How to communicate value of changing to modern web development practices to boss

[Onion]

I am currently a wordpress developer. I work for a small web design agency. I think I'd like to change the development practices and stack we use to being something like Next.JS with Tailwind so it uses a SSG and then only does dynamic stuff for very interactive components. it also means we can host it in a serverless way mostly using Vercel or SST. It also means we can do more high-end business and develop custom themes and command higher rates as we're not limited by some of the baggage wordpress brings. I also want a challenge by integrating things like version control, CI/CD, and self-hosting using tech like SST. Wordpress also just gets in the way, even with tech like Roots Sage as there isn't that good of hosting support for it other than hosting yourself. In that case, it's a security nightmare managing the server myself.

I'd like some advice on going about this. I am not a natural salesman. However, I am reading more about sales and influencing others, but I don't really know how to communicate the value of these better development practices to the boss who just sees the brief go to us, and we make it happen. I know we can deliver more quality and more robust solutions to the client, but I need some help figuring out how to do this. Another challenge I have is they underestimate how hard it can be to make wordpress and the current crop of themes work and to change them; they have a beginner level of elementor knowledge under their belt and they're very confident that they know best no matter what. They're overall a great boss but I don't think we can deliver better things unless we upgrade to more modern development practices. Any suggestions would be greatly appreciated.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Andreas Thiel]

I am currently a wordpress developer. I work for a small web design agency. I think I'd like to change the development practices and stack we use to being something like Next.JS with Tailwind so it uses a SSG and then only does dynamic stuff for very interactive components. it also means we can host it in a serverless way mostly using Vercel or SST. It also means we can do more high-end business and develop custom themes and command higher rates as we're not limited by some of the baggage wordpress brings. I also want a challenge by integrating things like version control, CI/CD, and self-hosting using tech like SST. Wordpress also just gets in the way, even with tech like Roots Sage as there isn't that good of hosting support for it other than hosting yourself. In that case, it's a security nightmare managing the server myself.

I'd like some advice on going about this. I am not a natural salesman. However, I am reading more about sales and influencing others, but I don't really know how to communicate the value of these better development practices to the boss who just sees the brief go to us, and we make it happen. I know we can deliver more quality and more robust solutions to the client, but I need some help figuring out how to do this. Another challenge I have is they underestimate how hard it can be to make wordpress and the current crop of themes work and to change them; they have a beginner level of elementor knowledge under their belt and they're very confident that they know best no matter what. They're overall a great boss but I don't think we can deliver better things unless we upgrade to more modern development practices. Any suggestions would be greatly appreciated.

How familiar are you with all the components involved?
It seems like you would have to work on several hundred little projects that eventually have to converge.

How many people are open to the idea at this point? Do you want to provide a proof of concept to get everybody on board?

I personally think the difficulty level of this is off the charts.
When I tried to play the thought of what creating a more modern alternative to WordPress could look like, it was pretty disheartening. Among other issues, the whole agile push means that creating something generic has gotten more difficult.

Why would developing custom themes become easier after these changes? I fear most of the hopes are pretty vague and it wouldn't be as easy as you think to make themes more easily customizable.

 

[EquinoxS]

I think it's great you're thinking about options to improve the business!

As always you should convey the value of your suggestions. I can speak from my experience about WordPress, Next and so on. Actually, I learned WebDev the other way around than most other people. I started with the MERN stack and later switched to NextJS. I have to point out that I'm not using those tools professionally. I'm a self-taught dev and have only created projects for fun so far. The idea of starting a web dev side hustle has been on my mind for quite some time, however. By now I'm quite proficient with NextJS including Tailwind.

Since NextJS is over-engineered for many simple applications I looked into WordPress, Webflow, and Framer.

With that said, here is my personal summary: WordPress is a pure nightmare.
It was so hard to get into it and I kept thinking to myself "Wow, it was that much easier just writing the code by hand". You have to manage a gazillion plugins, the whole dashboard is counter-intuitive and there are so many layers of menues and submenues... just a horrible developer experience.

Any of the other No-/Low-Code tools will be faster to develop, cheaper (costs for plugins!) and easier to maintain (again, plugins and compatability issues).

Webflow is great if you are more on the development (not design) side of things since it's closest to writing code. It's also more powerful than Framer.

Framer is great if you have design experience with Figma and really want to reduce code to a minimum. Best for sites with little functionality.

NextJS is awesome for site speed and probably necessary for complex applications. Probably not needed if you managed to work with WordPress till now.

