It’s 2018 – every half-baked Kickstarter and Instagram-famous dog has a website. It’s table-stakes for B2B, but by no means is every website built equally. As your audience’s go-to destination for information, you’re either converting with confidence or sending them on a wild goose chase.
Hosted by PathFactory’s Director of Content Marketing, Cassandra Jowett, this #B2BChannel Panel gathered some of the brightest B2B minds for an hour-long discussion on all things website optimization:
- Takeshi Young – Website and SEO Manager at Optimizely
- Tyler Lessard – VP of Marketing at Vidyard
- Shannon Vaughan – Digital Marketing at Rockwell Automation
Here’s a round-up of the most useful takeaways from that panel:
How are you currently using your website to generate leads?
Takeshi: Conversion rate is one of the key metrics we look at. We also look at down-funnel conversion rates too like quality of leads and how they perform further down the funnel.
Shannon: At Rockwell Automation, we sell a lot of different things included software and hardware. Selling hardware is easier but for the software solutions, it’s more complicated. So I like to really tell a story on our site and lead people down a path to reveal what’s best for their company and reveal the bigger picture.
Tyler: At Vidyard, we’re always trying to drive toward a final high-value call-to-action: A demo request, pricing request, a pre-recorded demo view, or chat conversation. However, these actions are never the first thing people are doing on a site. A big part of our strategy is having diversity in potential actions visitors can take to engage with us at different stages in their journey. Having lightweight CTAs keeps people engaged and allows us to surface the right at every stage of their lifecycle.
Cassandra: The conversation seems to have changed in recent years. Marketers are focusing more on the customer experience and how people are engaging with the content on their website.
It’s not as much about driving as many people as possible to the ‘request a demo’ form as it used to be. You have to make sure there are other types of conversion points on your website.
How is your website currently designed to help visitors spend more time with your content and provide a good customer experience?
Takeshi: We’ve evolved to serve more enterprise companies than we did before so working to optimize the website for these customers. Creating personalized experiences for our top prospects and a different experience for others who aren’t as good a fit.
Shannon: We’ve also evolved at Rockwell Automation. We used to gate EVERYTHING. So we would get a lot of leads but they weren’t all great leads. Now we’ve scaled back.
We encourage leads to digest more content, educate themselves, and find what they want. Only then do we present a form and ask them for information.
I know what you’re thinking…less gating sounds all well and good; however, you need a way to get unknown website visitors to identify themselves
In your experience, what are the best ways to convert unknown website visitors into known ones?
Tyler: We are creating a more user-centric experience on our site. Before, we asked for value (information) from the visitor first and then unlocked all the content. Today, we offer leads value first and then ask for their information. For this reason, our content has to be better. The ‘action’ or next step also doesn’t have to be a demo request form with 8 fields. It could be ‘subscribe to a weekly update’, for example, which is a much more lightweight ask.
Marketers can’t rely on click-bait anymore because the content has to be good enough to drive the desired action afterwards.
Shannon: It’s tricky because some people just don’t want to identify themselves. Everyone’s got ‘that yahoo email’ they use as a dummy email. Someone who is really serious about buying something and learning about your company will give you their information. We’ve also changed our forms to ask more thoughtful questions which has been working well.
Takeshi: We used to gate all of our case studies and customer stories and we made the decision to ungate them. Since doing that, we’ve seen a lot more engagement and organic traffic–that’s been a big win for us. In regards to emails, this year we actually restricted the ability to download content without a professional email, so you can’t use ‘that yahoo email’ anymore. That’s lead to a decrease in number of leads but increase in quality.
Kristen O followed up with Shannon’s response via the live chat. She asked: What are examples of questions you ask and what do you do with the info when you get it?
Shannon: We created an analytics workbook so some of the questions we asked were specific to plant operations. For example, how many plants do you have that use a certain type of machinery? We asked 5 or 6 questions and the last question was ‘do you want sales to contact you?’. If they answered ‘yes’ to that question that lead would go directly to sales. Do we get a ton of those? No. But the ones we do get are such high quality and we get really valuable answers.
