Terms for starting a digital project

A strategic session is a short paid working meeting in which a business reviews its goals, customers, operations and digital systems with outside experts and leaves with a prioritized plan: what to do first, what can wait and what to drop. It is useful before committing budget to a redesign, a new system or advertising. Ours is a two-hour video session with written recommendations within 48 hours; it costs $500, credited toward a project started within 30 days.

More: Strategic Session

A fractional CTO is an experienced chief technology officer who works for a company part time: deciding on stack, architecture, vendors and hiring, and reviewing the work of developers, without the cost of a full-time executive. It fits companies that depend on technology but cannot yet justify a full-time CTO. Our base price, from $3,500 a month, covers about one day a week and runs month to month.

More: Fractional CTO

An extended strategic session is a deeper form of the strategic session for larger or more tangled questions: the team first spends several days working through the company's data, systems and market, then holds a full working day with the client. It suits decisions where a mistake is expensive, such as choosing a new platform, entering a new market or rebuilding digital channels. Ours starts from $5,000.

More: How Peretz Agency Works: Services, Process and Prices

Business diagnosis is the work of finding what actually limits growth before anyone names a solution, separating the symptom a company notices from its root cause. Owners tend to blame the nearest suspect, such as the website or the agency, and pay to fix the wrong thing: a request for more traffic is often a symptom when visitors already come but do not convert. A useful diagnosis decides whether a problem is external or internal, temporary or structural, financial or operational, and looks for numbers that do not match and evidence outside official reports.

More: The Columbo Method: How to Find What's Actually Wrong With a Business

A problem statement is a short description of the problem a product or project must solve, written in the customers' own words and backed by evidence that someone would pay to have it solved. Projects often arrive with a solution already attached, such as "we need a CRM", before anyone has defined what should change in the business. Without one, a team can deliver a faster, more modern website that changes nothing, because the owner actually needed more booked appointments, not a prettier homepage.

More: Architecting the MVP: What Marketing Owes a Product Before Launch

Business process mapping is documenting how work actually flows through a company today: who does what, where information is entered twice, where it gets lost and where work waits. It comes before choosing software, because technology can execute business logic but cannot invent logic that was never captured. The map is then used to rank fixes by impact against effort; for most companies such a diagnostic takes one to two weeks.

More: Business Process Automation in Kirkland: What Digitalization Actually Means (And When You Need It)

Business architecture is the description of what a company's systems need to know and who is responsible for it: customers, products, prices, markets, suppliers and orders, and which system is the source of truth for each. The sound order is to understand the business, map its processes, define ownership and only then choose systems. An architecture is a map of responsibility and ownership; a technology stack is only a list of tools.

More: The Software Should Fit the Business

Competitor analysis is the study of other companies in a category to understand how it works: what customers are offered, how sites and stores are built and where the gaps are. It is for understanding, not copying, and it has limits: it shows only what competitors chose to expose and may describe a company that has already changed. In one benchmark we reviewed 30 brands and tied 18 of them to a specific platform architecture through technical fingerprints such as URL structure, metadata, CDN paths and payment methods.

More: We Design for Reality, Not for Ideal Scenarios

Market research before a build is checking whether a product has a business case before money goes into making it: is there demand, who exactly is the audience, can they afford it, how large is the opportunity and what does it cost to reach one customer. Answering these questions late is expensive. Markets also move while a company waits: a platform idea postponed for four years can meet a different market, with competitors already holding the search positions.

More: Wrong Move. Right Move. Too Late.

A product roadmap is a plan of what will be built and in what order over a period of time, tied to business goals rather than kept as a list of features. Product management owns the strategy, the priorities and the roadmap itself, while product design decides what each planned item should become for its users. A roadmap is revised as the team learns; our strategic session ends with a 90-day roadmap of first moves.

More: Product Design vs UX/UI: Where Does a Digital Product Actually Begin?

Quick wins are changes a business can make right away, with little effort and without starting a project, that bring a visible result: fixing a broken form, clarifying an offer or removing an extra step from a purchase. They buy time and trust while larger decisions are prepared. After our strategic session they are listed in the written recommendations separately from the work that needs a budget.

