Last Updated on April 4, 2023 by Alex Birkett
Capturing email leads is one of the primary goals of most content marketing programs. The money’s in the list, you get a million dollars back for every dollar you invest in email marketing, yada yada yada all the cherry picked statistics.
Anyway, you know it’s important or you wouldn’t have found your way here.
The way most people go email capture it is pretty rudimentary: throw up a popup with some seemingly tempting offer, and let the chips fall where they may.
Upon first glance, email capture seems straightforward, with little wiggle room. However, there are really endless possibilities for experimentation and creativity. Actually, I believe there are so many different ways to go about email capture, that it’s a bit overwhelming.
So, to help myself set up some parameters when auditing or optimizing an email capture program, I walk through these questions:
This is usually referred to as the Five Ws (or the Five Ws and How, 5W1H, or Six Ws). It’s commonly applied in journalism, though also in other inquisitive endeavors such as research and police investigations. I believe it can help you to set up an email collection program, but it can also help you audit an existing one and improve it. Without a guiding framework, it’s easy to get lost in millions of tactics, or worse, never know where to get started at all.
All that follows can be applied across industries – SaaS, ecommerce, personal blogs, etc. The specifics may be slightly different, but the game is the same. Emails are valuable for everyone, after all.
Who: Defining Your Target Personas and How to Reach Them
Sales people talk to customers. Customer Success talks to customers. But marketers don’t usually talk to customers. No bueno.
This is a shortcoming for many reasons, but for a concise argument, it’s because “who” you want to target helps you answer the rest of these questions. It all starts with the customer.
This “who?” question branches off in two directions:
- What is our target audience like (how can we model our target customer)?
- To whom will we actually show an email capture form?
The first question is a broader one; it affects every part of your marketing.
I know there’s a lot of bashing around personas, but it’s mostly because marketers have ruined them. The modern “persona” is almost a parody of itself. They’re created with no data, they include irrelevant or useless information, and they’re given silly names like “Meticulous Melvin.”
They are to marketing what the cheesy stock photo is to web design: a lazy placeholder for something that should actually have value.
All that aside, personas hold lots of value if you do them right. They’re a representative model – not 100% accurate, but an actionable approximation – of your target customer that you can use for various product and marketing decisions.
This is no place to dive into what I consider to be a good user persona creation process (summarized: use real data instead of made up stuff). But you should read this CXL article on the topic.
In general, before you create email capture offers, do some research on your audience and find out what they may actually respond to. It will save you tons of time and frustrations.
The second branch here is more local: to whom will you target your email capture form? You can choose to target offers based on:
- Traffic source
- Referral source
- URL/page targeting
- Number of pages views
- New vs. return visitors
- Mobile vs. desktop.
If you’re a Google Analytics user, this information could be found in both the “Audience” and “Acquisition” sections of your dashboard.
Now, the most basic implementation of this specific audience targeting is “everyone.” This is probably unsexy to personalization advocates, but oh well – you save on a lot of complexity by just throwing a static form on your blog asking people to subscribe. Most sites have some sort of static subscription mechanism like that:
When you do start running some A/B tests on your email capture forms – or at least looking into your historical content marketing analytics data – you may find that some audience segments are responding better or worse to different offers. This is when you might want to look into audience targeting.
Most popup tools you’ll use offer this sort of thing out of the box, at least with basic rules like device targeting (mobile vs. desktop) and URL inclusions or exclusions.
You can also build these rules out in a tag management platform like Google Tag Manager, if that’s what you’re using to fire your lead capture tools.
You can get pretty complicated with this stuff, but I always try to keep it simple, as complexity carries a management cost. On my own site, I basically just target all desktop users on their first visit (and their next visit afterwards, after a period of 2 weeks), after scrolling 50%.
I also have a few static forms on my site that everyone sees. There’s no special targeting.
It doesn’t need to be this simple though. For instance, when I was at CXL, we aligned several offers with different content categories, depending on the problem it solved (enterprise CRO program building, A/B testing ebook, CRO mastery guide, etc.).
HubSpot has a million content offers depending on the content topic, the blog where it lives, the desired conversion pathway, etc.
While I said your “who” is your starting point, in reality, it’s something that you’ll definitely learn more about after you create email capture forms and get data rolling in. No matter the pretty picture we paint with our personas, reality usually tells us the situation is different. As Mike Tyson once said, “everyone has a plan until they get punched in the mouth.”
