How long does a CMS migration take?

Single-site moves that finished took from a few months to well over a year; multi-brand programs run for years, a few brands at a time, and most are still running. Upgrades inside the same vendor can take as long as a move. In every project where buyers compared the schedule with an original plan, it slipped — sometimes through overruns, more often through pauses or postponements. What makes CMS migrations slow, from the buyers who ran them.

Based on verified interviews with the digital, ecommerce, marketing and web leaders who plan and run CMS migrations, at DTC and enterprise brands. Buyers are anonymized before publication; vendor names and views are reported as given. No vendor paid to appear or could edit this page.

Ask buyers how long their CMS migration took and few can give a clean start-to-finish number: the project ran in phases, started before they joined, or is still going. What they describe instead is where the time goes. Content and templates must be assessed, transferred, transformed or rebuilt; companies with many sites often migrate in waves; and features the old platform provided may need to be recreated. Delays also arise before implementation, when budgets change, priorities shift or the platform and agency decisions remain unresolved. None of the projects where buyers compared the schedule with a plan was reported on schedule — the same pattern buyers describe for email, commerce and loyalty platforms — and more of those slips were pauses than overruns. The companion reads are why companies switch CMS platforms, which covers what happens after the move, and what headless CMS buyers wish they'd known, where the ratings live.

3–4 mo
The quickest finished move here: a lift and shift to a headless CMS, which the buyer says carried the old site's problems forward. The fast end, not the norm.
0 on schedule
Among projects with a planned date and a later status, none was reported on schedule. Some finished late; more were paused or pushed before finishing.
Years
Every global or multi-brand program here that gave an elapsed time is still running, after one to more than three years.
Elapsed time buyers describe for CMS projects Five single-site projects that finished took about three to four months (a lift and shift), five to six months (a rewrite onto a headless CMS), nine to twelve months (moving content bucket by bucket), about a year (a new site) and fifteen to eighteen months (a version change inside the same vendor). Four global or multi-brand programs were still running at the time of the interview, after about a year and a half, about two years, two to two and a half years, and more than three years. SINGLE SITE · FINISHED Lift and shift 3–4 mo Rewrite onto headless 5–6 mo, about twice the plan Content bucket by bucket 9–12 mo New site about a year Version change, same vendor 15–18 mo, planned at 6 GLOBAL OR MULTI-BRAND · STILL RUNNING Global content move 1–1.5 yrs, still going → Dozens of brands in waves about 2 yrs → Multi-country rollout 2–2.5 yrs → Global, multi-brand 3+ yrs, still going → 0 1 yr 2 yrs 3 yrs
Each row represents one company's reported project. Finished projects are measured to the buyer's stated launch or completion milestone — for the rewrite onto headless, the go-live, with page rewrites continuing after it — and approximate where the buyer gave an approximate time. Running programs show elapsed time through the interview, not their eventual duration, and a partial launch or completed wave is not counted as completion. Plans are excluded except where a buyer compared the outcome with one, and commerce replatforms that carried the site with them are excluded. Examples from these interviews, not typical durations.
You've read the average. How healthy is your stack?A free, peer-powered physical for your tech stack, with a shortlist and your next steps.
Book my stack physical →

Why CMS migrations take longer than planned

01

Content and templates take more work than expected

The content workload is the source of effort buyers describe most. Depending on the old and new architectures, teams may need to rebuild templates, transform content models, move pages and archives, or rewrite material that cannot be carried across directly; some content moves in batches, while other material needs manual attention. Buyers with large sites and deep archives describe this work extending the project substantially. One B2B distributor calls its move “a very labor-intensive lift in rewrite of all of our pages that are moving to Optimizely.” At a global electronics company, “the last, I would say probably year, year and a half has been the migration of content over to this.” A consumer brand on a long migration sees no quick end: “We are slowly migrating, but I don't see that another two or three years down the line we'll be migrated fully,” because “there are lots of pages, there's not one page or two page.” Publishers carry archives: “we need to bring on legacy articles and content from the 70s all the way until modern day”. What a rebuild involves is on the Contentful stack.

Named in this context: Optimizely · Contentful · Adobe Experience Manager

02

Many sites become waves, and waves become years

Companies with many brands, divisions or markets cannot move everything at once, so they move in waves and the program runs for years. One company “picked certain divisions within the company to be the ones that are launching first and then, because it's just we can't do it all at once.” A home-products manufacturer rebuilding its sites says “we've got a lot of sites, so it'll be a while.” At one global company, “It's been three years and that AEM site is on-prem,” with the larger regional sites still to come “in at least couple of years because they're big.” Commerce replatforms roll out market by market in the same way; those are on ecommerce replatforming timelines.

