To Optimize Conversions, Focus On Everything That Happens Before
Looking to turn more website visits into conversions? Columnist Nick Iyengar explains why it's important to first map out each process a user goes through on your site, and then use analytics to find out if it's meeting its goals.
There’s an old saying: If you watch your pennies, your nickels and dimes will take care of themselves. The same is true with your website’s performance, regardless of whether you run an ecommerce site or a lead-generation site.
Over the past few years, marketers have acquired many tools that have helped them make the connection between “inputs” and “outputs” on their websites — that is, inputs like unique visitors (and the marketing dollars it takes to attract them) in relation to outputs, such as leads and revenue. For example, Web analytics tools make it easy to measure the marketing campaigns you’re running, and whether they produce the “macro” conversions you are about.
But from an analytics standpoint, there’s still room to grow in terms of measuring what happens in the process of converting those inputs into valuable output.
I like to think of a website almost like an assembly line. To turn your visitors into valuable conversions, a lot of little things need to happen first.
When you’re measuring the performance of all the individual steps in a process, it becomes a lot easier to know why you’re seeing a change in revenue, for example. And when you know why something’s changing, it’s a lot easier to take action.
If you’re looking to make your website more efficient in turning visits into sales or conversions, it’s important to map out the “assembly line” that your website represents.
For many websites, there are processes or sub-processes that a user must go through before they can accomplish your ultimate goal for them. And often, marketers and analysts don’t have the data they need to measure the performance of those processes.
By taking a structured approach to reviewing your website, you can identify all of the different steps in the conversion process — and fill in a lot of measurement gaps.
Every page on your website exists for a reason, right? Assuming that’s true, we should be able to map out all of our pages, determine why they exist, and then use analytics to measure whether or not they’re accomplishing their goals or not.
Using the analogy of an assembly line, building out this mapping of your website will help you identify areas where you can reduce waste, deal with bottlenecks and increase “yield,” whether that’s sales, leads or something else.
Mapping Out Your “Assembly Line”
No matter what kind of website you run, start by taking an inventory of the different types of pages on your site. Think about why these pages exist: What is a successful outcome when someone visits this page? Even if it’s something that seems quite obvious, you want to be sure that those outcomes are being measured closely.
For example, you might have category pages that showcase similar products or services. These types of pages should be pushing visitors further down the funnel — so a key outcome here would be to measure whether people view a specific product or service description after viewing the category page.
Review your site in detail, determine how many different types of pages you have and what purpose they serve in the process of “manufacturing” a conversion on your site. Once you’ve done that, you should be able to group your pages by task or by process — and you’ll have a great idea of all the specific outcomes you need to be tracking.
Segmenting For Deeper Insights
After you’ve mapped out your “assembly line,” you’ll have a great way to put structure around all of the various outcomes for which you need analytics. And getting those new data points is certainly a win, but if you want to go even deeper, you’ll need to be able to do some segmentation.
For example, you might find that the dropout rate of people navigating from your home page to a category page to a service description page is X%. Of course, you’ll want to work on improving that, but in order to prioritize your actions, it would help to know which services seem to have the highest dropout rate — or the lowest yield.
Depending on how many different products you sell online, or how many services you offer, you may need to start building out some hierarchy for your page types. For instance, it’s easy to say you that you have “Product Category” pages, but you may need to drill down and get more granular.
Within “Product Category,” are there sub-groups you could create? You may want to drill down from Widgets to Widgets for Small Businesses to Shiny Blue Widgets for Small Businesses. And of course, this hierarchy can be recreated across other page types, like product or service detail pages.
By mapping out your assembly line and then putting in place a hierarchy that lets you drill down for more specific insights, you’ll not only have a high-level sense of where you can improve performance, but you’ll also get specific ideas about how to get started.
What kinds of questions can this help you answer? For example: What was the click-through rate from service description pages to downloading a case study/white paper? What percentage of people who downloaded a case study later requested a quote? How did that vary across categories and subcategories of services?
Tools For Optimizing Your Assembly Line
Ready to start filling in those measurement gaps? Here are a handful of Google Analytics tools (Adobe Analytics has similar ones, by the way) that you’ll want to start getting familiar with.
1. Event Tracking
Event tracking is great for tracking all of the content and features of your site that aren’t simple HTML pages.
Think about all the content you’re busily producing for your visitors: videos, case studies, white papers and more. You’ll want to know whether people are using and downloading this content — and more importantly, whether consuming this content has a long-term impact on conversions (more on this in a bit).