Resource centres are the subject of a lot of debate right now – some see them as an effective way of delivering content, others merely a dumping ground. Where do you see a resource centre’s role today?
Takeshi: This is really top of mind for us right now as we’re undergoing a website redesign. Now we are trying to incorporate more content throughout the buyer journey. So it’s not just living in the resource centre, it’s offered on the home page, the product page etc.
We’ve been trying to personalize the content we show based on the industry of the visitor. This has actually led to a 20% increase in engagement with content.
Shannon: We do have a resource centre. Ours is segmented in terms of content type. However, we don’t really find those to be where our buyers tend to go to find information. We prefer to take them through a journey including a variety of content assets in all shapes and sizes-from blog posts to videos to eBook-that help them educate themselves.
Tyler: I totally agree. Having a centralized resource centre has intrinsic value. For example, it’s a good place for sales to find content. But you can’t expect that’s where people will go to binge on your content. People aren’t often looking for an ‘ebook’ they are looking to an answer to a specific question. At Vidyard, we let them filter content in our resource centre by persona, company size, question, type etc. Having said that, we feel it’s more about how you use those resources throughout your website and campaigns than just having them live in a content hub.
Tyler touched on the personalization piece but viewer Will V wanted to dive a bit deeper: How do people optimize and personalize their website to accommodate persona and use case?
Shannon: We’re going through a website refresh right now. We are working towards getting to a place where visitors can personalize their journey rather than having to go through a prescribed journey we dictate to them
Takeshi: We personalize on a number of different vectors including industry and job function. We tag all our content so we can serve it based on the persona. We also tag things as ‘general’ so some assets will be served regardless of persona.
There are a lot of different ways to measure a website’s performance – heat mapping, google analytics etc. – what indicators of website effectiveness have translated into the most meaningful improvements for your team?
Shannon: Lead gen is our main goal for the website but we do look at other things like where people are coming from, where they are going, what they are clicking on and where they are dropping off. This way we can tailor our content to our prospects’ actions.
Takeshi: I’m part of the demand gen team so we are super focused on conversion rates of top landing pages. I also look at down-funnel metrics, like how many of those leads become marketing qualified and sales qualified.
Tyler: There’s traffic metrics, bounce rates, and other key metrics we look at as well as specific conversion rates on key pages. We also look at heat mapping to see what’s happening in different places and see if there are better ways we can be drawing people to the high-value CTAs.
Traffic at all costs is not the best idea, especially in B2B. Always look at your downstream metrics to see how leads are resulting in revenue and pipeline.
Cassandra: We’re in an interesting situation here at PathFactory. We have a brand new domain and have to build our authority up from scratch. So traffic is great but we need to make sure people can actually find us and that people who are arriving on our site find what they need. We monitor things like bounce rates, Alexa ranking, and domain authority as we rebuild the website post-rebrand. So it’s a slightly different set of metrics at this stage but it’s solving a different set of challenges.
Of course, no conversation around company website would be complete without touching on experimentation and A/B testing:
What kind of A/B testing are you running on your website and what tool are you using for that?
Shannon: We do A/B testing but we don’t really use a tool. We will have 2 separate pages where we change one thing and see what performs better. Or we will run one page for a certain amount of time and then adjust something and run it again.
Tyler: We’ve been doing split testing for quite a while and we use Optimizely for that. We have optimization and growth experts that test different things on each page–everything from small things like changing the colour of the CTA button to bigger things like testing layout of the page and moving different blocks around. The ability to automate testing at scale is really powerful. I’m still blown away by how some small changes to the website can have a huge impact on conversion.
