Notes from Andy Budd’s UserTesting Webinar

Deconstructing project failure

• The difficult part is the doing
• Project kick off is already too late, seeds have already been sown for failure
• Starting is crucial
• Define the problems early

Problem: Selling the project

• e.g. time since re-design
• “this time it’s going to be different”
• Overselling is a risk – stakeholders expectations are high

Problem: Estimation bias - humans are terrible at estimating the potential issues (failure)

• resources
• design blockers

Solution: Become product rather than project centric

• learning cycle is too short with large/long projects
• start moving budgets into smaller changes - opex rather than capex
• shorter iterations, continuous cycles
• manage expectations more easily

---

Problem: We estimate at the point in the project when we know the least

• ballpark figure becomes the budget
• ballpark referred to later on

Problem: We didn’t have the right people in the room

• understanding organisations
• different value places into the involvement in the process
• involve the right people - the product team; developers, designers etc
• planning poker exercise for user stories
• timing is key

Solution: A good scoping phase

• explain the discovery stage
• talk to customers, stakeholders
• use personas, user journey maps, user stories
• trial and error
• sell back to the organisation
• manage expectations

---

Problem: Technology is surprisingly opinionated

• generic technology is opinionated
• opinions are baked into technology
• this has a knock-on effect

Problem: Selecting your technology platform too early can paint you into a corner

• added too early from a list of requirements
• CTOs love shopping and tech
• suppliers are great at selling tech
• miss-sold tech is rampant - once invested the flaws start to show
• hard to turn back once invested

Solution: Design from the outside in

0. design user journeys
1. prototype
2. then approach tech

---

Problem: Agile isn’t a cure-all

• you have company buy-in but …
• challenging in it’s own right
• successfully modern waterfall approach can work
• huge commitment going full-agile
• agile is an all-in company approach
• certainty to stakeholders on delivering
• difficult for ‘command and control’ teams
• referred to as “fragile”
• agile is cross-team and consensus agreement

Solution: Try being agile (lowercase a)

• start with critical project
• start slow and small
• develop stakeholder understanding
• building culture
• refer to the agile manifesto

---

Problem: Scope creep

• agile allows for scope creep
• new priorities, market change, timeline change
• bloated if not controlled

Solution: A kick-ass product manager with real authority

• being there 100%
• real ownership of the product and decisions
• continuous feedback to the company
• needs to be in-house
• hard to find
• sits between UX design, tech, business
• understands goals, budgets, user experience
• mediator between teams
• unique and valuable
• clear governance structure/program in place
• senior stakeholders own the outcomes but not the output - decides on the what not the how

---

Problem: Organisational silos

• improve efficiency
• breeds knowledge gaps
• design » dev handover void and process misunderstanding
• demoralising for any group within the organisation

Solution: Product teams and cross functional pairing

• work as product teams
• involve everyone
• user stories - a snapshot of a conversation and decision as a group
• shared understanding

Question: How should cross functional pairing change over a project duration or multiple sprints?

• Use your best judgement
• It depends on the project, stage and team
• Irrelevant if the teams are already committed to the culture

---

Problem: A talented but inexperienced team

• soft skills matter
• broad experience make great designers

Solution: Get external help and perspective

• bring in freelancers, agency, coaches
• share the risk
• bring in and work with the internal team
• mentorship
• extract the talent

---

Problem: You select the agency on pitch

• pitfalls when working with them
• big pitches = big promises
• underquote to please - sometimes intentional, sometimes not

Solution: Try working with them first

• they challenge you
• paid pitches
• solve a problem together
• bring teams together to discover relationships
• try a design sprint and integrate teams

---

Design is a culture problem

• change the way you do business
• hard for one person to change the whole culture
• find allies

Thanks for a very insightful talk Andy. Thanks to UserTesting for hosting the free webinar.