert. Another favorite trick of mine (only available for U.S. audiences) is to narrow this audience down to people who can potentially afford a high value product like Ladder.
If they don’t have an income of over $100k they probably aren’t in a position as an entrepreneur to afford a marketing agency yet, so this is a great proxy for ability to buy our services.
Remember to save it as a new audience:
We can obviously keep going with this, but I think you get the idea. Planning these things out using the actual data Facebook provides us can help us avoid picking an audience that’s too narrow or broad, surface any poor assumptions about our personas and can help us spot new audiences we never would have thought about.
Now we’ve used Facebook specifically for all these examples, but you can repeat this process for any channel. Most of the top self-serve platforms trying to get you to spend money will have some sort of way to estimate audience size, though most don’t have an audience builder as advanced as Facebook’s one. Often it’s just a case of building a campaign and seeing the number it spits out. We’re about to go through an example of this approach in the next section.
The process as we’ve described it so far works best for B2C; we’re lucky that a lot of our audience is very active on Facebook, but that’s not always the case for some B2B businesses. The answer can be to modify the process slightly to zoom out from a specific person’s likes / interests, to targeting the employees of companies that have the right attributes.
For B2B businesses who need to target by company attributes, there’s no better platform than LinkedIn. Because people enter in company information in order to get hired, meet potential clients and network with each other, the quality of the data tends to be much better than on Facebook. This increased accuracy and scope of targeting comes at a price; we tend to see anything from 5x – 10x the CPC on LinkedIn vs Facebook.
As we know from an analysis of our customers, we’re ideally looking for a business with 11-50 employees, in the computer software or internet industry. In order for them to have the authority to sign off on us, we typically need to be talking to VP level or above, and our customers have all been in a marketing or product role, or they’ve been the entrepreneur who founded the business. All of this targeting (and more) is available.
On LinkedIn you’ll find that the audiences are much smaller; our audience of 130k is right around the sweet spot, but you’ll still be fine above 20k-30k. You might even want to break out marketing roles vs product roles vs the entrepreneur, as they will respond to different messages. Once you’re happy with your audience, LinkedIn allows you to save it and re-use it later.
Should I want to make a similar audience but change a few key components, I can re-load it and save it as a new one. For example, say I now wanted to target the fashion & apparel industry; another area where we’ve had success.
In this case 18k is relatively small; I know from memory the ecommerce businesses we’ve worked with have all actually been smaller than our average client. This is potentially because they use Shopify and things like distribution. In this case we might want to adjust to this reality and target smaller businesses than we usually would.
Or we could branch out from fashion to other ecommerce style industries.
As you can see; the sky really is the limit here; you always want to be testing new audiences to see if something works better than what you already have. But before you plan to test the audience, you should always be checking the size in the platform. This will help you avoid campaigns that are too narrow and expensive, or too broad and inefficient.
Exploring these platforms will often give you ideas you hadn’t thought about, or didn’t know were possible. The major ones tend to update something every single week, and if you spot it first, you’re at a huge advantage to your competitors.
That’s it for the Target Audience Research portion of our market research document. Stay tuned — up next: Influencer and Competitor Research!
Need a marketing plan custom-built for your business?
Sign up to our newsletter to stay up to date with all the latest movements in the field.
More from Ladder Blog
The art world is being disrupted by generative AI, and artists aren’t happy. Generative AI models like DALL-E, Midjourney, and Stable Diffusion were trained by scraping millions of images from the internet, without permission. Artists claim this violates copyright, while the companies training these models argue this falls under fair use.
Read More →In the world of digital marketing and data-driven decision-making, creative testing is a pivotal tool in achieving business growth. Gone are the days of relying on gut feelings or guesswork; now, business decisions are powered by data-validated insights, meticulously collected, analyzed, and validated. This transformative process empowers businesses of all sizes, from established enterprises to budding startups, to thrive in an ever-evolving digital market. This article looks at the practical applications, challenges, and innovative transformations associated with creative testing, offering you valuable insights and actionable strategies to implement in your own digital marketing efforts for achieving growth and success.
Read More →