In other words, analyze the data once you have it; it can help you learn which offers convert better or worse on which pages and with which audiences.
This stuff is quite easy to model out at a basic level, and you can get an excellent look at user intent and how well your offer is aligned. Just line up all the associated impressions and conversion rates, and see which ones are lagging:
As I showed in my article on content optimization, I like to model out projects based on the assumption that we could possibly bring every one of these up to the average conversion rate.
To do that, just calculate the average conversion rate (=AVERAGE(E:E)) and put that in a new column.
Multiply the impression count in column C by the average conversion rate, and you have a feasible goal that is hypothetically attainable. You can then put some conditional formatting on them to isolate only the potential uplifts greater than zero (i.e. posts that have a below average conversion rate and could therefore improve if you brought them up to the average).
Then you can project out which email capture locations would bring the most value if you focused on optimizing it.
Note: when you do a report like this, it doesn’t necessarily mean that an offer or targeting is under-optimized for a page; it may just mean the page and the traffic it brings has low or now intent to sign up. I find this is a worthy heuristic, however, to explore optimization opportunities.
Sometimes, I lament, you can tweak and optimize a page forever for marginal or zero gains, and you’re better off simply moving on to “warmer” traffic.
You can also spin through a few custom reports in Google Analytics to weigh out which segments seems to be responding better or worse. A super common and easy one to start with is targeting by device. It’s probably the case that you should be targeting mobile and desktop users differently, especially with email optins:
Intuitively, ask yourself, what audience or acquisition segments would make a difference or matter in terms of what offer they see or when they see it? Ask those business questions, and then dig into your digital analytics data to see if there truly are any meaningful discrepancies between segments. Then use that to form a hypothesis driven experiment, where you can maybe eke out some additional email captures.
No matter what route you choose as to whom you target, just make sure that’s it’s something you truly consider before throwing up some whack, generic popups. This often goes overlooked, but user intent and who gets your offer is just as important as what your offer is and when you deliver it. In fact, it’s difficult to craft a good offer if you don’t understand who you’re targeting.
What: Aligning the Offer with your Audience
I’m wading into murky waters with this one, because it’s still a problem I’m working on, struggling with, and chipping away at bit by bit.
You know your ideal audience. You have defined parameters for your audience targeting. You have some idea of what you do with email subscribers and/or leads as soon as they opt-in.
Now what the hell do you offer them in the first place?
Let’s back up a bit first, and define the basic stuff. What I’m really talking about here is sometimes called a “lead magnet.”
A lead magnet is an exchange of value for information. It’s the promise of an ebook, an email course, a quiz result, a discount, or just regular content updates, in exchange for an email address and potentially other personal information.
In this step, then, we’re talking about maybe the meatiest part of lead generation: your side the of the bargain. What are you offering this anonymous stranger in exchange for their kind offering of their email address?
As it turns out, this is a really hard problem with multiple dimensions:
- What format should your offer be (PDF, video, web content, other)?
- Do some formats convert better than others? Do some contribute to greater down funnel conversion rates (product signup, purchases, upgrades, etc)?
- How do you determine which audiences get which offers (if you’re doing different offers at all)?
- What formats produce the best qualified leads?
There are a million questions here, and seemingly no universal answers.
The best thing I can do here is give two insights that I’ve learned along the way:
- Learn, to the best of your abilities, about the user’s intent and match your offer the best you can.
- Continually update your beliefs and offer alignment based on what the data tells you.
I recognize that can sound vague taken solely as palabras, but I’ll try to get you an example or two to illustrate.
Learning about user intent
How are users coming to your page? If you don’t know that, head over to your Google Analytics account and go to Behavior > Site Content > Landing Pages, and then set up a secondary dimension of “Source / Medium.” This, at a high level, will show you how people are arriving at your site.
Let’s pretend you’re setting up a content offer / content upgrade for a blog post of yours that gets a lot of traffic. Most of the traffic comes from organic search. What to do? Look at the keywords that it’s ranking for (that are presumably bringing visitors in).
In our hypothetical case, we’ve got a page addressing the safety concerns of CBD oil. Here are the keywords it’s ranking for:
Now, candidly, we may not have needed to go through this process to get to the user intent, but it’s such a fast process that I figure, “why not?” And you might learn something new, like finding out a ton of traffic may be coming from a term you didn’t even know you ranked for.
