A (not) cinderella story of how we optimized a B2B startup lead-gen campaign

How we optimized a B2B startup lead-gen campaign
About G2M team

We offer 360 degrees global startup marketing services, from strategy to implementation, tailored to startups in 'post-seed to round-B stage' with none to small internal marketing teams. Our services include CMO services, PR, content marketing, social, lead-gen, ABM, paid media, marcom and a lot more. Learn more

This is the story of an optimization effort around a gated content landing page (that is, a landing page that requires email signup before the download can start) for a B2B tech company in the cybersecurity arena. Unfortunately, its name must be kept undisclosed.

When marketers work on optimizing a landing page (which is quite common), they actually try to increase both traffic to that page, and conversions to leads. In this process, they use proven methods and helpful tools such as analytics, landing page templates and more.

As gated content promotion – webinars, gated video, a white paper or an ebook – is (still) one of the most successful tactics to acquire quality leads in B2B marketing, landing page optimization is an important task that marketers need to master.

Usually, gated content landing pages are promoted in several marketing channels (blog articles, social channels, paid campaigns and more), and optimizing the entire conversion funnel can save a lot of money.

We all read Cinderella stories

“A single change led to a 300% increase in conversions. We closed the day and when we opened the mailbox again in the morning we were shocked to see hundreds of new leads that included our best target prospects.”

I don’t believe in these stories. More importantly, I usually can’t learn anything from them.

If what you’re after is another Cinderella story about a bunch of brilliant superstar marketers that made a single landing page skyrocket traffic by 3000% overnight, delivered thousands of top-quality leads within a week, and managed to make the entire sales division worship them ever since – click back. You won’t find it here.

If, on the other hand, you want to learn how things work in the real world, you will find this case study valuable.

In reality, marketers sweat over-optimizing their lead-capture landing pages, and improvements are slow to come.

If you ask me, B2B startup marketers struggle even more due to the limited traffic they usually work with. The audiences that B2B startups target is (usually) much smaller than in other industries. They struggle to reach enough data points, sufficient enough to run any optimization activities. Sometimes they can’t even run decent optimization activities until their campaigns run long enough (which can take quite some time in cases of low promotion budgets) to collect the necessary insights

This is a good description of our day to day life, as we mainly work with young, seed to round B startups in the B2B industry. Below I describe a specific optimization effort that we’ve recently gone through and may help you gain a few actionable insights.

Here goes:

A couple of weeks ago we were working on an ebook for one of our startups (unfortunately, we can’t tell you which one, which is why we also had to blur some of the visuals included in this article).

Anyway, after the content was ready we had an internal debate on whether we should use the marketing automation landing page (LP) template (Hubspot in this case) or create a WordPress landing page as part of the website.

First, you should know that there are pros and cons to each approach.

True, building a LP as part of your website may require some initial graphic and dev resources, but you end up with a lot more control over the way your landing page looks and behaves.

Building your landing page based on a marketing automation template is perceived as easier, because there’s no need to rely on external designers and developers. You are, however, limited by the template (in terms of design and settings).

That’s why I usually prefer to create a LP template as part of the websites I build. When I inherit a website, I decide on a case-by-case basis.

In this case, both timing and budget constraints made us build the landing page using the marketing automation platform’s template.

That was the starting point (see below). It didn’t wow us, but we wanted to move on quickly and decided it was good enough for the time being.


The landing page was published and promoted both organically (via direct channels, at no cost) and using paid campaigns.

We had to wait for sufficient data before we could reach any conclusions about the performance.

When enough data was gathered, we saw that the page was converting at ~20% across all channels – paid, direct, organic and others… That meant that 20% of the page visitors left their email in return for the content, turning into leads. A detailed analysis showed that at some point, paid campaigns showed slightly over 11%. That is low.

We didn’t feel over the top

So the overall performance was not bad, but it wasn’t amazing, either. Paid channels’ performance was concerning.

Since we believe in promoting content only in ultra-focused quality channels and communities, and since we run campaigns based on lists, and since the content has proven to be an interesting and appealing one, we expected more.

We started thinking that maybe the root cause is that the page just wasn’t good enough.
We were on the verge of recreating the landing page. After all, if the landing page doesn’t show amazing results, it’s time to work on its structure, copy (text) and creative (visuals), right?
Well, not just yet.

First thing first – find proof for your hypothesis

We first wanted to be sure that the fault was indeed in the landing page.

All we had at this point was a general conversion rate of ~20%. It didn’t provide enough evidence.

