Two-thirds of large technology programmes miss their targets on time, budget, or scope. That's BCG's number, from 2024. And globally, something like $2.3 trillion gets wasted on digital transformation that doesn't deliver. Trillion, with a T.
So here's the question that bothers me: if the failure rate is that high, and the money being burned is that staggering, why do most organisations still choose their digital experience platform based on a two-hour demo and a feature comparison spreadsheet?
I think I know the answer. Because the demo feels like evidence. You've seen the platform do the thing. You've watched the vendor's solution architect click through a beautiful interface, pull up personalisation rules, show you content scheduling, drag-and-drop page building, analytics dashboards. It all works. Of course it works - it's their demo environment, configured by their own engineers, running on their own infrastructure, with unlimited time and budget behind it.
What you haven't seen is what happens six months into implementation, when your team of three is trying to replicate what that solution architect built, your Salesforce integration is half-finished, and the partner you hired is learning the platform alongside you.
I've seen that film. More than once. It doesn't end well.
Full disclosure before we go further: Distinction is a Kentico partner. We also work with Umbraco and Payload. We don't have formal partnerships with Sitecore, Optimizely, or WordPress VIP - which, honestly, gives us a reasonably independent view of those platforms. I'm going to be as honest as I can about all five, including the one we're partnered with. If this comparison feels rigged, close the tab. I mean that.
Every mature DXP can do roughly the same things. Personalisation? Yes. Multi-channel content delivery? Yes. Workflow approvals? Yes. A/B testing? Yes. Marketing automation integration? Yes.
The differences between platforms aren't in the feature list. They're in how those features get implemented, who needs to maintain them, what they cost over five years rather than one, and whether the partner ecosystem around that platform can actually deliver for a firm your size.
About eighteen months ago, I sat in a selection meeting at a mid-market professional services firm - roughly 200 people. The evaluation team had built an enormous spreadsheet. Forty-seven features across four platforms, each scored one to five. Kentico and Sitecore came out almost identical on the grid. But the TCO difference over three years was going to be somewhere in the region of £300k-£400k. That didn't appear on the spreadsheet anywhere.
Nobody had modelled it. Nobody had asked.
The buyers who make the best platform decisions have stopped asking "does it have this feature?" and started asking "can our team and our partner actually deliver this feature, and what will it cost to maintain over the life of the platform?" That shift in question changes everything.
After twenty-plus years of watching platform decisions play out - some brilliantly, some disastrously - we've landed on five factors that consistently predict whether an implementation will succeed in practice. Not in the demo. In practice.
Total cost of ownership over three to five years. Not year-one licence cost. Total cost. That means implementation, licensing, hosting, maintenance, internal resource to support it, and the partner hours you'll need for ongoing development. The platforms with the lowest licence fees sometimes have the highest total cost because they require more development effort, more plugins, more custom work. You need to model this honestly over a realistic horizon - because you're going to live with this decision for five to ten years.
Team fit. This one gets overlooked constantly. Is the platform's development model aligned with the capability of your internal team or the partner you can realistically afford? A platform that requires senior full-stack developers to maintain is a very different proposition depending on whether you have those people. If your internal team is two marketers and a part-time IT manager, that changes the equation dramatically compared to a firm with a six-person dev team.
Integration realities. Does the platform's integration architecture actually work with your existing technology stack? Or does it require significant middleware, custom connectors, or - worst case - ripping out and replacing adjacent systems? I've seen implementations where the platform itself was delivered on time and on budget, but the integration work doubled the total programme cost. One project I was close to - a financial services firm, mid-sized, not a client of ours - ended up spending more on connecting their new DXP to their CRM than they'd spent on the platform itself. The demo doesn't show you that.
Vendor stability. Is the vendor financially viable? Are they actively investing in the product? And - this is the one people forget to ask - are they genuinely committed to the mid-market segment, or is it a secondary concern behind their enterprise clients? Because if you're a £20m professional services firm and your DXP vendor's roadmap is being driven by Fortune 500 requirements, your priorities will never be their priorities.
Implementation partner quality. This is the factor that gets systematically underweighted in almost every platform evaluation I've seen. A great platform badly implemented will underperform a good platform well implemented. Every single time. I'll come back to this one.