In any case, knowing that people come to the site searching “is cbd safe,” and “is cbd oil safe,” we can cater our offer with this knowledge in mind:
- Maybe we offer a beginner’s guide to CBD oil dosing safely
- Maybe we offer a 10% discount off their first CBD order (if they order today!)
- Maybe we offer them a fact sheet on the health benefits of CBD
- Maybe offer them a quiz or survey that rates their CBD knowledge
There’s no right answer here; it’s something you have to consider in each individual case, stake out a rational strategy to get started, and then test and optimize with time.
Something of interest with all of this talk about “aligning for user intent,” is that there may be a good programmatic way to do this when you’re doing keyword research. IPullRank wrote an excellent post covering it (it’s quite technical). I’ve fought my way through about half of the process, so I can’t speak completely to its efficacy yet, but it seems quite cool.
If you don’t want to learn R and Python, though, you can do what most people do when aligning content and keywords to user intent: just eyeball it and make a gut decision. It’s usually going to be pretty close to accurate.
Or, let your creative juices flow. This is actually the part of the process where you get to have some fun, and most marketers don’t take advantage of it.
The world probably doesn’t crave another ultimate guide ebook; do something weird and attention grabbing. Offer a free vape pen with their first CBD oil if they order in the next 25 minutes (and if they tell you their favorite 90s music video). I don’t know, I’m riffing, but I think it’s fair to say that you’re not going to break down any barriers by swimming in the red ocean of the millions of ebooks and webinars out there.
Look at the data and adapt based on what you learn
Here’s the important part: just because you’ve looked at user intent and decided on an offer doesn’t mean it needs to say like that forever. Look at the data and adapt.
One thing I’ve worked on at HubSpot is “historical blog optimization.” We take a bunch of high organic traffic blog posts, reconsider their conversion paths, and try out new CTAs to see if we can get any wins. The interesting thing? The conversion rates are usually super variable, even though all the posts, in my mind, seem to be similar intent.
I wish I could show you actual data, but I’ll just make up data that looks really similar to show you what I mean:
This may look trivial – conversion rates vary, obviously – but it’s more surprising when you consider that, at least in my mind, all of these posts were super similar intent. They had similar keywords, were written about similar topics, and were published on the same company blog. I put the same CTA that lead to the same landing page, but got such different results.
The average conversion rate that your tool will likely spit out on reports, in this case, doesn’t even matter that much. One post brought in a .06% conversion rate, while two were in the 2.8% range. That’s so different!
Your results may look different, but I highly encourage you to pull these numbers, if for no other reason than to learn that the same offer and same conversion path can have much different results depending on the page. From here, you can move on and attempt to optimize individual pages and offers (again, my post on content optimization).
A quick example of an action I would take on this information is the following: let’s say I put a bunch of product CTAs on a bunch of blog posts. They lead you right to a product page, where you can sign up free. It’s higher intent than a related ebook, but lower intent than requesting a sales demo.
If I then go back and see that some posts are converting very well, I leave those alone for the time being.
But if i see that some posts are converting super low, say two standard deviations lower than the mean, I’ll consider swapping the CTA out for something lower effort – for example, an ebook on “email productivity.” Perhaps I’ll try to go for a simple email subscription instead of an actual product sign up or even an ebook signup.
I’ll usually play with offers on only the highest traffic posts, since that’s where you’ll see impact.
Sometimes, you’ll find out, the user intent on the page is just too damn low or irrelevant to craft a compelling offer. For example, if you wrote a post on “59 ways to increase your Twitter followers,” but you sell car insurance, there may not be a ton of ways to capture those emails.
If that’s the case, I find, it’s best to move forward on net new content and offer creation. Also, reconsider your content strategy and stop playing to vanity metrics like page views.
When: Beyond Simple Exit Intent Popups
When does your offer fire? This is where we talk about one of my favorite topics: behavioral targeting.
Behavioral targeting is just what it sounds like: delivering a given experience based on behavioral indicators (as opposed to demographic, psychographic, etc.).
It’s a broad term, but through lots of good work by companies like BounceX, behavioral targeting has mainly become associated with conversion optimization.
Even more specifically, it’s usually known now as mouse and scrolling behavior on a website.
The most common example of behavioral targeting is probably the exit intent popup. I’m going to go to GrowthHackers.com and click on a random post, and it’s almost certainly going to have an exit intent popup. Watch this:
First try! It’s super common, at least in the marketing space.
