Library · Reference
Plain-language digital product terms with practical examples. Examples are fictional.
61 of 61 terms
Metrics & experimentation
A comparison that randomly assigns eligible users or other units to different experiences and measures a defined result. With a sound design and analysis, it can estimate whether the change caused a difference.
Example: A fictional notes app randomly assigns new accounts to two welcome screens and compares first-note completion using a pre-agreed analysis plan.
Prioritisation & delivery
Specific conditions a piece of work must meet to be accepted. They make expected behaviour testable and help people spot gaps before delivery.
Example: For a saved article, the title must appear in the list and reopening it must show the correct article.
UX & accessibility
Designing and building products so people with disabilities can perceive, understand and operate them. It requires attention to code and interaction as well as visual appearance.
Example: A fictional reading app labels its controls for screen readers and lets people reach every action using a keyboard.
Metrics & experimentation
The point where a new user reaches a defined early experience of the product's value. Teams must choose and test the behaviour and time window that make sense for their product.
Example: A fictional recipe app counts activation when someone saves a recipe and builds a shopping list in their first week.
AI products
A structured assessment of how well an AI feature performs on representative tasks and failure cases. It should cover the whole user experience, using defined criteria and appropriate human checks as well as automated scores.
Example: A fictional help team tests whether answers match its articles, cite valid sources and admit when the answer is missing.
Technology & platforms
An agreed way for one piece of software to use another piece's capabilities or data. It describes available operations and how to request them, rather than being a complete developer toolkit.
Example: A fictional reading app requests book details from a catalogue through its API.
Technology & platforms
Checking that a person or system is the identity it claims to be. Signing in proves identity to the service, but does not by itself grant every permission.
Example: A fictional workspace app verifies a member's sign-in before opening their account.
Technology & platforms
Deciding what a person or system is allowed to access or do. These permissions are separate from the process that checks identity.
Example: A fictional workspace lets a signed-in viewer read a project but only its editor can change the title.
Technology & platforms
The server-side software that handles shared data and application rules behind an interface. It commonly checks access, stores information and answers requests from user-facing applications.
Example: A fictional book-club app's back end stores group membership and returns only lists that the requesting member may view.
Metrics & experimentation
The loss of users or customers during a defined period under an explicit rule for leaving. Inactivity-based user churn and cancelled-subscription churn measure different things.
Example: A fictional paid app starts a month with 200 subscribers and 10 cancel during it, giving 5% subscriber churn using the starting subscriber base.
Metrics & experimentation
A group of users who share a defined characteristic or starting event. Comparing groups over time can reveal patterns hidden by a whole-product average.
Example: A fictional lesson app compares week-four return rates for people who joined in June and July.
AI products
The amount of information a model can work with in a single interaction, usually measured in tokens (small chunks of text or other input). Depending on the model, the limit must accommodate input and generated output rather than unlimited conversation history.
Example: A fictional research helper selects relevant passages instead of sending an entire large archive in one request.
Metrics & experimentation
The proportion of an eligible group that completes a defined action. Both the starting group and action must be specified so comparisons are meaningful.
Example: If 40 of 500 fictional event-page visitors register during a week, that visitor-to-registration conversion rate is 8%.
UX & accessibility
A shared set of design principles, reusable interface parts and usage guidance. It helps teams create consistent experiences while still checking what works for each situation.
Example: A fictional platform uses the same tested form fields, error messages and spacing rules across its tools.
AI products
A numerical representation of content that captures useful learned relationships. Comparing these representations can help find related items, but similarity is not proof that their meanings are identical.
Example: A fictional article app finds pieces related to a reader's question even when the wording differs.
Technology & platforms
Recording defined actions or system occurrences so a team can understand product behaviour. Useful tracking needs consistent names and properties, with care over what data is collected.
Example: A fictional lesson app records lesson_started and lesson_completed events without storing the learner's written answers.
Prioritisation & delivery
A control that switches a feature on or off without requiring a separate code release for each change. Teams can use it to limit exposure, compare experiences or disable a problematic feature.
Example: A fictional notes app enables its new editor for a small group, then turns it off after finding a saving bug.
AI products
Additional training that updates an existing model using selected data for a task or behaviour. It changes the model, unlike adding instructions or retrieving reference material for a single request.
Example: A fictional catalogue team trains an existing model on reviewed examples so it follows its category-labelling format more consistently.
Technology & platforms
The part of an application that presents information and handles interactions on the user's device. It often exchanges data with a server, but some features can work entirely on the device.
Example: A fictional reading app's front end displays saved titles and reacts when someone selects one.
Metrics & experimentation
A sequence of steps people take towards a goal, measured to see where they stop. Falling numbers show where to investigate, but do not explain the cause on their own.