Thus, your line of argument should be centered around time, maintenance costs and to a lesser extent user experience. Maybe it's just me but I can almost immediately tell if a site is made with WordPress. It just looks old-fashioned.

 

[rpeck90]

I'm in a similar position to yourself - several clients who are stuck with a monolithic web stack (one uses Wordpress and several use Shopify). I've suggested the same to them (the term you're looking for is "headless", with another being JAMStack) but there is little you can say that will persuade them to change.

Whilst they are likely amenable to new ideas, their primary concern will be getting stuck with a stack that a) they don't understand, b) will have to pay expensive developers to fix / upgrade, c) not be able to manage properly.

It's a phenomenon I call the "adoption cost", which is basically how "expensive" it is to switch to a solution beyond the monetary impact. In this case, the adoption cost is to move away from a "proven" solution to something that most don't understand and will, therefore, be incapable of integrating effectively into their business.

The only way around it is to develop the technology yourself and demonstrate that it's a better solution, whilst being able to interface with the current stacks the company currently uses. In order to do this, you need to lead with the one thing most people care about, which is money. You demonstrate how your solution can built a richer, more secure, powerful and effective web front-end whilst still benefitting from the admin backend provided by Wordpress/Shopify etc.

You can see part of an email I sent to one of my larger clients on the topic here: -



This company is at the point where they need to consolidate their marketing, technology and backend processes into a more substantive system that can be used to help accelerate the business's growth.

Adopting a headless design pattern would contribute to that by giving the company the opportunity to design an interface exactly how they need and (most importantly) push them to think systemically around both the technology and team aspects of its adoption.

They don't need "headless" on its own. It needs to form part of a wider solution which will "transform" their business into something more robust, substantive and effective. That's basically what they're looking for.

To do this, I have written a large document explaining how they need to fix various parts of their system to accommodate the growth. To be clear, these have NOT pursued this avenue presently, so I am not advocating you do this - really just showing what I've been doing to address the issue.

It's their choice as to what they end up doing: -



And, remember, the above is coming from a guy who gets paid to do Shopify + Wordpress code. You can see a child theme I created for another client I work with who primarily use Kadence here: -



This streamlines the development process of their WP sites by standardizing the various attributes of the WP backend, namely plugins, CPT's and some other aspects. The aim with it is to be able to replicate instances of Wordpress sites by using OCDI.

The basis of the theme is Redux, which allows us to create a large number of customization options, which can be managed through the backend code: -





A bit off-topic, but as you're a WP dev, one of the really cool things about the above is that it integrates natively into Kadence's "dynamic content" system. This means that we can use the above inputs inside the Kadence Gutenberg blocks: -



If you have further questions, I'd definitely be interested in talking. I have been working on an engine to provide the means to develop & deploy "headless" sites whilst interfacing with existing Shopify + Wordpress data.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Choate]

A startup I work at is possibly making the transition from WordPress to Next.js because there are a handful of our other websites using the React system and it makes sense to standardize when you have a user dashboard, logins, three or four different websites, etc. all under one brand. Using React components makes it easy to standardize styles and lower costs by having a team of engineers who can work on a single technology. This is like a one year+ process for a single business and with a handful of the best full stack engineers working on it.

With that said, it does not make any sense at all for a small web design agency to make the transition to Next.js. The needs that a small web dev agency fulfills are way different than what you'd do with Next.js.

On the business side of things, the companies that your agency supports might not even have the budget to increase their web spend - so when you think of stakeholders you need to sell this idea to, it's not only your boss, but all of the businesses you serve. On the development side of things, you don't need things like version control for a local business like restaurants, service businesses, franchises, etc. Most small businesses don't need to overengineer their website which is what a transition to Next.js would be.

It also means we can do more high-end business and develop custom themes and command higher rates as we're not limited by some of the baggage wordpress brings. I also want a challenge by integrating things like version control, CI/CD, and self-hosting using tech like SST. Wordpress also just gets in the way, even with tech like Roots Sage as there isn't that good of hosting support for it other than hosting yourself. In that case, it's a security nightmare managing the server myself.

Hosting is the easiest thing ever with WordPress. Managed hosts like WP Engine, Kinsta, Flywheel, etc provide a premium service that can be easily baked into retainer fees. You can also get into middle of the road hosts like SiteGround where you can throw 20 small websites on a $400/year plan. And if you prefer to do it more yourself, Cloudways w/ Digital Ocean or just straight up Digital Ocean are both options.

