As I dive deeper into this mini-degree, it’s quite impressive the breadth of valuable content that is inside of it. I continue to pick up valuable nuggets of applicable information that I can immediately apply to businesses that I’m working with.
It’s been made very clear to test based on data from customers, and not from opinions of “experts” or “hippos” (highest paid person’s opinion) inside the company.
I was reviewing the ResearchXL Research process, which I found incredibly valuable. It’s a six-step framework for gathering incredibly valuable insight from your customers and also developing great hypotheses.
Step 1: Technical Analysis
- This includes Cross browser testing, cross-device testing, speed analysis tests
- Is anything on the app or site broken?
- Where on the site or app is it broken?
- Which browsers?
- Which devices?
- Which pages are slow?
Step 2: Heuristic Analysis
- Identify key areas of interest, check key pages for relevancy, motivation, and friction.
- Relevancy: If I’m clicking on a link about a certain kind of boot, is the page relevant to that?
- Clarity: Do I clearly understand specifically and clearly what this website is about, who this is for, why should I buy here and not from some other place?
- Motivation: What is being done on the page to increase my motivation to take action and what things make it difficult to take action?
- Photos should improve the clarity of the value proposition and not just be two people in suits shaking hands or a picture of the front of a skyscraper
- There was a great survey idea thrown out during this section: “How many websites did you look at before making a purchase from this website?” I will definitely be using this survey in the near future.
- Every single page should have one desired action on that page, and this includes the home page. The purpose of that could be to get them further down to funnel on another page. If it’s ecommerce and it’s a category page, the desired action is to click on a product. If it’s a product page, the purpose is to add to cart. If it’s an add to cart page, the purpose is to proceed to checkout, etc.
- They briefly reviewed the Fogg Behavioral model, which seems like a great mental model for reviewing websites. That model is a desired behavior that happens when three things converge (motivation, ability, and trigger).
- A question to ask yourself given this model is: “how much friction can I add into the sales process considering their level of motivation?” I think a great example of this is job applications. There is sooo much friction in online job application processes. If it’s a well-qualified and busy candidate, they aren’t even going to bother about uploading their resume, cover letter, and then filling out their entire work history which is already included on their resume step by step, giving references, etc. It’s clunky and I feel that companies lose out on the best candidates this way.
Step 3: Digital Analytics
- Analytics health check, identify leaks, setup measurements for KPI’s
- Where are the leaks on the site/app? (Make sure you’re referencing the customer journey for this step. If you don’t have one, then get one created as soon as possible)
- Which segments?
- What are users doing?
- Which actions correlate with higher conversion?
- MEASURE EVERYTHING. There’s a saying my wife has that I apply to certain situations that definitely applies here: “It’s better to have it and not need it then to need it and not have it.” You’ll likely need it at some point down the road, so put in the work now.
- Ask yourself as you review this: “How many different ways are there to achieve this goal?”
Step 4: Mouse Tracking & Form Analytics
- Heat maps, click maps, scroll maps, user session video replays
- Where do they click?
- How far down do they scroll?
- Differences between devices?
- Session Replays?
- One of the easiest ways to get someone to continue to scroll is to keep the background color the same (amazon does it, which doesn’t mean to copy everything Amazon does, but they do it for a reason because color changes can signal the end of the content area subconsciously for the potential client/lead).
- Make sure you are using analytics to see where users are dropping off.
- Put form field that are easy to fill out first, so it creates momentum for the customer.
- Also use form field analytics to see if users are dropping off at a certain point on the form.
Step 5: Qualitative Surveys
- Customer surveys, chat logs, interviews, web traffic surveys, and cart abandoners
- Buyer groups — people who just purchases, and then people who left and didn’t buy
- Which problem are they solving?
- How are they deciding?
- What’s holding them back?
- What else do they need to know?
- Quantitative data tells you what is happening, where and how much. It doesn’t tell you why. Qualitative data answers the why.
Step 5: Qualitative Surveys
- It is recommended to use open-ended survey questions for customers surveys (people that bought the product) because you may have blind spots that you may not be aware of and asking closed-ended questions will only get you answers within your own awareness of problems.
- “What is one thing that kept you from buying?”
- “What kind of questions do you have about the product?”
- “Tell me something about yourself”
- Ideally get 200–300 responses for your open-ended survey and then categorize them. More than 300 is pointless and just creates more work
- “What was the biggest doubt you had about buying from us?”
- Another question to ask is on a page in the funnel that is struggling with the conversion action: “What is holding you back from completing this purchase right now?”
Step 6: User Testing
- Identify usability, clarity issues, and sources of friction
- What’s difficult to understand?
- What’s difficult to do?
- What goes wrong?
- In user testing, you want to give them three tasks:
- 1. Find a specific product or piece of information
- 2. Find something broad. Find a pair of jeans under $50
- 3. Buy a product, sign up, etc.
I find this framework extremely valuable and actionable and am excited to implement it. I’ve implemented pieces of it, but I love it as a whole. The number one step to do in conversion optimization is conduct research because if you don’t do that correctly, everything else is a waste of time.