More: Strategic Session

Investment sequencing is the order in which a business puts money into its website, SEO, content and advertising. Spending the whole budget on the website leaves nothing to bring visitors to it, and an online store captures demand but does not create it. Depending on its stage, a company may first sell through Instagram and influencers, prove the demand and only then invest in its own store.

More: Wrong Move. Right Move. Too Late.

A discovery phase is a paid stage before development in which an idea becomes documents that a price and a deadline can rest on: requirements, user flows, a clickable prototype, architecture, integrations, risks and a backlog with an estimate. It matters when quotes for the same idea differ widely. Our discovery phase takes two to four weeks, starts from $4,000, and its documents are written so that any team can work from them.

More: Discovery Phase

Product discovery is the part of product management that reduces the most dangerous uncertainty about a product while it is still cheap to change, moving from a question to evidence, an interpretation and a decision. It is a way of working, not a workshop. Discovery is enough when the team can say what problem it solves, for whom, how people solve it today, which constraints and risky assumptions exist and what the first version will not do. It is not the same as product discovery in e-commerce, which means how shoppers find products through search and filters.

More: Product Discovery: What Should You Know Before You Start Designing?

Requirements gathering is collecting and documenting what a client believes should be built: functions, users, rules, integrations and constraints. It records a belief, while discovery tests whether that belief is correct. Requirements also tend to move: the assumption that they will stay unchanged through a project is easy to make and hard to maintain, so a good process decides in advance how changes will be handled.

More: Product Discovery: What Should You Know Before You Start Designing?

Technical discovery is the part of discovery that studies the existing technical landscape: which systems a company already runs, where its data lives, which system is the source of truth, which APIs and limits apply and whether to build, buy or integrate. It has to inform product design from the start rather than be bolted on after the screens are drawn. Our architecture discovery goes through seven stages before any technology is recommended: business problem, assumptions, customers, existing system, cost of change, better questions to AI, and only then technology.

More: Product Discovery: What Should You Know Before You Start Designing?

Assumption mapping is a discovery technique in which the team writes down what it believes about users, the market and the solution, then ranks each belief by importance, uncertainty and the cost of being wrong. The fundamental, least proven assumptions are researched first. The method separates evidence (observed behaviour or data) from assumptions (believed but not shown) and hypotheses (still to be tested), and a simple "how do you know?" keeps a passing comment from becoming a requirement.

More: Product Discovery: What Should You Know Before You Start Designing?

Customer discovery interviews are conversations with potential users about how they deal with a problem today, held before a product is designed. The questions are about the last time the problem happened, not whether people would use or pay for something: past behaviour is evidence, future intention is only a hypothesis. Such interviews often reveal that a spreadsheet, an email chain or a WhatsApp thread already solves part of the problem in a way the proposed product has not understood.

More: Product Discovery: What Should You Know Before You Start Designing?

Stakeholder interviews are conversations with people inside the client company, from owners and managers to the staff who will use a new system, held at the start of a project. They reveal the teams, handoffs and decisions that shape the work, including those that are political rather than technical. This matters because a product can fail even when it works, simply because the organisation cannot adopt it.

More: Product Discovery: What Should You Know Before You Start Designing?

Hypothesis testing is checking a business assumption with the smallest possible experiment before investing in a full product. Before building, the team decides which hypothesis comes first, what minimum product can test it and which metric will show that it works. A hypothesis can often be tested for $500 instead of $50,000: a landing page, a small ad budget and analytics show who comes and which offers lead to inquiries before the system exists.

More: MVP Isn't a Minimal Product. It's the Cost of Entry.

Idea validation is a loop of claim, evidence, experiment, result and decision that checks whether a product idea deserves investment. A real test must be able to fail: showing a prototype to five people who say "looks great" is validation theatre, not evidence. The loop also has to change something, since an experiment after which no decision changes has produced information but not progress.

More: Product Discovery: What Should You Know Before You Start Designing?

A demand test, or pre-order test, is a landing page that offers a product before it exists and asks visitors for an action that costs them something: a deposit, a pre-order or card details. The success threshold is fixed before launch, and the problem is described in the customers' own words. It addresses the most common cause of failure: according to CB Insights, no market need is cited in 42% of startup post-mortems.