Another common one: scroll depth triggered popups. That’s how I’ve got mine set up. It triggers when you’ve scrolled 50% of the way down a page:
Pro tip: set up scroll depth tracking with Google Tag Manager. It’s super easy to do and you can then see how far most people are scrolling down on a given blog post. Your report will look something like this:
Only 324 out of 1081 hit 50% on this page, meaning perhaps I should consider a better triggering time in order to reach more of my audience?
Another common one: a popup upon first arrival. This is especially true in ecommerce, where you’re usually trying to quickly capture the attention of casual shoppers. Many stores use a discount offer to do so. This is an extreme example, but not uncommon:
Sometimes, your “when” doesn’t actually need to be explicit, active behavior like scrolling or moving your mouse to exit the page. Sometimes it can just be based on time on page. Often, marketers trigger opt-ins with some combination of all of these:
- After 10 seconds on site
- After scrolling 50%
- Excluding visitors from Hacker News
- Excluding mobile visitors
…And on and on. In my mind, it’s not so much the spread of your impressions but the accuracy of your targeting. The fewer uninterested or irrelevant people you can target, the better, simply because it’s a better user experience for everyone involved. Also, because you don’t want a bunch of unqualified and uninterested people on your email list. Read this CXL post on unsubscribing 83000 emails to learn why. More isn’t always better.
The best timing trigger, in my mind, is based on a behavioral signal that implies strong intent. For instance, if a user clicks on a link that says “download our free spreadsheet template,” there’s a lot more intent than simply throwing an exit popup with the same offer for everyone.
People usually call that a “content upgrade,” but it’s really just a prequalified click that triggers your form
Here’s a glimpse of one form of behavioral targeting I’m working on right now…
At HubSpot, we have tons and tons of “templates” blog posts. They cover topics like “sales email templates,” “follow up email templates after a networking event,” and “follow up email after job interview.” They get tons of traffic, and there’s lots of room to test out different conversion pathways and timing triggers.
Anyway, I’ve set up a test where, when a user copies text to their clipboard from a given email template, they get a message that tells them the message was successfully copied (kind of delightful, right?), and that we have a tool for templates, tracking, and automating emails (sell it!). It looks like this in action:
I think there’s a lot of creativity regarding the “when” of a content offer. I gave a few common examples and a somewhat more creative one here, but this is one area where marketers are continuously innovating. I think there’s a lot of room to optimize here, especially beyond the common “exit intent” and hit-you-right-away email popups.
Where: Placement and Real Estate
Where you place your email capture form matters a bunch, too. It’s hard to speak for specific websites, but Andrew Anderson, one of the smartest CRO people I know, had this to say about patterns he’s seen regarding split tests:
“Real Estate usually has a much higher beta and longer half life than copy changes. Spatial changes tend to be better than contextual changes for long term monetary value. Even in those cases you better be designing your efforts to see if that is true in this case and going forward.”
In other words, the space that website elements occupy on a page really affects user behavior and conversion.
Before you run off testing a million and one different locations for your email capture form, though, remember this CRO heuristic: meeting customer expectations is a simple best practice. Do what people expect and you’re most of the way there.
Users expect to navigate a site in a certain way, with the broadest categories branching out into specific filtering (e.g. Home > Category > Subcategory > Product). They expect clear, action driven CTAs, that accurately explain the next step. They expect lead generation forms to be in certain places on your site.
A few common locations include…
On the blog sidebar:
Bottom of the page:
On a dedicated “subscribe page”:
On a dedicated landing page (above the fold):
Lightbox popup:
Scroll box popup (or anything on the bottom right or left side):
My advice: use the expected placements before you start experimenting with creative real estate changes. People generally expect forms to be in those places, so don’t reinvent the wheel, at least right off the bat.
At CXL, we used to have quite a few badass and creative email capture forms, thanks mostly to working with BounceX, who crushes it with creative lead generation methods. They’re not up live on CXL’s site anymore, so I can’t get a screenshot, but we had:
- A hover over lead capture trigger. When someone hovered their mouse on the header image, it turned into a CTA.
- A left sidebar pull away that took you to the left side of the screen (hard to describe without an image, but it was cool and dynamic).
- A callout quote box in the middle of articles with Peep pitching an ebook offer we had.
However you do it, please try to refrain from doing that full screen takeover, welcome mat bullshit. It’s bad UX.