Right. Let me walk through the five platforms that come up most often in mid-market DXP evaluations.
Sitecore is a powerful, enterprise-grade DXP with deep personalisation and marketing automation capabilities. It's also, bluntly, overengineered for most mid-market firms. The licensing model has shifted to a composable, cloud-based approach - Sitecore XM Cloud, plus various add-on products - which gives flexibility but makes cost prediction genuinely difficult. Implementation typically requires specialist Sitecore developers who command premium rates. If you're a 500-person financial services firm with a dedicated digital team and complex personalisation requirements, Sitecore might be worth the investment. If you're a 150-person consulting firm, you'll likely be paying for capability you'll never use. TCO over three years tends to be the highest of the five platforms here. The partner ecosystem is mature but weighted toward large agencies - finding a Sitecore partner who genuinely understands mid-market budgets and constraints is harder than it should be.
Optimizely (formerly Episerver) sits in similar territory - strong enterprise pedigree, robust experimentation and personalisation tools, and a pricing model that can escalate quickly. Their recent acquisitions have broadened the platform significantly, which is a strength if you want an integrated suite and a headache if you don't. Like Sitecore, the development talent pool is specialised and relatively expensive. And here's the thing about Optimizely that I don't think gets said enough: the experimentation heritage is genuinely differentiated, but most mid-market firms don't run the kind of structured testing programmes that justify it. I've watched mid-market firms choose Optimizely, use maybe 30% of its capability, and spend three years feeling like they're underutilising their investment. That's not a great position to be in - you've paid enterprise money for a platform you're running like a basic CMS.
Umbraco is open-source, built on .NET, and offers probably the most flexibility of any platform on this list. Licensing costs are low - free for the open-source version, modest for Umbraco Cloud. The developer community is large and active. For firms with strong internal .NET development capability or access to a good Umbraco partner, it can deliver exceptional value at a fraction of the cost of Sitecore or Optimizely. The trade-off is that Umbraco is a CMS, not a DXP. Out of the box, you don't get the marketing automation, personalisation, or built-in analytics that the enterprise platforms include - you build or integrate those capabilities separately. For some firms, that modular approach is exactly right. For others, it creates integration complexity that erodes the initial cost advantage. We work with Umbraco and rate it highly for the right use cases. But "right use cases" is doing a lot of work in that sentence.
WordPress VIP is the enterprise tier of the world's most widely used CMS. If your primary need is content publishing at volume - serious volume, with multiple editors, complex workflows, and global distribution - it's very strong. The talent pool is enormous, which keeps development costs competitive. But WordPress VIP is fundamentally a content platform. The DXP capabilities that Sitecore, Optimizely, and Kentico offer natively - personalisation, integrated analytics, marketing automation - require plugins or third-party integrations in WordPress. That plugin ecosystem is both WordPress's greatest strength and its most significant risk. I've written separately about the fragility of digital foundations, and WordPress plugin dependency comes up constantly in those conversations. Plugins get abandoned. Security patches don't land. One client we assessed had 47 active plugins, six of which hadn't been updated in over two years. Nobody had noticed. For a mid-market B2B service firm that needs structured content delivery, CRM integration, and progressive lead nurturing, WordPress VIP can work - but you'll be assembling those capabilities from multiple sources rather than getting them from a single platform.
Kentico - specifically Xperience by Kentico - sits in what I'd describe as the mid-market sweet spot, though I'm conscious of my bias here. It's a unified DXP built on .NET: content management, marketing automation, personalisation, and analytics in a single platform with a single licensing model. The TCO is typically significantly lower than Sitecore or Optimizely over a three-to-five year horizon, partly because of the licensing structure and partly because the implementation effort tends to be more contained. The platform is genuinely designed for the mid-market - Kentico's product roadmap reflects mid-market needs rather than being driven by enterprise feature requests. The trade-off? The partner ecosystem is smaller than Sitecore's or WordPress's. The talent pool of experienced Kentico developers is narrower, which means your choice of implementation partner matters even more. And if your requirements are genuinely enterprise-scale - complex multi-brand architectures, deep experimentation programmes, global content distribution at massive scale - you may find the platform's ceiling before Sitecore or Optimizely's.
If you want a deeper look at Kentico's positioning specifically, there's a companion piece on why Xperience by Kentico is a smart choice for mid-market firms.