Another challenge I have is they underestimate how hard it can be to make wordpress and the current crop of themes work and to change them; they have a beginner level of elementor knowledge under their belt and they're very confident that they know best no matter what. They're overall a great boss but I don't think we can deliver better things unless we upgrade to more modern development practices. Any suggestions would be greatly appreciated.

This is a workflow, mastery and tool issue. WordPress is not inherently difficult, in fact it's probably the easiest and most flexible way to quickly deliver websites at low cost.

Workflow - Something is probably lacking in terms of how websites are imagined, designed, and developed

Mastery - Any tool will get the job done, even Elementor, but without a full mastery things will get difficult

Tools - There are better solutions than Elementor available. Most of this plays into mastery, as in any tool will get the job done, but for me personally I find that YooTheme Pro and Bricks are the best currently available.

I'm in a similar position to yourself - I have several clients who are stuck with a monolithic web stack (one uses Wordpress and several use Shopify). I've suggested the same to them (the term you're looking for is "headless", with another being JAMStack) but there is little you can say that will persuade them to change.

Whilst they are likely amenable to new ideas, their primary concern will likely be getting stuck with a stack that a) they don't understand, b) will have to pay expensive developers to fix / upgrade, c) not be able to manage properly.

Headless is a very complex and niche solution that would have very little benefit for a generalized small web dev agency. Similar to OP's desire to transition presumably dozens of small business websites to pure Next.js, there's no reason to overengineer things. It will not transform their business into anything that it's not already.

---

If OP wants to work with modern stacks and is capable of using Next.js, they are underemployed working at a small web design agency using Elementor. The solution is more to start a side hustle building websites the way OP wants to build them and then transition that into their own agency—not rebuild someone else's company completely in a new technology.

 

[Onion]

How familiar are you with all the components involved?
It seems like you would have to work on several hundred little projects that eventually have to converge.

How many people are open to the idea at this point? Do you want to provide a proof of concept to get everybody on board?

I personally think the difficulty level of this is off the charts.
When I tried to play the thought of what creating a more modern alternative to WordPress could look like, it was pretty disheartening. Among other issues, the whole agile push means that creating something generic has gotten more difficult.

Why would developing custom themes become easier after these changes? I fear most of the hopes are pretty vague and it wouldn't be as easy as you think to make themes more easily customizable.

Thank you Andreas for replying.

I'll admit, half of my original post was half-baked. Sorry about that. And yes, you're right, a lot of what I said is vague and is convincing no boss in its current state.

With how we currently do things, we're usually picking a theme from themeforest or some theme provider. Then we're being told to use the page builder which is attached to the theme to build the pages. The boss and clients then want us to make lots of styling, content, structure changes and add dynamic elements here and there and override a lot of the themes core structure. Depending on the page builder, this is very doable, but a lot of the time it is slow and cumbersome. Not to mention, these page builders don't usually support version control (Elementor has page revisions but it is buggy). The development pipeline alone using something that is closer to the code is always nicer, for me at least. I can setup visual regression tests to make sure things don't break. I can setup CI/CD pipelines with version control. Using something like tailwind also feels like there is less thinking about what to call things and more styling. And I'm a lot more grateful for hot reload after having to work with page builders trying to change something. It sucks to wait for 15 seconds for the page builder to load 1000 things, then see it not right, then rinse and repeat.

I'm a open book. I'm still learning my craft and if you think I'm missing something here, please let me know. I want to master what I do, and I'm more than open to feedback to get better.

 

[Onion]

I think it's great you're thinking about options to improve the business!

As always you should convey the value of your suggestions. I can speak from my experience about WordPress, Next and so on. Actually, I learned WebDev the other way around than most other people. I started with the MERN stack and later switched to NextJS. I have to point out that I'm not using those tools professionally. I'm a self-taught dev and have only created projects for fun so far. The idea of starting a web dev side hustle has been on my mind for quite some time, however. By now I'm quite proficient with NextJS including Tailwind.

Since NextJS is over-engineered for many simple applications I looked into WordPress, Webflow, and Framer.

With that said, here is my personal summary: WordPress is a pure nightmare.
It was so hard to get into it and I kept thinking to myself "Wow, it was that much easier just writing the code by hand". You have to manage a gazillion plugins, the whole dashboard is counter-intuitive and there are so many layers of menues and submenues... just a horrible developer experience.

Any of the other No-/Low-Code tools will be faster to develop, cheaper (costs for plugins!) and easier to maintain (again, plugins and compatability issues).