Similarly, if you’re building interactive features on your site — for example, tools for your customers to explore and research your products — you’ll want to be able to answer the same questions.
Without event tracking, you’ll have blind spots on any piece of your website that isn’t automatically tracked when a page loads. So make sure all of your great content, tools and features are tagged with events, and you’ll really enrich your dataset.
2. Custom Dimensions
Content consumption and feature usage are great, but let’s not forget: We’re not trying to get PDF downloads for their own sake; we’re hoping that PDF downloads are one step in a process that will eventually lead to conversions.
So while event tracking helps us figure out if people are consuming content, we’ll need to go further to determine whether these events actually have any long-term influence on conversion.
This is where custom dimensions come in. In addition to tagging videos, PDFs and interactive features with events, you can tag people as having interacted with all of those aspects of your site.
What’s the difference? Take a look at two sample insights:
- Insight #1: 4 percent of our users downloaded either a case study or a white paper this quarter.
- Insight #2: 4 percent of our users downloaded either a case study or a white paper this quarter, and people who downloaded a white paper have been twice as likely to request a quote as those who downloaded a case study.
See the difference? Insight #1 is certainly better than nothing. We can try different strategies to get from 4 percent to 5 percent.
But with Insight #2, we have a very clear next step. We should invest less in producing case studies invest more in producing white papers, and we should find ways to get more people to download our existing ones.
To get from Insight #1 to Insight #2, you’ll need to use event tracking and custom dimensions together.
3. Content Groupings
Web analytics tools generally track your website one page at a time, and they don’t do much “critical thinking” to group pages together based on ways in which they might be similar.
If you’ve mapped out your entire site, thinking about all the processes a person might need to go through before they convert on your goals, you’ll likely have identified several different types of pages. Examples might be Corporate/Reference pages (home page, about us, privacy policy, etc.), category pages, product or service detail pages, checkout pages and many others.
Each of these types of pages has a job to do, and as a result, you’ll want to measure these pages using different performance indicators. Let’s say you’re tracking “Add to Cart” clicks on a product detail page. Your true goal probably isn’t to optimize the Add to Cart rate on just your most popular product. Rather, you’re probably looking for ways to improve Add to Cart performance across all product detail pages, or at least all of the pages in a certain category. That’s where Content Groupings are helpful.
With Content Groupings, you can use simple, logical rules to have Google Analytics lump similar pages together. For example, any page that starts with “/catalog/shoes” will automatically be labeled as part of the “Shoes” category. But any page that starts with “/catalog/shoes/mens/tennis” can be labeled as part of a much more specific Men’s Tennis Shoes category.
Now you can take all the great event tracking data you’re collecting and use it to analyze the performance of entire groups of pages, rather than just individual pages one at a time. As a bonus, you don’t even have to use any new code to get this done.
4. Enhanced Ecommerce
This feature, of course, is geared toward businesses that are actually doing ecommerce. I’ve written about this at some length recently, so to get all the details, feel free to refer to my earlier article. But it’s worth calling out again, because the checkout process is, of course, one of the most important processes for any ecommerce website.
Prior to Enhanced Ecommerce, it was fairly straightforward to track things like revenue and measure people as they navigated through various stages of the checkout process. And while that certainly covers the basics, Enhanced Ecommerce gives us much more insight.
For example, what’s going on before someone enters the checkout process? Which products are viewed most often? Which products, after being viewed, are most (or least) likely to be added to the cart or to be purchased?
These kinds of questions are crucial. After all, optimizing your checkout process has limited potential for improvement if your highest-margin products are underperformers in terms of even being added to the cart in the first place.
And that’s a classic example of the benefit of carefully mapping out every stage of the processes on your site. We need products to be in the cart before we can sell anything — so while revenue is your “macro” conversion, simple “Add to Cart” clicks should measured just as rigorously.
Turning Your Site Into A Finely Tuned Machine
At the end of the day, there are plenty of ways to get more conversions out of your website. For instance, you could just pour a lot more marketing dollars into attracting people to your site.
But that’s the easy (albeit expensive) way to do things. Instead, you can focus on every little process that stands between your visitors and the conversions you care about.
Methodically map out each process and what the success metric for each would be. In many cases, you’ll need tools like events, custom dimensions and content groupings to start collecting and analyzing those metrics.
But if you can get each station of your assembly line humming more efficiently, seeing an improvement in conversions is all but assured.
Contributing authors are invited to create content for MarTech and are chosen for their expertise and contribution to the martech community. Our contributors work under the oversight of the editorial staff and contributions are checked for quality and relevance to our readers. The opinions they express are their own.
Related stories