Here's the thing that drives me slightly nuts about platform evaluations: firms will spend three months rigorously assessing four DXP vendors, scoring them across dozens of criteria, running demos, checking references, negotiating contracts - and then spend about two weeks choosing the implementation partner.
That's backwards.
Only 25% of organisations say their platform migration delivered expected value within the first year. And 60% attribute failure to the absence of an end-to-end plan. Those aren't platform failures. Those are implementation and planning failures.
I had a conversation last year with a COO at a professional services firm who'd just come out of a failed Sitecore implementation. The platform was fine. The partner had done maybe eight Sitecore projects before, which sounds like enough until you realise that the integration with their CRM was unlike anything the partner had done before, and they'd quoted for it anyway. Eighteen months in, the integration still wasn't working properly, the content team couldn't publish without raising a ticket, and the board was asking why they'd spent £600k on a website that performed worse than the one it replaced. The platform wasn't the problem.
The implementation partner determines whether the platform's capabilities actually get delivered. Whether the CRM integration works reliably. Whether your content team can publish independently. Whether the site performs under real traffic, not demo traffic. Whether the project lands on time and on budget, or joins the two-thirds that don't.
When you're evaluating partners, apply the same rigour you'd apply to the platform itself. Technical capability in the specific platform - not "we've done a couple of projects" but genuine depth. Sector experience that means they understand your regulatory environment, your content model, your buyer journey. A delivery track record you can verify with references. And commercial alignment - meaning their business model doesn't depend on the project running over or on locking you into perpetual dependency.
A platform like Kentico, which requires competent .NET implementation to deliver well, will perform very differently in the hands of a specialist partner versus a generalist agency that's used the platform twice. The same is true of every platform on this list. Sitecore implemented by a top-tier Sitecore partner is a different product from Sitecore implemented by a web agency that added it to their capabilities page last year.
So, practically, what should you do differently?
Stop leading with the feature grid. Build your evaluation around the five factors I've described: TCO, team fit, integration realities, vendor stability, and partner quality. Weight them based on your specific situation. If you have a strong internal dev team, team fit matters less. If you're running a complex technology stack with multiple integrations, integration realities should be weighted heavily. If you're a 100-person firm with a limited budget, TCO might be the dominant factor.
Ask better questions. Instead of "does the platform support personalisation?", ask "what level of internal resource is required to configure and maintain personalisation rules, and what does that cost annually?" Instead of "can it integrate with Salesforce?", ask "show me a reference implementation of a Salesforce integration at our scale, and tell me what went wrong during that project." That last bit - what went wrong - is the question most people don't ask. It's also the most revealing.
Evaluate the implementation partner with equal rigour. Ask them about projects that went badly. Ask what they'd do differently. Ask how they handle scope changes. Ask for references from firms your size, not their biggest logo. If they can't give you specific examples from your sector at your scale, that tells you something important.
If your evaluation has surfaced an architecture question - headless vs hybrid, composable vs monolithic - there's a separate piece on headless or hybrid worth reading before you finalise. And if you've arrived at this comparison without the broader strategic context for why the CMS decision matters in the first place, I'd suggest reading why your next CMS decision matters more than you think. For legal sector readers specifically, there's a sector-specific framework on future-proofing your law firm's technology stack.
We've done a full RFP. We've seen demos from four vendors. We know what each platform can do. We're ready to decide.
I hear this a lot. And I get it - you've invested significant time and energy, and you want to move forward. But knowing what each platform can do is only a fraction of what you need to know. Can your team operate it? Can your partner deliver it? Can you afford it over five years, not just one? Is the vendor going to be around and investing in the product in 2030?
These aren't abstract questions. They're the questions that determine whether you'll be writing a case study about a successful implementation or commissioning a rescue programme eighteen months from now.
If you want a structured way to apply these five criteria to your own evaluation - across any platform shortlist - we've put together a DXP evaluation scorecard that makes the comparison rigorous rather than impressionistic. It's genuinely vendor-neutral, and it should produce a useful output whether you end up choosing Kentico, Umbraco, Sitecore, or something else entirely.
The platform decision is a five-to-ten year commitment. Give it the evaluation it deserves - not the one the demo encourages.