Webflow is great if you are more on the development (not design) side of things since it's closest to writing code. It's also more powerful than Framer.

Framer is great if you have design experience with Figma and really want to reduce code to a minimum. Best for sites with little functionality.

NextJS is awesome for site speed and probably necessary for complex applications. Probably not needed if you managed to work with WordPress till now.

Thus, your line of argument should be centered around time, maintenance costs and to a lesser extent user experience. Maybe it's just me but I can almost immediately tell if a site is made with WordPress. It just looks old-fashioned.

Thank you Equinox for replying.

I really appreciate the suggestions for those tools. I'll check them out.

From the above comments, I can see the consensus is that Next isn't needed for these smaller projects.

I really appreciate the suggestions around the time, maintenance costs and the user experience. Like @Andreas Thiel pointed out, this may be difficult to get across. I may reflect on this some more.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Onion]

I'm in a similar position to yourself - several clients who are stuck with a monolithic web stack (one uses Wordpress and several use Shopify). I've suggested the same to them (the term you're looking for is "headless", with another being JAMStack) but there is little you can say that will persuade them to change.

Whilst they are likely amenable to new ideas, their primary concern will be getting stuck with a stack that a) they don't understand, b) will have to pay expensive developers to fix / upgrade, c) not be able to manage properly.

It's a phenomenon I call the "adoption cost", which is basically how "expensive" it is to switch to a solution beyond the monetary impact. In this case, the adoption cost is to move away from a "proven" solution to something that most don't understand and will, therefore, be incapable of integrating effectively into their business.

The only way around it is to develop the technology yourself and demonstrate that it's a better solution, whilst being able to interface with the current stacks the company currently uses. In order to do this, you need to lead with the one thing most people care about, which is money. You demonstrate how your solution can built a richer, more secure, powerful and effective web front-end whilst still benefitting from the admin backend provided by Wordpress/Shopify etc.

You can see part of an email I sent to one of my larger clients on the topic here: -

View attachment 53275

This company is at the point where they need to consolidate their marketing, technology and backend processes into a more substantive system that can be used to help accelerate the business's growth.

Adopting a headless design pattern would contribute to that by giving the company the opportunity to design an interface exactly how they need and (most importantly) push them to think systemically around both the technology and team aspects of its adoption.

They don't need "headless" on its own. It needs to form part of a wider solution which will "transform" their business into something more robust, substantive and effective. That's basically what they're looking for.

To do this, I have written a large document explaining how they need to fix various parts of their system to accommodate the growth. To be clear, these have NOT pursued this avenue presently, so I am not advocating you do this - really just showing what I've been doing to address the issue.

It's their choice as to what they end up doing: -

View attachment 53276

And, remember, the above is coming from a guy who gets paid to do Shopify + Wordpress code. You can see a child theme I created for another client I work with who primarily use Kadence here: -

View attachment 53277

This streamlines the development process of their WP sites by standardizing the various attributes of the WP backend, namely plugins, CPT's and some other aspects. The aim with it is to be able to replicate instances of Wordpress sites by using OCDI.

The basis of the theme is Redux, which allows us to create a large number of customization options, which can be managed through the backend code: -

View attachment 53278

View attachment 53280

A bit off-topic, but as you're a WP dev, one of the really cool things about the above is that it integrates natively into Kadence's "dynamic content" system. This means that we can use the above inputs inside the Kadence Gutenberg blocks: -

View attachment 53281

If you have further questions, I'd definitely be interested in talking. I have been working on an engine to provide the means to develop & deploy "headless" sites whilst interfacing with existing Shopify + Wordpress data.

Thank you rpeck for replying.

I always have questions, so I'll send you a PM. Thank you for the offer.

Thank you for reminding me about Jamstack. I've looked through For fast and secure sites | Jamstack a couple of times previously, just didn't think to mention it. There's lot of good options.

I haven't touched much Redux very much. However, in reading your response, I've watched some overview videos on it. That does seem helpful to be able to change a lot more options in the Customiser Page. I'll try implement it myself.

 

[Onion]

A startup I work at is possibly making the transition from WordPress to Next.js because there are a handful of our other websites using the React system and it makes sense to standardize when you have a user dashboard, logins, three or four different websites, etc. all under one brand. Using React components makes it easy to standardize styles and lower costs by having a team of engineers who can work on a single technology. This is like a one year+ process for a single business and with a handful of the best full stack engineers working on it.

