We had a client come in the other day, and during our discussion, he asked a question that almost all our clients do: does your company offer custom website development, or do you also do template-based website building? I told him we only do custom development, and we will never, ever do template-based work. Then I gave him a real-life example from our actual business operations, not a made-up one.

Years ago, we were in talks with a certain association in Beijing – I can’t name it specifically. It was our first time working with this client, who’s based near North 3rd Ring Road. When we first met, the usual process followed: they checked out our company, reviewed our quotes, looked at our past projects, and assessed how well we could understand and deliver on all their requirements. After about a week to half a month of discussions, they finally chose us, and we set out to develop the website for them with full sincerity.

Now, we have a partner company that specializes in template-based website building systems. They’d been telling me for ages to give their system a try. They kept saying the biggest downside of custom development is the long lead time – that while we provide open-source code, a project that would take us around 20 days could be finished in just 1 or 2 days with their template system. As fellow tech professionals, I knew template systems have their drawbacks and limitations, but I wasn’t entirely sure what those limitations entailed, so I’d always declined to try it. But they kept pushing it – we have mutual friends, so they kept asking me to give it a shot.

Their main motivation, I knew, was that our company has a lot of high-profile successful cases, including projects for state-owned enterprises and the Ministry of Education. They thought if we used their system for these well-known clients, it would be a great promotional tool for them. I saw right through it, but still refused. Then this small project with the association came along, and it seemed to fit all the client’s requirements perfectly. With them hounding me nonstop and our mutual friends putting in a good word, I thought fine, I’d give it a try.

So I used their template system for this association’s project. And sure enough, as they’d said, a project we initially estimated would take about 30 days was completed in just three to five days for the back-end development – we still did the front-end design the usual way, of course. The client was really satisfied with the end result; all the functions were implemented, the website went live and was working smoothly, and I actually thought the system was pretty good. After that, I started to feel template-based systems might not be as bad as people said.

I’d planned to use the system for future projects, but that whole year, we hardly got any projects that suited it. Most of our work was custom software development, e-commerce platforms and other complex projects, so I never got the chance to use it again.

Then early the next year, this association client came back to us for a second round of upgrades. They needed a single landing page for a national campaign – just two pages total, a promotional page and a registration page. When users clicked the sign-up button at the bottom of the promo page, they’d be directed to the registration page to submit their information to the back end. Simple enough, right? But here’s the ridiculous catch with that template system: user registration was forced into three separate steps across three pages. First, username and password; second, detailed personal information; third, a verification step. These three steps were hardcoded into the system – you couldn’t separate or modify them, and every registration process on sites built with that system had to follow this exact flow.

The client asked why we couldn’t change it, and I had no good explanation to give them. We went back and forth for days, but it was impossible to modify the system, and the client got really frustrated. They thought we were incompetent, that we either couldn’t or wouldn’t make the change for them. In the end, we had to make do and get the campaign page up as best we could, but right after the campaign wrapped up, the client dropped us on the spot and hired a new company to rebuild the entire website from scratch.

This incident really left a lasting impression on me. If we’d lost the client because of our own technical limitations – if we truly couldn’t meet their requirements – I’d have accepted it without a word. But this time, it was all because we used a template system. A feature that would’ve been trivial for us to build with custom development was impossible to implement with the template, and it cost us the client.

That’s why I always say template-based website building will eventually lead to a falling-out with the client. Sooner or later, the rigid limitations of the template will make it impossible to meet the client’s needs, and they’ll reject all the work you’ve done for them. After that incident, I told everyone in the company that we would never, ever use template-based tools again. Even if custom development takes longer and is more challenging, we stick to fully custom-built solutions for everything we do – no templates, ever.