Takeshi: Ya that’s something we definitely do here at Optimizely… and we use Optimizely [laughs]. The most impactful experiment this year a test we did that led to a form update. Previously, our forms had 7 different fields. We’ve now split it up into a 2-parts. On the first page we ask for an email and on the second we ask for the rest of the information. We’re using a technology called Clearbit to automatically enrich the data on the second page of the form based on the email address. That’s been a huge win for us this year in terms of A/B testing.
What are some of the most recent website tests you’ve run that yielded interesting or unusual results and how did that inform your strategy?
Takeshi: As I mentioned, we are about to launch a new website (wish us luck), so we’ve been testing out a lot of design elements to a subset of visitors to validate various hypothesis. For example, our design team had the hypothesis that people would respond better to photos of people rather than company logos when presenting case studies. We tested it and actually found that the logos performed better. Sw we can take that into account when we launch our redesign.
Tyler: When doing these tests, a lot of people forget to document them to refer back to later. Having a documented repository of best practices and dos and don’ts from previous experiments goes a really long way. We just use Google Sheets, but Takeshi, how do you guys handle testing documentation?
Takeshi: Optimizely actually launched a tool called Program Management earlier this year to help document experiments and ideas. You can list the ideas out, rank them in terms of impact and effort and then record the results when they’re done.
Cassandra: One of the things we’ve been doing is testing different content titles. Obviously, all content has a name and it’s tempting to keep the name it’s been given after it’s written but that’s not always the best way to get someone to click on it. When we launch new content tracks we test different formats, lengths and styles of writing to see what gets more engagement.
There are a million ways you can be A/B testing all the time, it can be overwhelming. How do you decide which changes you’ll make in each test? Is there a method to the A/B testing madness?
Takeshi: We always look at tying all of our A/B tests back to our company objectives. At the beginning of the year, it was improving lead quality so all of our tests would ladder up to that.
Always make sure the A/B experiments you’re running are tying back to the objectives that matter to your team.
Shannon: We ran a test where we were trying to rank higher in Google searches. So all of our tests were measured based off how much they affected our Google rankings.
Tyler: It’s a bit of a patience game. You have to have a plan–maybe you have a monthly or quarterly theme. You have to be methodical in your approach to A/B testing.
Cassandra: It’s definitely a long game. You’re tempted to test everything right away but you need to build a strategy around it and make sure each test actually means something and will add value to your team.
Lead quality is a source of contention between sales and marketing at most B2B organizations. How can you design the website to generate a better class of lead to send to sales?
Tyler: Marketers need to make sure both sides have the data to validate if leads are effective. In the early days, if someone did anything on the website they were flipped to the sales team. Inevitably this causes a sales efficiency problem because the majority of leads are rejected. At Vidyard, we now have a more mature lead scoring model that takes into account activity scores, persona information and company demographics. In order for a lead to be passed to sales they not only have to engage with our content, they also have to be the right fit from an individual and company perspective.
Cassandra: Definitely. Adding the content engagement piece, demographic, company size etc. can be so critical to understanding the value of leads.
Live chat. So hot right now. Olivia from TouchBistro wanted to know: Is live chat something that increases conversion or is it just a nuisance for visitors?
Tyler: We use it and it works really well for us. Having said that, when you add net new CTAs you have to monitor potential cannibalization of other channels. Our top 4 CTAs from a performance and revenue perspective on our website are Demo Request (live demo), Pricing, Watch Demo (gated, pre-recorded), Chat. Based on what we’re seeing, these Tier 1 CTAs are most successful.
Takeshi: We’ve jumped on the chat bandwagon as well and are piloting Drift. It’s been working out great. We aren’t using the live chat, we are just using the chatbot right now. We currently have it set up to only appear for enterprise accounts and our target accounts. The chatbot allows visitors to set up calendar meetings with our SDRs in one step. This way, we speed up the sales process.
Cassandra: It’s all about how you use it. If you use it right you’re giving people the experience they want and reducing friction and helping visitors get what they need faster.
Watch the whole discussion here:
The Channel Panel is over but you can still watch all 6 recordings here!