Named in this context: Adobe Experience Manager · Contentful · headless platforms

03

What the old platform did has to be rebuilt

Personalization, testing and recommendations that the old suite handled have to be rebuilt or relearned on the new one, and that work lands on the schedule. A financial services company moving off a suite expected to “learn the intricacies of doing things like personalization, things that we could do in Sitecore, because we've been using that system for years.” How long it takes to get back to where the old site was, after launch, is covered on why companies switch CMS platforms.

Named in this context: Sitecore · Adobe Experience Manager

04

Budgets and priorities pause the project

The most common shape of slip here is not a build that ran long but a project stopped before or during the build. A healthcare distributor says “We were supposed to migrate four websites on Sitecore this year, but it got pushed due to budget constraints, and we don't know when” — “So we're in a waiting pattern at the moment.” A food retailer had hoped to move that year, “but as of probably about a week or two ago, I was told to put all developments on pause while budgets were being reevaluated”. Elsewhere a migration “got delayed a little bit because of some internal changes that happened,” and only later “is starting to be reignited, let's say, and taken off pause”.

Named in this context: Sitecore · Liferay · Adobe Experience Manager

05

Platform and agency decisions delay the start

Some companies spend months or years evaluating a replacement before implementation begins. That extends the overall modernization program, though the selection period is distinct from the migration's build and cutover. One media company's editors complained for years before an RFP was run; another company hired an agency to choose its platform after months of evaluation; a hotel group's replatform moved from one year to the next while it changed agencies and reran its selection. As with loyalty programs, the clock starts long before the build.

Named in this context: Sitecore · Kentico · agencies

06

New hands: a new agency or a new architecture

A change of who builds the site, or how, sets the pace. A hotel group waiting on its new agency put it this way: “We're still on Sitecore, but when the new agency starts, I believe we're moving to a new platform, but I'm not sure what that is 100% yet.” Two buyers say a team new to headless delivery slowed them; one says the move to a headless architecture “has really impacted our ability to deliver quickly at scale”. What headless changes for a team is on what is a headless CMS.

Named in this context: Sitecore · headless platforms · agencies

How long does a CMS migration take?

These interviews do not establish a typical duration: few buyers give a full start-to-launch time, and the ones who do describe different projects. Among the single-site moves that finished, buyers describe anywhere from three or four months — a lift and shift that carried old problems forward — to well over a year for a full rewrite or a version change. The one multi-market program that finished took about a year for its home market and about another year for the rest. Every global or multi-brand program that gave an elapsed time was still running, after one to more than three years. Single-site moves sit in about the same range as moving a single-brand online store, which is on ecommerce replatforming timelines. Read the figures as examples, not benchmarks.

Is a CMS upgrade easier than switching?

Not necessarily. Version changes inside the same vendor run as migration-sized projects in these interviews, because the site has to be rebuilt for the new version: a move from one generation of Sitecore to its headless version, planned at six months, took about fifteen months to the interview and about eighteen to finish, and a Kentico version upgrade took about a year of dedicated staff time to move the content. A third buyer says a Sitecore update that skips several years of versions can take many months. The switching page notes that an end of support tends to keep companies with the same vendor; staying is not always the cheap option it looks.

Do CMS migrations stay on schedule?

Not in the cases with usable timing comparisons here. Seven projects have both an identifiable original plan and a reported outcome or later schedule status, and none was reported to be on schedule — the same pattern buyers describe for email, commerce and loyalty platforms. Three finished or launched later than planned: one rebuild took about twice its plan, the version change above took about three times its plan, and a new website launched six or seven months after the date the team had expected. The other four were paused or pushed back before or during implementation — twice for budget, once for internal changes, and once while the company changed agencies and reran its selection — and had not necessarily finished by the interview. This small, selectively reported group cannot establish how often CMS migrations miss their dates overall. A slip has knock-on costs: one team bought a testing tool for its new site and could not use it until the site launched.

Due for a stack physical?

Book mine →

The stories behind the timelines

A healthcare services company spent much of a year waiting on its own website. Among its priorities were “relaunching our website, some brand stuff, standing up like a voice of the customer function.” The team bought a testing product ahead of the launch — “we sort of prematurely bought the Optimize product because we wanted to do A/B testing with the new website” — and then waited: “The website was very delayed in launching, so it was like months after we, you know, maybe six or seven months after we anticipated we would be able to start using it.” By the time it went live, “we have a lot of unused opportunity wasted.” Director, lifecycle and product marketing · healthcare services company
A media company with dozens of brand sites is moving them a few at a time. The decision followed years of complaints from editors about “the difficulties that they had using the platform over a two to three year period.” The new platform is going in gradually: “Keep in mind, we're not pulling the plug on all of them right away.” Instead, “We are moving three to five brands at a time.” Each wave is a checkpoint: “what we're going to do is we're going to evaluate our performances, see how we're doing, and then if the other ones do make sense to change, then we'll do so.” Digital executive · media company
A consumer products company moved its websites home market first. “We phased it in, and the whole process took us about probably a year to phase our US web operations and get all the design and everything done.” The other markets followed: “And then over the next year, we brought each of those markets.” The company has since used the same pattern elsewhere in its stack — “We did the same thing when we adopted Sitecore.” Senior marketing director · consumer products company

