--- title: "Customer Journey Research: Which Changes to Prioritize | antedote" description: "A journey map shows plenty to improve. Which change deserves investment depends on what it alters for the customers who matter and the evidence behind it." url: "https://antedote.com/thinking/where-to-invest-in-the-customer-journey/" published: "2026-10-28" updated: "2026-10-28" author: "Riki Neill" --- # Which part of the customer journey is worth changing? A journey map shows plenty to improve. Choosing where to invest depends on which obstacles change what customers do, for which customers, and whether the business can change them. Article on Research strategy, 4 min read Reviewed by [Riki Neill, Partner](https://antedote.com/why-antedote/#partners) on October 28, 2026. Published October 28, 2026. The map is complete. It shows delays, frustrations and moments customers value. Your team can see plenty to improve. The harder question is where an investment would make a worthwhile difference. The part of the customer journey worth changing is where an obstacle alters what your most important customers go on to do, and where the business can make a change that removes it. The most visible frustration and the longest stage are good places to look. Neither is automatically the answer. ## Three candidates for one budget Consider a hypothetical company that sells electric bikes online and through independent dealers. Its journey map shows three strong candidates. Shoppers find the specifications hard to compare. Booking a test ride is clumsy. After purchase, new owners aren’t sure whom to contact when something needs adjusting. Each problem is real, and each has a champion in the business. The website team can point to the volume of complaints about the specification pages. The retail team can show that deliberation, the weeks between first interest and purchase, is the longest stage. The service team can describe the frustration of new owners in detail. ## A complaint, a gate and a change you can make Those arguments describe different things. A frequent complaint tells the team what people notice and mention. Some will grumble about the specification pages and buy anyway. A gate is a condition for moving on. If people who never get a test ride rarely buy, the booking problem may matter more than its share of complaints suggests. The third thing to know is whether the business can change the obstacle. The test ride may be the gate, yet independent dealers run it. Improving that moment could mean persuading partners, a different investment from redesigning a web page. A worthwhile change has a large consequence for customers and a real way for the business to act. Length is a weaker guide than it looks. In [research on patients’ memories of painful procedures](https://pubmed.ncbi.nlm.nih.gov/8857625/), what people remembered largely reflected the worst moment and the final minutes: “lengthy procedures were not remembered as particularly aversive.” Medicine is far from bike buying. Still, it’s a reason to doubt that the longest stage shapes how customers judge the whole. ## The answer can differ by customer A gate for one group can be a formality for another. A first-time buyer may need the test ride to feel confident. Someone replacing a bike they already know may skip it entirely. A commuter replacing a stolen bike in a hurry is in a different buying situation from someone who has been saving for months. That is why one map can mislead when a business serves several kinds of buyer. In [journey research for a kitchen and bath brand](https://antedote.com/explore-our-work/mapping-a-kitchen-and-bath-buying-journey/), the brand already had a segmentation of its buyers, which raised a harder question: would each segment need a journey of its own? A national survey of recent fixture buyers, across every segment, was designed to measure each stage, so the team could see whether one journey served them all. ## What each kind of evidence can show Observation shows what people actually do: the workaround, the pause to consult a partner, the step nobody mentions because it feels ordinary. Customer accounts explain why a moment mattered to them. Quantitative patterns show how common an obstacle is and how strongly each stage relates to what people go on to do. Each has limits. Observation covers few people, and accounts are reconstructed after the event. A relationship between a stage and likelihood to buy is a pattern, and it is no proof that improving the stage would change sales. In the kitchen and bath work, the qualitative and quantitative research were designed together, so one built what the other measured. Digital and in-home ethnography produced a framework of the journey’s stages and the gates between them. The survey measured each stage, and a model of how each gate relates to likelihood to buy gave every candidate starting point the same measure. ## Compare the ideas as well as the stages A map tells the team where to look. The investment usually concerns something more specific: a comparison tool, a booking system, a check-in with new owners after their first month. Once those ideas exist, they can be judged against the same understanding of the journey, by the customers each one is meant to help. After the kitchen and bath journey research, the brand’s redesign program asked us to rate eight of its own improvement ideas against the journey. A rating of that kind is no forecast of return. It can show which ideas customers see as answering the obstacles that matter to them, and it can expose a favorite that answers a complaint without touching a gate. The decision that follows might be to invest in one idea, to rework another so it reaches the gate it was meant for, or to try a change with one customer group before committing more widely. Whichever it is, the team is choosing between specific changes for specific customers, knowing what the evidence supports. If you’re weighing improvements like these, that comparison is where [strategic research](https://antedote.com/what-we-do/strategic-research/) can help. ## Read next - *Article* · 3 min read ### [Who isn’t in the research?](https://antedote.com/thinking/customers-your-research-brief-leaves-out/) Research describes the people it includes. - *Article* · 4 min read ### [AI can make an idea look finished before the thinking is.](https://antedote.com/thinking/ai-concept-development-polish/) AI can make a concept tangible sooner. The risk is a polished image that settles product and commercial choices the team hasn’t yet made. ## How we help [Strategic research](https://antedote.com/what-we-do/strategic-research/) Qual, quant and data science, designed together, and the strategic thinking to turn the evidence into your next move. ## Working through this question now? Bring us the version your team is facing; half-formed is fine. We’ll say how we would approach it, and whether we’re the right fit. [Talk to us](https://antedote.com/talk-to-us/) · [hello@antedote.com](mailto:hello@antedote.com) A partner or director will reply.