We decided to start by looking at some data and begin optimizing all steps in the prospect journey, prior to working on the landing page. We asked some specific questions:

  • Are we really targeting the right people? Since we already had numerous campaigns running on LinkedIn and Facebook, we had some evidence for the quality of our audience. Still, we always double-check (and so should you). It happened before that Facebook or LinkedIn was buggy and messed up a campaign’s targeting, which led to lame results. But this time, after checking, we couldn’t find any issues. We were targeting the right people. As a strategic decision, we usually hunt for more expensive but higher-quality clicks, rather than cheaper ones. This was our strategy in this case as well. It meant reaching out to an ultra-targeted and therefore a relatively narrow audience. This strategy leads to pricey impressions and clicks, but conversions are usually high and that’s the payoff. It’s another reason why we weren’t satisfied with a ~20% conversion rate.Before I move on, just a quick tip – UTMs (in paid and direct channels campaigns) will help you explore the quality of the traffic in Google Analytics. In this case, we used UTMs to measure the behavior of the audience and ensure that prospects were indeed exploring the landing page as they should have (based on the “time on page” measurement).
  • Are the ads satisfying?
    We get clicks. If you get clicks, it means that your ads are appealing, right? Yes, it makes sense. But it’s not only about being good looking; it’s about being fully aligned with the landing page. If people click without expecting to get to a sign-up download page, they are surprised, and not in a good way. This will probably lower your conversion rates. I, therefore, recommend that you be very clear with your ad messaging. We did finetune the ads a bit, but we thought that they were OK, to begin with, and the root-cause lays elsewhere. One tool that I love using is The Emotional Marketing Value Headline Analyzer. It’s a free tool that lets you test different variations of your headline. It simply presents a percentage score telling you whether your headline is more intellectual, empathetic, or spiritual, and what emotional marketing value it creates.
  • Are there major differences between different traffic channels?
    Attention, please. This one is important and tricky. We decided to segment the data and look at the performance of each channel and medium: organic, direct and specific paid channels. We figured that if all channels perform the same, then it must be the landing page. If there are major differences – well, we need to think. We did find some major differences.
    Different channels presented different rates. More accurately, while LinkedIn conversion rates reached 30%, and both organic and direct channels didn’t fall far behind, it was Facebook that dropped the averages.Now please don’t think what many B2B marketers think: “of course, Facebook doesn’t make sense in heavy B2B tech industries. The prospects just aren’t there”.We have endless proof of the effectiveness of Facebook B2B campaigns. We run great demand gen activities using Facebook, and in fact, many times we enjoy the fact that our competitors just don’t believe in Facebook. Even the biggest B2B ad channels admit that on average, 15 touchpoints across multiple channels need to occur before a prospect becomes a lead (meaning that we have to be in their face, everywhere).Besides, Facebook has over 2 billion active users across its multiple ad domains (Facebook, Instagram and its huge audience network of partners). Facebook can show ads to whomever it likesBut I’m not here to argue. Sometimes a specific ad channel just doesn’t work. It may be the context or the specific type of content

Long story short

Anyways, back on our end, we will have to re-evaluate our actions and do some thinking.

Now, we know that organic traffic, direct and Linkedin ads traffic performed well. It was only a single Facebook campaign that dropped the averages.

Should we give up on Facebook ads, or insist on continuing to use it? Well, we won’t give up on Facebook. It’s way too significant to be left out of the game. But, we can’t continue as is. We first will have to run some more experiments and add some more gated content pieces to the blend because we have a hypothesis that Facebook’s context might not be optimal for this specific content. We are currently busy working on some more content pieces that will be used to test just that.

This table is taken out of the marketing automation system and demonstrates the huge difference in conversion rates between Facebook and LinkedIn:

In other words, and since we did see a reasonable conversion rate on LinkedIn, we learned that while optimizing the landing page is always a good practice, it wasn’t an emergency in this case.

Still, as we were already at it, we did take some actions (such as optimizing the landing page and the ads) that added up to an overall improvement.

This is the new design we created (we had to blur it, which misses part of the point. We are sorry for that). But don’t get too attached, as it might change a few more times. After all, testing is a never-ending process.


The results so far

Well, after all the above-mentioned actions took place, such as re-doing the look of the landing page, finetuning ad copy and creatives and so forth (and let’s not forget – we also lowered the Facebook budget until we are ready with additional content) we can already see a nice increase of about (linkedIn, for instance, climbed to over %27! It was ~20% when we started).

As I said, we are working to add more content pieces to enrich our lead-gen machine, as it’s a critical element in every B2B performance marketing operation. We’d love to share with you the results when we have them.

Conclusion: Optimization is a never-ending story

This journey taught us a couple of important lessons: 

  1. Our initial belief that we are facing a multi-channel (across the board) performance issue was prooved wrong, once we sliced and diced additional data. 
  2. Our tendency to optimize landing pages, ad creatives, and promotion copy proved to be very beneficial in this case (even though the major problem was related to a single channel, optimizing other elements still helped a lot uplifting the conversion rate). 
  3. We learned (all over again) that no matter how experienced you are, there’s always room for optimization. In other words, don’t expect to get anything optimal at first. You can hope for it, but don’t count on it.
  4. We learned when to let go and decide that for now, the optimization has reached a ‘good enough’ level. Reaching a threshold and shifting resources to other activities it’s the right thing to do, but we learned how to feel cool about it. 
  5. We learned that some questions must be left unanswered and we just have to solve some challenges with tests (without being able to provide a satisfying answer to the “why” question) – in this case, why Facebook sucks? Still no clear answer, but we have a plan for the coming next experiments that will try and solve this issue. 

Happy to get your feedback! 

Einav Laviv
Einav Laviv is a tech marketing expert, a startup founder and the CEO at G2Mteam, that supports Israeli startups with full stack global marketing services. She lives, breathes and loves startups & marketing: B2B, B2C, Mobile, online and product
Einav Laviv
Einav Laviv
hello [at] g2mteam.com