Example: A fictional reading app compares visitors, account creations and first saved books to find its largest drop-off.
Metrics & experimentation
A measure watched to catch unacceptable side effects while improving a main goal. Teams decide in advance how much deterioration would stop or change a release.
Example: A fictional search team tracks errors and response time while testing a layout intended to increase useful clicks.
AI products
An AI output that presents invented or unsupported information as though it were reliable. A confident tone does not establish that the answer is true.
Example: A fictional book assistant invents a sequel title, so the team checks suggestions against its catalogue before displaying them.
AI products
A workflow that includes a person's review or decision at a defined point in an AI-assisted process. The person's responsibility and ability to correct or stop the result need to be explicit.
Example: A fictional community newsletter tool drafts summaries, but an editor checks and approves every summary before publication.
Strategy & discovery
A specific belief that can be checked with evidence. It states what you expect to happen and under which conditions.
Example: A fictional notes team predicts that a sample notebook will help more new users save their first note within a day.
AI products
Using a trained model to produce an output from new input. This is different from training, which changes what the model has learned.
Example: A fictional recipe app sends a dish description to a model and receives suggested tags without retraining it.
Strategy & discovery
A way to understand the progress someone wants to make in a particular situation. It focuses on their goal and circumstances rather than a requested feature alone.
Example: A commuter wants to use a delayed train journey to practise a language without needing sound.
Metrics & experimentation
A measure chosen to track progress towards an important goal. It needs a clear definition, time period and owner to support useful decisions.
Example: A fictional course team tracks the percentage of new learners who finish a first lesson within seven days.
AI products
A machine-learning model trained on large amounts of data to process and generate language. Its fluent responses can still contain errors and need checking for the intended use.
Example: A fictional note-taking app uses an LLM to suggest a title, then lets the author edit it.
Technology & platforms
The delay between an action or request and the response at a specified measurement point. Teams should state what is timed and examine slower experiences as well as averages.
Example: A fictional search team measures the time from tapping Search until results appear, including slow mobile connections.
Prioritisation & delivery
A deliberately limited version of a product used to learn whether a core idea delivers value in real use. It should serve its intended early users well enough to test that idea, rather than simply being unfinished.
Example: A fictional book-club service starts with group creation and one shared reading list, leaving recommendations for later.
Technology & platforms
An application built for a particular operating system using that platform's tools and capabilities. Its installation and device access differ from a website opened in a browser.
Example: A fictional reading app builds a platform-specific mobile experience with downloaded books and operating-system sharing controls.
Metrics & experimentation
A central measure intended to reflect the value users receive from a product. It helps align a team, but needs supporting measures to reveal trade-offs and harm.
Example: A fictional collaboration app tracks weekly teams that complete a shared project milestone, alongside reliability and support measures.
Strategy & discovery
A goal-setting method that pairs a clear, ambitious objective with measurable key results showing whether it was achieved. Key results describe changes in outcomes, rather than a list of tasks to complete.
Example: A fictional learning app sets the objective 'Help new learners build a lasting habit', with key results to raise week-four retention from 25% to 35% and increase weekly practice days from two to three.
UX & accessibility
The experience that helps new users understand a product and reach useful first results. It can include guidance, setup and examples, and should not be reduced to a welcome tour.
Example: A fictional notebook app offers a sample note and a clear first writing task instead of five explanation screens.
Strategy & discovery
A change in user behaviour or a result the product helps achieve. It describes the effect of the work rather than the thing the team ships.
Example: More new members finish their first lesson, after the team simplifies the starting experience.
Strategy & discovery
Something a team produces, such as a feature, design or document. Finishing it does not automatically mean that users are better off.
Example: A fictional app team releases a new lesson picker; whether more people learn is a separate question.
Prioritisation & delivery
Deciding which work deserves attention first when time and resources are limited. Good choices weigh expected benefit, effort, risk and the team's goals.
Example: A fictional event app fixes failed registrations before adding animated profile pictures.
Prioritisation & delivery
An ordered collection of possible product work, including improvements, fixes and research. It is more detailed than a roadmap and is not a promise to deliver every item.
Example: A fictional map app keeps a broken search fix above a proposed saved-route feature.
Strategy & discovery
The work of learning which user problems are worth solving and which solutions might work. It reduces uncertainty before and during development through research and testing.
Example: A fictional recipe team interviews cooks and tests sketches before building a pantry feature.
Prioritisation & delivery
A shared view of the product's intended direction and major priorities over time. It explains where the team is heading and can change as evidence develops.
Example: A fictional library app plans to improve finding books now, borrowing next and reading groups later.
Strategy & discovery
A set of choices about whom a product serves, which problems it addresses and how it will succeed. It guides what the team will deliberately leave out.