More: Landing Page Development

A fake door test is an experiment that offers a feature or product that does not exist yet, through a button, a page or an ad, to measure how many people try to use it. Those who click learn that it is coming, and the click rate shows real rather than stated interest. It is one of the cheapest ways to test an assumption without a prototype, alongside interviews, spreadsheets and pricing conversations.

More: Product Discovery: What Should You Know Before You Start Designing?

A concierge MVP is an early version of a service in which a person manually performs what the future software is meant to automate, such as matching requests, preparing reports or answering customers. Customers get the result, and the team learns which steps really matter before writing code for them. Sometimes the best prototype is a person pretending to be the product.

More: Product Discovery: What Should You Know Before You Start Designing?

Product-market fit is the point at which a product meets real, repeatable demand in a market large enough to sustain it. Without it, marketing does not fix a weak product; it only amplifies the weakness. A CB Insights analysis of 385 failed startups named product-market mismatch as the leading cause in 43% of cases, while running out of money was usually a later symptom.

More: MVP Isn't a Minimal Product. It's the Cost of Entry.

A project brief is a short document describing what the client wants to achieve, for whom and why, written before an agency or team starts proposing solutions. A good brief saves time; a good diagnosis saves the project. For a redesign aimed at leads, the brief defines one action per page type, the question each page answers, the proof and where it sits, the shortest form, a redirect map and tracking from day one. In AI work, a detailed brief often decides whether a pilot becomes a real product.

More: Why Most AI Pilots Stall Before Production, and How to Get Yours Live

A technical specification is a document that describes what a system should do and how: functions, user roles, integrations, data and constraints. It captures someone's current understanding of what should be built, not how the company actually works, which is why five agencies estimating one specification are estimating the same assumptions. A long specification is useful when research stands behind it; AI now makes it easy to write an impressive one that only describes the surface.

More: How to Choose a Web Development Company

A scope of work is the written description of a project's work with its stages, deliverables, dates and price, agreed before the work starts. Changes to it are agreed before they are made, not afterwards. Scope should follow the business goal: a project can be delivered exactly to contract and still create no value if the scope described the wrong thing.

More: How Peretz Agency Works: Services, Process and Prices

Scope creep is the uncontrolled growth of a project's scope after work has started, usually through small additions that each seem reasonable. One feature reveals an integration, the integration a data problem, the data problem a structural one, and under hourly billing the client carries the cost. Scope creep driven by indecision differs from intelligent adaptation to better information. It is often blamed for overruns, though hidden technical debt is the more usual cause.

More: How Much Does a Custom Website Cost in 2026?

A change request is a formal request to change a project's agreed scope, estimated and approved before the work is done. It separates new decisions from defects: "the checkout button doesn't work" is a bug to fix, while "redesign the checkout before launch" is a new decision with its own cost and timeline. Handling changes this way keeps the budget and the launch date predictable.

More: How Long Does an E-Commerce Build Actually Take, and What the Client Has to Do

A project estimate is a forecast of a project's cost and duration based on what is known about its scope at the time. Two agencies may quote $40,000 and $100,000 for the same brief because they are estimating different projects; a number given without questions about the client's processes reflects assumptions, not requirements. A complex project cannot be honestly estimated after a one-hour call or from a PDF. On a web application, a stage is ready to estimate once its users, roles and connected systems are written down.

More: How Much Does a Custom Website Cost in 2026?

Comparing quotes means bringing proposals from different agencies to the same scope before comparing their prices. Quotes of $18,000, $65,000 and $190,000 for one project usually describe three different projects. The check covers discovery, UX/UI, information architecture, migration, integrations, payments, tax, shipping, SEO migration, analytics, testing, documentation, training and post-launch support; with a large team, it is worth asking what each person will do.

More: Why E-Commerce Development Quotes in Seattle Range From $25 to $300 an Hour

A product backlog is an ordered list of the work still to be done on a product: features, improvements, fixes and technical tasks, with the most valuable items at the top. It is reordered as the team learns and is the main tool for deciding what goes into the next release. Our discovery phase ends with a backlog with a budget and timeline per stage, and outstaffed developers take their tasks from the client's backlog.