With that said, it does not make any sense at all for a small web design agency to make the transition to Next.js. The needs that a small web dev agency fulfills are way different than what you'd do with Next.js.

On the business side of things, the companies that your agency supports might not even have the budget to increase their web spend - so when you think of stakeholders you need to sell this idea to, it's not only your boss, but all of the businesses you serve. On the development side of things, you don't need things like version control for a local business like restaurants, service businesses, franchises, etc. Most small businesses don't need to overengineer their website which is what a transition to Next.js would be.



Hosting is the easiest thing ever with WordPress. Managed hosts like WP Engine, Kinsta, Flywheel, etc provide a premium service that can be easily baked into retainer fees. You can also get into middle of the road hosts like SiteGround where you can throw 20 small websites on a $400/year plan. And if you prefer to do it more yourself, Cloudways w/ Digital Ocean or just straight up Digital Ocean are both options.



This is a workflow, mastery and tool issue. WordPress is not inherently difficult, in fact it's probably the easiest and most flexible way to quickly deliver websites at low cost.

Workflow - Something is probably lacking in terms of how websites are imagined, designed, and developed

Mastery - Any tool will get the job done, even Elementor, but without a full mastery things will get difficult

Tools - There are better solutions than Elementor available. Most of this plays into mastery, as in any tool will get the job done, but for me personally I find that YooTheme Pro and Bricks are the best currently available.



Headless is a very complex and niche solution that would have very little benefit for a generalized small web dev agency. Similar to OP's desire to transition presumably dozens of small business websites to pure Next.js, there's no reason to overengineer things. It will not transform their business into anything that it's not already.

---

If OP wants to work with modern stacks and is capable of using Next.js, they are underemployed working at a small web design agency using Elementor. The solution is more to start a side hustle building websites the way OP wants to build them and then transition that into their own agency—not rebuild someone else's company completely in a new technology.

Thank you Choate for replying. I'm really grateful for your detailed feedback.

As I've noticed with the other comments and others I've replied to, doing this for a small web development agency isn't the right option. And you're also right about the workflow, mastery, and tools sections. These are probably the root of my annoyance. I really want to master what I do, and I need honest feedback in what I'm doing wrong.

I suppose I could start my own agency. There is really nothing stopping me. I could begin by taking on some small projects on the side in tandem with my current job. Or a different system entirely (MJ's system post). Any advice on this area would be helpful.

I love brutal feedback; I want to master what I do. Let me know if you're willing to let me PM you with any more questions I have.

 

[rpeck90]

Thank you Choate for replying. I'm really grateful for your detailed feedback.

As I've noticed with the other comments and others I've replied to, doing this for a small web development agency isn't the right option. And you're also right about the workflow, mastery, and tools sections. These are probably the root of my annoyance. I really want to master what I do, and I need honest feedback in what I'm doing wrong.

I suppose I could start my own agency. There is really nothing stopping me. I could begin by taking on some small projects on the side in tandem with my current job. Or a different system entirely (MJ's system post). Any advice on this area would be helpful.

I love brutal feedback; I want to master what I do. Let me know if you're willing to let me PM you with any more questions I have.

I think both my and Choate's posts alluded to the idea you should be looking at developing a solution yourself.

Rather than trying to get the agency to agree to it, it would likely be more prudent for you to do the work and then showcase the benefits of what you have created. The agency's focus is on sales / profit and if your solution is able to help improve that, they may be interested in implementing. Wordpress etc are only used because they're mature, widely supported and relatively "safe". Custom solutions are not.

I shared the anecdotal stuff in my post because I wanted to give you an idea of how difficult it can be to win support for this sort of thing. Most people don't care about how you achieve results, so long as they are legitimate, replicable and advantageous to the business. Whether you use React, Next.js, Gatsby etc is irrelevant. If you can demonstrate how the solution is going to give them an extra $500k in revenue, do you think they'd be interested?

To respond directly to your assertions above, don't start your own agency. Focus on solving the problem you have identified. The problem is you are annoyed at the antiquated systems which are generally used by web agencies and you want to make use of the various modern tools. Think about ways you could attempt to solve this problem with the current flow you have (IE working at the agency).

Could you, perhaps, design a theme which can be "offloaded" from the main site (IE hosted separately), with a custom plugin used to define various options for it?

To give an example, the child theme with Redux I shared was my own invention. The company didn't know how to implement a solution, only that they wanted a particular result (the ability to replicate versions of Wordpress sites). It was up to me to come up with a viable way to achieve it. Perhaps you could pursue a similar path.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.