# Choosing a tech stack for a first product: no-code, WordPress, Shopify or custom code Source: https://sarang.pro/choosing-tech-stack-first-product/ Author: Sarang | Published: 25 October 2026 The first technical decision most founders make is also one of the stickiest: what to build the product on. Pick a tool that is too limited and you rebuild within a year. Pick something too ambitious and you spend your runway on engineering before you know whether anyone wants the product. This guide compares four common starting points for a first product: no-code builders, WordPress, Shopify and custom code. It gives a way to choose between them based on what you are testing, who will maintain it and how easily you can leave. ## Start with the question, not the tool The UK Government's Service Manual has a short guide to [choosing technology](https://www.gov.uk/service-manual/technology/choosing-technology-an-introduction). Its first priority is to make choices that let you "change your mind at a later stage". It also tells teams to aim to "minimise the total cost of ownership", avoid "getting locked in to long contracts for specific tools and providers", and "make sure you always have full control of any data you store". Those principles were written for government services, but they suit a start-up well. A first product is an experiment. The stack should make the experiment cheap to run and cheap to change, and it should let you take your customers and data with you if you outgrow it. Before comparing tools, answer three questions in writing: - What exactly are we testing? Demand for a product, a new way of working, or a technical idea that has never been built before? - Who will run it day to day? A founder with no coding skills, a freelance developer, or an in-house team? - What must we own? Customer lists, content, payment relationships, source code? ## The four options at a glance OptionBest forMain trade-off No-code builderTesting a workflow or internal tool quicklyLimits on custom logic and on moving away WordPressContent-led sites, blogs, membership and simple shopsYou or your host must keep it updated and secure ShopifySelling physical or digital products onlinePlatform rules, app costs and fees shape your margins Custom codeProducts whose core value is new softwareHighest cost and time, and you need people to maintain it ## No-code and low-code builders No-code tools let you build apps by arranging components and rules on screen instead of writing code. Microsoft's [Power Apps overview](https://learn.microsoft.com/en-us/power-apps/powerapps-overview) describes its product as letting users "create feature-rich, custom business apps without writing code", while also allowing professional developers to extend it with code and custom connectors. Many website builders, database tools and automation services work on the same idea. No-code is a strong choice when the product is mostly a workflow: forms, approvals, a simple marketplace, an internal dashboard. You can put something in front of users in days. The risks are about the edges. Complex pricing logic, unusual integrations or heavy traffic can push past what the tool allows. Check early whether you can export your data in a standard format, and assume the screens and logic themselves won't come with you. If you later move to custom code, plan on rebuilding the product rather than migrating it. ## WordPress WordPress is the most widely used content management system on the web. W3Techs, which surveys websites, reports in its [CMS usage figures](https://w3techs.com/technologies/overview/content_management) (dated 24 October 2026) that WordPress runs on 39.6% of all websites. That reach means a large pool of developers, themes and plugins. The WordPress project describes itself on its [about page](https://wordpress.org/about/) as "open source software" licensed under the GPL, which gives you the freedom to run, study, change and redistribute it. Practically, this means you can host it anywhere and move between hosts. The flip side is responsibility. The official [installation guide](https://developer.wordpress.org/advanced-administration/before-install/howto-install/) expects a web server with PHP, a MySQL or MariaDB database, and someone to create the database, edit the configuration and upload the files. Managed WordPress hosts take on much of this, but someone still has to decide which plugins to install and keep them updated. Plugins and themes are code written by third parties, so each one you add is another thing to keep updated. Budget time for this every month. WordPress fits when content drives the business: articles, courses, guides, a portfolio. It can run a shop through plugins, but a store with many products and frequent orders usually runs more smoothly on a dedicated commerce platform. ## Shopify Shopify is a hosted commerce platform. It handles hosting, checkout, security updates and much of the payment setup, and it has a theme and app marketplace for extra features. W3Techs puts it on 5.3% of all websites. For a product business, this removes a lot of work. You can launch with a theme, add products and take payments without a developer. Read the commercial terms carefully. Shopify's help centre explains that [third-party transaction fees](https://help.shopify.com/en/manual/payments/third-party-providers) "apply to all transactions processed through third-party and alternate payment gateways", even when Shopify Payments is turned on, with an exemption for PayPal when Shopify Payments is your main provider. Monthly app subscriptions also add up. Model your margin with platform fees, payment fees and app costs included before you commit. Shopify suits founders whose advantage is the product, brand or marketing, not the software. If you need very unusual checkout logic or a non-retail business model, you may spend more effort working around the platform than with it. ## Custom code Custom code means developers build the product with a programming language and framework, hosted on infrastructure you choose. You get full control over features, data and user experience. It is the right choice when the software itself is the product and no existing platform does what you need. It is also the most expensive route. You pay for design, building, testing, hosting and ongoing maintenance, and you depend on developers who understand the code. If you go custom, reduce the risk. Own the code repository in a company account. Use common, well-supported frameworks so you can hire for them later. Use hosted services for the commodity parts, such as payments, email and authentication, so your developers spend time on what makes you different. ## A worked example An example (illustration only). A founder wants to sell monthly boxes of specialty tea and later add a feature that recommends teas based on a short taste quiz. - What is being tested: whether people will subscribe and stay subscribed. The recommendation engine is a later idea. - Who runs it: the founder, with a freelancer for a few days a month. - What must be owned: the customer list, subscription data and the brand. A sensible first stack is Shopify with a subscription app, plus a quiz built in a no-code form tool that tags customers by preference. The founder checks that customer and order data can be exported as CSV. Only once retention is proven does it make sense to pay for a custom recommendation service that plugs into the store. Had the founder started with custom code, the first few months and much of the budget would have gone into building checkout and subscriptions, which are already solved problems. ## Common mistakes and how to decide - Choosing for the company you hope to be in five years. Choose for the next experiment, and make sure you can leave. - Letting an agency or freelancer own the accounts. Domain, hosting, platform and code accounts should belong to the business. - Ignoring running costs. Plugins, apps, hosting, fees and maintenance often cost more over a year than the initial build. - Building custom versions of solved problems. Payments, login and email are rarely where a start-up wins. - Skipping the exit check. Before you commit, find out exactly what data you can export and in what format. As a rule of thumb: sell products, start with Shopify. Publish content, start with WordPress. Test a workflow, start with no-code. Build custom code when the software is the thing customers pay for and you have the budget and people to look after it.