More: Web Application & SaaS Development

A project timeline is the schedule of a project's stages from start to launch, including the work and decisions on the client's side. As a guide, a store on a hosted platform with a theme takes weeks, a custom storefront with integrations and migration several months, and a custom professional website usually 6 to 10 weeks. Delays gather before development (scope, content) and right before launch (testing, approvals), so a realistic timeline shows the client's track too: decisions, product data, access, feedback and acceptance.

More: How Long Does an E-Commerce Build Actually Take, and What the Client Has to Do

An MVP, or minimum viable product, is the first version of a product with just enough features for real users to try it and pay for it, so that demand is tested by the market rather than by opinions. A good MVP is cut in scope, not in quality: its data model and code should survive growth. Our MVP projects start from $20,000 and usually take two to four months.

More: MVP Development

An MVP review is an independent check of a first product version that people already use, often built quickly or with AI tools, covering its code, security, product logic and market. It ends with a written verdict: keep, fix or rebuild. A narrow internal tool without payments or sensitive data can usually be kept and a working product with growing users and an understandable structure fixed, while payments or personal data on a base without access control call for a rebuild. Ours starts from $3,000, most cost $3,000 to $5,000, and the price is fixed before the review starts.

More: MVP Review

Phased delivery is splitting a large product into stages, each with its own scope, estimate and release, instead of building everything for one launch. The first stage carries the flow that earns money or saves the most time, and each stage is designed to carry the next. When a good idea is too big to fund at once, staging it and cutting whatever the first test does not need is often the way forward.

More: Web Application & SaaS Development

A go-to-market strategy is the plan for how a product reaches its customers: through which channels, with what message, on what budget and in what order. It is tested in parallel with the build: where the target customer spends attention and whether that channel can be reached at a viable cost. An excellent product with no viable distribution path is not an MVP but an expensive project with few users, and a new website without a marketing budget or traffic plan is a costly object nobody sees.

More: Architecting the MVP: What Marketing Owes a Product Before Launch

Build vs buy is the decision whether to develop custom software or use a ready-made product, usually software as a service (SaaS). Both sides have changed: platforms do far more than they used to, and custom development has become faster with modern frameworks and AI. The choice is often not permanent: a company can buy and configure first, learn how its process really works, then extend or build.

More: The Build vs. Buy Equation Has Changed. Most Companies Have Not Recalculated It.

SaaS sprawl is the uncontrolled growth of a company's software subscriptions: tools added by different teams, overlapping functions and licences nobody uses. A $500 subscription is rarely a $500 system, because connectors, extra modules and higher tiers accumulate around it. According to Zylo's 2026 index, SaaS spending grew 8% year over year while 36% of licences went unused.

More: The Build vs. Buy Equation Has Changed. Most Companies Have Not Recalculated It.

Total cost of ownership, or TCO, is the full cost of a system over its working life, not only the price of building or buying it: subscriptions, apps and transaction fees, hosting, support, upgrades, testing, further development and the cost of working around platform limits. Comparing a one-off build quote with a monthly subscription without TCO is a common mistake. In our platform comparisons we look five years ahead.

More: Leaving Magento in 2026: Upgrade, Shopify Plus or Custom Laravel?

Post-launch operating costs are what a website or application costs to keep running after release: hosting, security, updates, monitoring and fixes. A common planning figure is 10 to 20% of the initial build cost per year. For a professionally maintained small business website, hosting, maintenance and updates come to about $1,200 to $3,600 a year, of which the domain and hosting alone are $200 to $500.

More: What Actually Makes E-Commerce Development Expensive

Per-seat licensing is a pricing model in which software is paid for per user per month, common for CRM, ERP and collaboration tools. It looks cheap at the start and grows with the team: at 2026 list prices, 20 users over five years cost $120,000 to $234,000 on Salesforce Sales Cloud and $109,500 on HubSpot Professional, while a custom CRM starts from $90,000. At 10 users HubSpot, at $55,500, comes out cheaper, and a $30-per-user CRM becomes expensive once it needs $50,000 of implementation.

