An SEO Consultant Reviewed Our CMS. Here's What Changed by Teatime

An email arrived on Tuesday morning from Mandy Cochrane, an SEO consultant and a friend of mine. She'd spent an hour in the demo of our CMS at my invitation and written up what she found: what she liked, what puzzled her, and a few things she thought were missing. By teatime, four of those were built, tested and released. I'm writing it up because it answers a question I get asked in a roundabout way all the time: what happens when the WordPress plugin doesn't do what you need?

A bit of background first. Mandy spent thirteen years working with another CMS platform day in, day out, and has recently been relearning WordPress, so she looks at an admin screen the way a mechanic looks at an engine. She wasn't reviewing it for a client. She was curious, and so was I.

Looking through her notes

The one I'll slow down on is the character counter.

Every page on a website has a title and a short description that Google shows in its results. Make them too long and Google trims them mid-sentence. The CMS tells you the recommended length next to each box, and Mandy wrote: "if you don't mind a suggestion, it would also be handy to include a character counter as the user types."

There was one. It had been there for months, in small coloured text after the recommendation, so it read as part of the note and she looked straight past it. I could have replied "it's there, look harder". But an SEO professional went looking for a counter and didn't find one, and that is the bug. It's now a coloured pill directly under the box, green, amber or red, showing the count against the target. Titles show their width in pixels too, because Google measures a title by width, so one full of wide letters gets trimmed sooner.

The second was sharper. The site settings have a box for a Google Analytics ID and another for a Google Tag Manager ID. Her question: "if they add both IDs, is there any risk of duplicate GA4 tracking?" There was. Tag Manager is a container that nearly always loads Analytics itself, so filling in both meant every visit was counted twice, and your visitor numbers read double with nothing looking wrong. Both boxes now warn you to use one route or the other, and the dashboard's site health card shows amber while both are set.

Third, hidden pages. Anyone who can edit a page can tick "Hide from search engines". Useful for a thank-you page, and a problem when it's ticked on the services page by mistake, because a hidden page looks perfectly normal on the site. Mandy liked that ordinary editors could do it, "as long as they understand how to use it safely!" The trouble was there was no list anywhere of which pages were hidden. There is now, with each one named, so someone can check they were all meant.

Fourth, the form builder. She expected the preview to update as she edited the fields, and found "it only updates once saved, which is slightly less exciting!" It has a Refresh preview button now. A click instead of a save.

By teatime

Each of those was small. None of them went on a list. By mid-afternoon all four were built, run through the CMS's automated tests (just over 3,700 of them) and released. Every site on the platform picks it up with its next routine update.

That turnaround is the case for building on something we own, and it's the bit I'd ask about if I were choosing a platform.

What happens when the plugin doesn't do what you need?

When you're sold a WordPress site, you're told there's a plugin for everything. That's roughly true, and you'll be running seven of them before anything specific to your business is built. What you're not told is what happens when the plugin is nearly right. The counter exists but you can't see it. The analytics loads twice and nothing warns you. The feature you want is on the developer's list, behind the features more customers asked for, because that's how roadmaps work. Your options are to wait, to pay someone to bolt a fix onto code they don't own, or to live with it.

Owning the platform changes the shape of that conversation. A suggestion made on a Tuesday morning was in the weekly update by the afternoon. No vendor to lobby, no release cycle to wait for, no "we've logged it". The person who wrote the CMS is the person you email, and I say that as the person you email. It's the argument the Motley CMS page makes in the abstract, and Tuesday was the first time I could point to a dated, specific example with a witness.

None of which makes WordPress the wrong choice. For some sites it's the right one, and I've written about how to tell. It does mean the question to put to any agency, whatever they build on, is not "is there a plugin for that?" but "what happens when the plugin isn't quite right, and who fixes it?"

The notes that didn't need a fix

Not everything she raised needed changing. She asked whether the CMS shrinks images on upload, because clients forget to. It does: every upload is resized to five widths and saved in lighter modern formats, the page serves whichever suits the visitor's screen, and the original is kept and never sent to anyone. She asked about syncing stock with eBay or Amazon. It doesn't do that, though the shop can push orders and stock changes out to other systems. And she was sceptical of her old platform's AI blog generator, "a pretty blunt tool" that "encouraged users to create sloppy content". I agree with her. The CMS has no post generator and no image generator, a few tools that suggest rather than publish, all of them off by default, and nothing goes live without a person looking at it.

Where her words ended up

Mandy gave me permission to quote her, and one line from her email is now on the CMS page, because it says what I've been trying to say about the platform better than I have:

"The thing that struck me more than anything is the intuitive usability of the CMS. Clear explanations on editable fields, and helpful tips make such a difference, and show just how geared you are towards your users."

A content management system is only as good as the afternoon someone in your office needs to change something, and the morning someone tells you what's wrong with it.

If you're choosing a platform for a new site and want to talk it through, get in touch. I'll say so if ours isn't the right fit.