Episode 154

scandiweb: Why AI Will Break the Billable-Hour Model

Alfreds Genkins
Alfreds Genkins
CTO

In this episode we talked about:

  • How Alfreds went from coding at 11 and interning at scandiweb at 16 to CTO, and how the company grew from under 100 people to roughly 500 or more
  • The internal experiment where a project manager beat the CTO at building an app with AI, and why technical expertise can slow you down
  • Why AI is moving from coding assistant to something that can solve business problems
  • Why AI lowers the cost of change, and what that means for the Shopify versus Magento tradeoff of simplicity and flexibility
  • Why scandiweb is building products like Expedio, which Alfreds says reached roughly 6 times faster responses
  • The conflict of interest AI creates for agencies that sell hours, and why a 100-hour retainer that now takes 20 hours is the elephant in the room
  • Three directions for agencies: smaller merchants and projects, more value for existing clients, and charging for value instead of hours
  • Why easier agency switching makes proactive service and human relationships more important
  • How Ari plans, writes, tests, and prepares Magento changes in an isolated environment, with humans involved as much as the merchant wants

🎧 Listen now on Apple Podcasts, Spotify, or YouTube

Episode highlights:

01:20 From 11-Year-Old Coder To scandiweb CTO

02:44 When A Project Manager Beat The CTO At Building With AI

06:35 Why AI Lowers The Cost Of Change Across Platforms

10:00 The Elephant In The Room: AI And The Billable Hour

13:53 Why Switching Agencies Gets Easier, And Relationships Matter More

16:48 Ari: An AI Agent That Plans, Builds, And Tests Magento Changes

Alfreds’ bottom line: AI is forcing agencies to rethink what they sell. Alfreds Genkins, CTO of scandiweb, says that when AI can deliver 100 hours of work in 20, the merchant should not keep paying the same bill, and agencies that sell hours need a new model. He sees three paths: serve smaller merchants and projects that never made economic sense before, deliver far more value to existing clients, and move toward charging for value instead of development hours. He also expects switching agencies to get easier as AI reads unfamiliar codebases, which makes proactive service and real relationships the reason clients stay. Scandiweb is putting this into practice with Ari, an AI agent that connects to a Magento codebase, plans and writes the change, tests it in an isolated environment, and prepares it for release, with humans involved as much or as little as the merchant wants. His advice to merchants: dream bigger, because AI makes ideas possible that were previously too expensive to attempt.

FAQ

Alfreds Genkins is the CTO of scandiweb, an ecommerce agency based in Latvia that has been around for more than 10 years and traditionally focused on Adobe Commerce and Magento. He started coding at 11, did a summer internship at scandiweb at 16, and became CTO about ten years later. When he joined, scandiweb had fewer than 100 people; today it has roughly 500 or more.
In an internal experiment, Alfreds and a project manager each built the same app with AI. The project manager finished faster, with more features and acceptable code quality. Alfreds realized his technical expertise slowed him down because he spent too much time telling the AI which architecture, technologies, and code structure to use. The project manager, used to turning business requirements into tasks, simply described what needed to be achieved and let the AI handle much of the implementation.
Alfreds’ idea is that capabilities that once took years to build are becoming much cheaper to reproduce with AI. Historically, Shopify offered simplicity and Magento offered flexibility at the price of complex development and expensive maintenance. As AI lowers the cost of change, those platform limits become less permanent. Because Magento is open source, scandiweb can study the entire codebase and bring in the best ideas from other platforms, such as Shopify’s rendering approach, which he says helped its Expedio project reach roughly 6 times faster responses.
If an agency has billed 100 hours a month to maintain a store and AI lets it do the same work in 20, the merchant would reasonably expect the bill to fall by about 80%. For an agency that sells hours, that means cutting its own revenue, while it still needs to pay the people doing innovation work. Alfreds calls this the elephant in the room and says agencies need a sustainable model rather than ignoring the problem.
First, open the doors to smaller merchants and projects that agencies used to turn down because the economics did not work, accepting smaller average project sizes but many more clients. Second, deliver more value to existing clients, including becoming proactive when AI lets an agency finish a year’s backlog in two months. Third, rethink pricing by moving away from selling development hours toward charging for the value delivered, with merchants choosing how much human expertise they need.
Alfreds expects so. Switching used to be painful because a new agency needed weeks to learn the codebase, customizations, and integrations. AI agents can now analyze an unfamiliar codebase and start on tasks almost instantly. He says that makes human relationships more important, not less, and he sums up the future of agencies as the survival of the friendliest: the winners will be the ones that understand their customers, create value proactively, and give exceptional service.
Ari is scandiweb’s AI agent for Magento stores. A merchant describes a change in natural language, such as improving checkout, fixing a promotion, or optimizing performance. Ari connects to the codebase, investigates how it works, asks non-technical questions to align on expectations, then plans the work, writes the code, tests it, and prepares the release. It runs in an isolated environment so the live store is never affected, and when something needs a business judgment it asks the merchant instead of assuming.
Yes. Alfreds says merchants can bring Ari to their current development partner so they get the productivity benefits of AI while keeping the people they already know and trust. Humans can stay involved in production releases, and merchants decide how much human involvement they want, whether that is their existing agency, an internal team, or scandiweb.
Dream bigger. He encourages merchants to bring their most ambitious ideas to their technology partners and challenge limits they have accepted for years, because AI makes it possible to solve problems that were previously too expensive to attempt. In his words, the future of ecommerce is not about writing more code but about making better ideas possible.
Toolbox listeners get an exclusive free trial of Noibu’s AI agents for CRO, A/B testing, performance, accessibility, ROAS, and bug resolution. The agents start by auditing your store and ranking the biggest opportunities, from slow pages to broken journeys. Then they build each change, wait for your approval, ship it, and measure what it did. The trial runs until you have at least two changes live and measured. Get started at noibu.com/podcast-audit.

Want to be featured on The Ecommerce Toolbox?

We feature senior ecommerce leaders with real things to say. If you’ve led something worth talking about, we want to hear from you.

Apply to be a guest →

Other episodes

See all episodes

Audit your site today

Supervised AI agents find what's broken, slow, or leaking revenue, do the work, and hand it back for your approval.