Why: The Most Important and Underlooked Question
In retrospect, I should have started the article with this one, but editing and formatting is for chumps.
Your “who” is definitely important, but you should probably define this question first: Why are you collecting emails?
What do you plan on doing with the information? Haven’t you considered what your “lead nurturing” experience would be (what a terrible phrase “lead nurturing” is).
Essentially, drawing out a customer journey map, starting at A and ending with Z, helps you formulate a smart and gentle plan to capture emails and information in a way that balances succeeding at your business goals with providing a good user experience. Considering your end goals helps you craft your content, the content offer, how many email capture forms you really, and how much information you ask for within your forms.
To that end, the whole “why” thing is a pretty big topic. It usually leads me down a rabbit hole of digging into core marketing systems and email automation flows, instead of simply the email capture mechanisms. I recommend starting out reading some materials on customer journey mapping and following it up with some good content on lead nurturing (there’s gotta be a better word for that?).
How: Form Optimization (and Moving Beyond Basic Forms)
So let’s say you’ve reached a point where you have a pretty pristine idea of whom you’re targeting, with which offer, when, and where on the page. You’ve also got a pretty sweet email automation workflow that’s converting leads into customers like gangbusters.
Another angle you can use to improve your email capture results it by looking at “how” you’re asking.
- How have you designed your lead capture form?
- How many form fields are you using?
- Can you optimize the sign up flow?
- Can you get rid of traditional web forms altogether?
There’s a level of creativity that expands beyond simple lead generation forms (give me your email, I’ll give you an ebook on the thank you page, etc.). Most of it starts with implementing some good form analytics solution to know your current form performance. I like using a dedicated tool, something like Formisimo. You can, however, also use Google Tag Manager to set up some pretty sweet form tracking implementations.
In any case, you’ll want to know when and where people are dropping off of your forms. This allows you to make intelligent decisions in culling form fields or simply reordering them to make the process smoother.
Form optimization is a massive subject. Should you use single or multi step forms? Single or multi column? How much information is too much to ask for?
Maybe it’s a cop out, but I’m just going to link to my favorite resources we published at CXL. No need to reinvent the wheel here:
- 13 Empirically-Backed No BS Form Design Best Practices
- 7 Web Form Optimization Insights from Testing Hundreds of Forms
- 14 Steps to Building Sign-up Forms That Convert
- Should You Really Reduce Form Fields?
- Form optimization (paid course)
Another topic is on the lead form itself. Nowadays, many companies are testing out bots or other types of interactive/conversational forms.
There are a bunch of tools you can use to help you design a chatbot/conversational form yourself (there’s probably a difference between the two, technically, but right now they seem to be used interchangeable. Here’s one such solution:
I haven’t tested conversational forms or chatbot much myself, but I can say that, generally speaking, I mostly hate chatbots. I’ve seen a few good ones, but most of the time I’d much rather just use a simple form. Or if I want to talk to people, live chat. Chatbots seem to be a weird hybrid model that is just unsatisfactory. Maybe it’s just been bad execution, but they all seem to be written with cutesy, chatty copy that I find annoying and patronizing. I don’t know, feel free to change my mind.
Something I do like, though: interactive quizzes and surveys. I think those are completely underrated as a lead generation mechanism. Think of Buzzfeed quizzes, but ones that ask for email addresses in exchange for the results. This Greenpeace one is a good example:
There are also voice forms and lead generation to think about. I don’t know enough about that to write about it, but it’s a thing. Read more here if interested.
Point is, the world is iterating on the basic idea of a form. I’m sure there will continue to be inspiring and interesting ways to collect visitor information in the future. If you’re just getting started, there are a lot of basics to cover first. But if you’re passed that point and want to start thinking about “how” you’re actually capturing information, there’s a fast moving world of innovations to look into.
Conclusion
Email capture is a huge topic. Frankly, this guide is probably too long as is, but I haven’t even covered lead magnet creation and the various types you can create, and I haven’t even touched upon things like GDPR, which do matter for information collection now.
That said, this guide should give you a core understanding of how to build out an email capture strategy, and it should give you more than enough ammo to know how to begin optimizing an existing set. Just ask yourself the questions every good journalist or police investigator (or Ludacris) does:
- Who?
- What?
- When?
- Where?
- Why?
- How?
If I’ve missed anything crucial, be sure to call me out in the comments! Hope this helps you capture some emails leads & subscribers.