More: Custom CRM Development

Website cost tiers are the typical price ranges for a site depending on who builds it and how. In the Seattle market in 2026, website builders cost $20 to $80 a month, freelancers $1,500 to $6,000, small agencies $5,000 to $15,000, custom sites $15,000 to $50,000 and more, and enterprise projects $50,000 to $150,000 and more. At $5,000 a client pays for execution; at $20,000, for strategy, custom design, conversion work, SEO architecture and an experienced team.

More: How Much Does a Website Cost in Seattle in 2026?

Outgrowing a platform means reaching the point where a website or system built for an earlier stage of the business no longer fits how it works. Typical signs: new services do not fit the navigation, documents are prepared by hand, inventory lives in spreadsheets, nobody trusts the CRM integration and SEO has plateaued. Businesses rarely regret starting on a website builder; they regret staying on one after outgrowing it.

More: Why Growing Bellevue Businesses Eventually Outgrow Their Websites

Cost of change is what it takes to modify a system after launch, and it is the real test of an architecture. The first launch says little; what matters is the price of the twentieth change, such as a fifth market, a new dealer flow or new pricing logic. An architecture that keeps this cost low lets a business adapt without rebuilding, while one that raises it turns every new idea into a separate project.

More: The Software Should Fit the Business

Portability is the ability to move a company's domain, code, database, content, customer records, analytics history and workflows to another vendor or platform without losing them. It is the practical side of ownership: holding data is not enough if it cannot work anywhere else. The cost of leaving a vendor is worth knowing before signing up, not discovering when the relationship ends.

More: What You Own vs What You're Renting: The Digital Ownership Audit

Digital assets are the domains and DNS, hosting, source code, CMS licences, analytics, ad accounts, CRM and email systems, APIs and user accounts a business runs on, together with its content. Each of them should be held on a company account rather than a personal one. Maintained assets such as articles, documented processes and case studies gain value over time instead of depreciating, which becomes visible when a company is sold.

More: What Happens to the Website After an Acquisition?

Digital ownership is a company's actual control of its domain, DNS, hosting, code repository, analytics, Search Console, payment and email providers through accounts it holds itself. An admin login or a data export is access, not ownership: a contract can transfer a website while passwords and keys stay with people who have left. Each component can be sorted as owned, rented (an understood dependency) or dependent on a person or vendor nobody else can reach. Who will own the site after launch is best settled before deciding how it is built.

More: Most Companies Don't Own Their Website. They Own a Collection of Dependencies.

Vendor lock-in is a dependence on a supplier that makes leaving it costly or impossible within a reasonable time. A simple test is whether the company could replace its developer, agency, host, CRM or platform without losing critical functions. A data export can be useless without the proprietary logic around the data. The risk is practical: Bitrix24 announced that its cloud would stop working in Ukraine on 1 June 2023, and users stood to lose their portals and data.

More: What You Own vs What You're Renting: The Digital Ownership Audit

Audit before quote is the practice of examining an existing site or system before agreeing a price for work on it. The usual order of discovery call, proposal, contract, audit and development fixes the price before hidden technical debt is found, and the surprises then arrive as extra invoices. Running the audit first, in days rather than months, gives a quote based on what is actually there.

More: The Hidden Technical Debt That Makes Website Redesigns More Expensive Than Planned

Technical due diligence is an independent review of a company's code, infrastructure, security, licenses and control of domains and accounts, carried out for a buyer or investor before a deal. It shows what is really being bought, what depends on specific people and what the fixes will cost. Our session costs $1,500 and a full review starts from $7,500; most engagements take five to ten business days.

More: Technical Due Diligence Session

A digital ownership audit is a review of what a business actually owns and what it rents or depends on across its website, code, data, tools, vendor contracts and key people. It answers one question: what breaks if a developer, an agency or a platform disappears tomorrow. Ours ends with a map of ownership and dependencies and a priority list; it starts from $1,000 and takes five to ten business days.

More: Digital Ownership & Debt Audit

Digital debt is the accumulated financial and operational burden in a company's digital estate: redundant subscriptions, unused services, uncontrolled vendor access, code nobody understands and basic ownership gaps, such as a domain tied to someone's personal email address. Unlike technical debt, it is not only about code. It grows quietly and tends to surface during a sale, an investment round or the departure of a key person.

More: Digital Ownership & Debt Audit