Example: A fictional reading app focuses on busy commuters and short listening sessions, rather than adding a social network.
Strategy & discovery
A description of the better future a product aims to create for its users. It gives long-term direction without prescribing every feature.
Example: A fictional learning app aims to make everyday language practice easy to fit into a short break.
Strategy & discovery
A situation where a product meets a strong need for a particular market and users keep choosing it. Teams look for several signals rather than treating one survey score as proof.
Example: A fictional design tool sees teams repeatedly return and recommend it because it solves a recurring collaboration problem.
Technology & platforms
A web application that can offer app-like capabilities such as installation and offline use where supported. Those capabilities depend on its implementation and the user's browser and device.
Example: A fictional recipe PWA lets supported devices install it and keeps selected recipes available without a connection.
AI products
The input and instructions given to a model for a particular task. Changing a prompt can change its response without changing the model's trained parameters.
Example: A fictional writing app asks for three short title suggestions and includes the draft as context.
UX & accessibility
An early representation of an idea used to explore or test how it might work. It may look interactive without being a functioning product ready for real use.
Example: A fictional event team links a few screen mockups to test whether organisers understand the invitation flow.
Technology & platforms
An approach that adapts a website's layout and presentation to the available screen space and device conditions. It helps the same content remain usable across different display sizes.
Example: A fictional glossary shows three columns on a wide screen and a single readable column on a phone.
Metrics & experimentation
The extent to which a defined group continues to use or return to a product over time. A retention measure must specify who is included, what counts as use and the return period.
Example: Of 100 fictional learners who started in May, 35 complete a lesson in June, giving 35% next-month retention under that definition.
AI products
An approach that retrieves relevant material and supplies it to a model when generating an answer. It can ground responses in selected sources, but does not guarantee correctness or retrain the model by itself.
Example: A fictional help assistant finds current setup articles before drafting an answer and links the retrieved passages.
Prioritisation & delivery
A scoring method that compares reach, impact and confidence against estimated effort. Its usual calculation is reach times impact times confidence, divided by effort, and the result supports judgement rather than replacing it.
Example: A fictional learning team scores two onboarding ideas using the same monthly audience and effort units, then checks whether its assumptions are credible.
Technology & platforms
A collection of tools, libraries and guidance for building software for a particular service or platform. It may make an API easier to use but is not the API itself.
Example: A fictional mobile team uses a map provider's SDK to display maps without building the drawing tools from scratch.
Prioritisation & delivery
A fixed-length period in Scrum during which a team works towards an agreed goal. It includes planning, building, review and reflection, rather than just a deadline for coding.
Example: A fictional two-week sprint aims to let readers create and rename their reading lists.
Prioritisation & delivery
Extra future work created when technical choices make a system harder to change or maintain. Some debt is a deliberate trade-off, but leaving it unmanaged can slow later improvements.
Example: A fictional app duplicates search logic to launch quickly, then plans to consolidate it before adding filters.
UX & accessibility
How effectively, efficiently and satisfactorily particular people can achieve their goals with a product in a given context. Testing real tasks can reveal obstacles that opinions about a design may miss.
Example: In a fictional test, readers try to change text size without guidance while the team observes errors and confusion.
UX & accessibility
The overall experience a person has while using a product to achieve a goal. It includes usefulness, clarity, accessibility and how the product feels across the whole journey.
Example: A fictional ticket app considers finding an event, registering and locating the ticket on arrival as one connected experience.
UX & accessibility
The controls and information through which someone interacts with a product. It includes elements such as buttons, menus, text and feedback, and is one part of the wider experience.
Example: A fictional music app's play button, track list and volume control form part of its interface.
UX & accessibility
Systematically learning about people's needs, behaviour and circumstances to inform product decisions. Different methods answer different questions, so interviews and usage data are often combined.
Example: A fictional recipe team watches people plan meals and asks about the workarounds they already use.
Prioritisation & delivery
A short description of a need from the perspective of the person using a product. It helps start a conversation about value and usually needs more detail before development.
Example: As a workshop organiser, I want to see who has replied so I can stop chasing people who already answered.
Strategy & discovery
Gathering evidence to assess whether a proposed problem, solution or assumption holds up. Positive feedback can support a decision, but does not by itself prove that a change caused an improvement.
Example: Five readers successfully use a prototype to save articles, supporting further testing of the idea.
Strategy & discovery
The main benefit a product offers a particular group of people. It explains why they might choose it over their current way of doing things.
Example: A fictional scheduling tool promises community organisers fewer messages to agree a meeting time.
Technology & platforms
An automatic message one system sends to another when a subscribed event occurs. It lets the receiver react to changes without repeatedly asking whether anything happened.
Example: When a fictional form service receives a workshop response, it sends the organiser's app a webhook so the attendee list can update.