The counter-current: the moves that went quickly

The quick or smooth moves here share a shape. A lift and shift, with the site moved largely as it was, finished in “three months, or three to four months” — with the old site's problems carried across, as the switching page records. Phasing by content type inside a group went well: one brand has been “migrating onto Amplience throughout this year, just kind of content bucket by content bucket,” a move a later interview confirmed was complete, and finds “It's a whole lot less buggy than AEM was.” A pilot can de-risk a global program — “We had run a pilot for six months, and that was a very successful pilot.” And small sites can now take weeks: one manufacturer has “developed two to three new websites in a matter of weeks versus going out to a developer and taking months.” Nothing quick here involved rewriting every page onto a new architecture.

What this means for your migration plan

In these interviews the length of a CMS migration is set by the content that has to move, the number of sites, and whether the project keeps its budget and its team, more than by the platform. Our recommendations, not practices buyers describe: count the pages, templates and archives and decide what moves in batches, what needs rewriting and what can be left behind before setting a date, since content is where buyers say the time goes; decide which features of the old platform — personalization, testing, recommendations — must work on day one and plan their rebuild as its own track; if you have many sites, plan the waves and the period of running two platforms, and pilot the first wave before committing the rest; secure the budget for the whole program, since pauses caused more of the slips here than overruns; and treat a major version upgrade as a migration, with the same planning. For what to expect after launch, see why companies switch CMS platforms.

Common questions

How long does a CMS migration take?

These interviews do not establish a typical duration. Among single-site moves that finished, buyers describe anywhere from three or four months — a lift and shift that carried old problems forward — to well over a year for a full rewrite or a version change. Multi-brand and global programs run for years, a few brands or content types at a time, and most of the ones here are still running.

Why do CMS migrations take longer than planned?

Buyers most often describe the work of moving, transforming or rewriting content and templates, many sites turning into waves of migration, features the old platform had having to be rebuilt or relearned, budgets and priorities pausing the project, and platform and agency decisions that delay the start. Among projects where buyers compared the schedule with a plan, none was reported on schedule — some finished late, and more were paused or pushed before finishing — though this small group cannot show how often migrations miss their dates overall.

Is a CMS upgrade easier than switching platforms?

Not necessarily. Buyers describe version changes inside the same vendor that took about a year or longer, one of them planned at six months, because the site had to be rewritten for the new version. An upgrade can be a migration-sized project.

Should you migrate all your websites at once?

Buyers with many sites in these interviews move a few brands, divisions or content types at a time and review the results between waves, and the phased moves here are among the ones that went smoothly. Phasing also stretches the program over years, so the old and new platforms run side by side for a long time.

Does moving to a headless CMS slow the migration?

Two buyers in these interviews say a team new to headless delivery set the pace, along with rewriting every page. Most of the timed projects here moved to headless platforms, so there are too few others to compare, and these interviews do not show whether headless moves take longer.

How long does a CMS migration take compared with an ecommerce replatform?

In these interviews, single-site CMS moves that finished took about as long as moving a single-brand online store, from a few months to about a year, and both run for years when many brands or markets move in waves. Neither category establishes a typical duration from the examples buyers give.

A physical for your tech stack

Peer-powered: built from hundreds of verified buyer interviews, and fitted to the tools you already run. You get a shortlist for your stack and your next steps.

Book my stack physical

Free · no password · your answers are anonymized before they inform research like this.

Want to come back to this?We'll email you a link to this page, so it's easy to find later or forward to your team.
Methodology. Alium conducts verified interviews with software buyers — the digital, ecommerce, marketing and web leaders who select and run these platforms. This page aggregates what buyers in that corpus, through October 2026, say about the time it takes to move to, build on or upgrade a content management system. Projects are confirmed in the buyer's own words and counted once per company; each is coded as completed, underway, planned, pushed or reversed, and plans are kept apart from what happened. Commerce replatforms that carried the website with them are excluded and covered on the ecommerce replatforming page. Few buyers give a full start-to-launch time, so durations are reported as examples with their population and status rather than as typical ranges; counts are of companies, and a project can have several causes of delay. Ratings for CMS platforms are on the headless CMS lessons page rather than here. Buyer identities are verified at interview time and anonymized before publication; vendor names are reported as given. No vendor paid to appear or was able to edit this page.