Heather Barmby
Head of CRO
LinkedIn Profile
Visualsoft is an ecommerce agency with more than twenty five years behind it, supporting over 800 active clients across UK retail. They work across the full commercial picture for the brands they partner with, from build and migration through to paid media, SEO, design, and conversion rate optimisation.
CRO sits as its own specialism inside the business, with a dedicated team, an in-house testing platform, and a methodology that runs the same way on every account. Analysis and research first. Then hypothesis and ideation. Then testing and validation. Then iterate, because the work does not stop when a test wins.
As a Shopify Plus Partner, the team spends its days inside merchant stores looking for the difference between what a site is doing and what it could be doing. That gap is rarely where anyone assumes it is, which is why the research stage carries so much weight in how they operate.
Most retailers are not short of data. They have analytics, session recordings, heatmaps, and a dashboard for each. What they are short of is explanation.
Analytics can tell you that a third of visitors leave a product page without adding to bag. It cannot tell you whether the price was wrong, the sizing information was missing, or they were never the right customer for that product. That gap matters most when a testing programme begins, because every A/B test starts with a hypothesis and someone has to write it. Without customer input, that hypothesis is a guess.
Aggregate data also treats every visitor as the same visitor. For a retailer whose customers range from complete novices to working professionals, averaging them into a single number hides the thing worth knowing.
For The Cake Decorating Company, the UK's largest retailer of cake decorating supplies, the Visualsoft CRO team built a research layer that sits on both sides of the purchase. Before the sale, an exit intent survey appears once a visitor has been lingering on a page for forty five seconds. The popup itself was custom built for the store rather than taken off the shelf, so it matched the theme exactly and stayed fast, with an embedded Grapevine Surveys survey placed inside it asking visitors what kind of baker they are. After the sale, a post-purchase survey on the order confirmation page captures how the experience went and whether anything on the site got in the way.
Every response arrives in Grapevine with the page the visitor was on when they answered it, and that is what turns a headcount into something the team can act on. It is not just that a share of the audience are hobbyists. They can see that an absolute beginner was sitting on a professional grade release spray, or that a hobbyist lingered long enough on a lace mat to be asked. That starts a map of customer type against individual products, and it raises the question worth chasing: is somebody hesitating because they cannot tell whether this product is meant for them? If so, that is a wording problem, and wording is something they can fix.
Our job is to make every visit work harder, and you cannot do that without knowing who is actually on the page. Reaching customers while they are still browsing tells us what kind of baker we are designing for, and the post-purchase questions tell us how the site performed once they got there. Together that gives us something to build a testing roadmap on rather than a backlog of assumptions.
Three questions across two surveys, and each one has a specific job to do. We worked out what we needed to know before we wrote any of them, so every answer maps back to a decision someone on the team has to make.
The skill level question tells us which type of customer is looking at which products, which is genuinely difficult to establish any other way.
For a retailer selling everything from beginner starter kits to professional grade equipment, that is the difference between writing product page copy for an audience and writing it for the person actually reading it. Where a product is bought overwhelmingly by absolute beginners, the description should explain what it is for and how to use it. Where professionals dominate, that explanation is noise and the specification matters more.
It also gives us a segment to test against. We are not just asking whether a change worked, but who it worked for.
The two questions do different jobs. The score gives us a number to track, and the open text gives us the reason behind it.
We review responses weekly, because a single comment is an anecdote and a pattern across a week is a signal. When the same friction point comes up from customers who have nothing else in common, it becomes a candidate for the roadmap.
Those reviews feed a monthly action list. Some items are content or UX fixes we implement directly. Others become hypotheses for the testing programme, which is where the CRO survey data earns its place. It is early days here, but the process is the point. Listen weekly, act monthly, and let the customer tell you what to look at next.
This is just one example of the many different ways you can capture feedback and gain valuable insights from your customers. We love being part of your journey - from seamless setup to turning insights into action!
So if you're looking to start collecting more insights and you need a bit of help, you know where we are!
See what other brands are doing with Grapevine Surveys to collect insights and improve their business
Find out how Emma Sleep Captures Customer Insights Across Europe with Grapevine Surveys
Find out how Calisi Beauty is using Grapevine Surveys to gain insights from potential customers who made a purchase and those who abandon their carts.
Find out how Nathan James uses Grapevine Surveys to capture vital customer insights to monitor marketing performance and consumer sentiment