Century Tech Alternative UK: A Post-16 Buyer's Guide

CENTURY sells mainly to schools and multi-academy trusts, so many people searching for a Century Tech alternative work in one, and for them the like-for-like options are other school practice and homework platforms. We have already compared those in our Sparx Learning alternative guide, which lays out Sparx, MathsWatch, Dr Frost Maths, Third Space Learning and CENTURY side by side. This guide is written for a different reader: the Head of Learning and Development, the COO or the curriculum lead at a further education college or an independent training provider, where CENTURY's FE product offers Functional Skills, GCSE resit content and initial assessment.

That buyer faces a sharper question than a school does. CENTURY is a mature platform with a large in-house library, a published evidence base and post-16 tooling for GCSE and Functional Skills. Dilr Academy does a different job: it builds a course from your own curriculum and then teaches it one to one. Choosing between them is less about which is better and more about which problem you are paying to solve, which is why we start from the job rather than the feature list.

Scope, stated plainly. This post covers what CENTURY is, how it decides what a learner studies next, how to read its published impact studies, where Dilr Academy differs, and how a provider should pilot. It cedes the mechanics of knowledge tracing to our guide to adaptive learning AI and the UK curriculum, the supplier safeguarding checklist to our AI tutor safeguarding checklist, and the wider market map to the AI teacher buyer's guide.

This guide is shipped by the team behind Dilr Academy, an AI tutor that builds interactive, multilingual courses from an institution's own curriculum and teaches each learner one to one. Or see DATS, our five-stage method for placing AI inside an organisation and running it.

What is CENTURY Tech, and who is it built for?

CENTURY Tech is a London edtech company whose platform combines assessment, short personalised lessons called nuggets and teacher dashboards, mainly across English, maths and science. CENTURY sells versions for primary and secondary schools, multi-academy trusts, further education colleges, apprenticeship and training providers, tutoring businesses and home learners, and says its AI recommends which nugget each learner should study next.

The company trades as Century-Tech Limited, registered in England and Wales under company number 08482934, and its home page says it is trusted by primary, secondary and post-16 providers in over 70 countries. For a college buyer, the relevant product is the one described on its FE colleges page, which calls CENTURY an AI-powered initial assessment, teaching and learning platform for FE colleges. That page is specific about scope: GCSE courses covering the major specifications for England, Northern Ireland and Wales; Functional and Essential Skills courses from Entry 1 to Level 2; and Digital Skills courses covering the Essential Digital Skills Qualification, which the page says are available for English and Welsh specifications.

Three features matter most for post-16 provision. The first is Smart IA, an adaptive initial assessment for English and maths Functional Skills from Entry 1 to Level 2, with an AI-marked writing task and results visible to staff only. The second is the library itself: CENTURY says every nugget contains videos, slideshows and self-marking questions, and that all materials are created in-house by experienced teachers. The third is the data layer, with a Leadership Dashboard, weekly emails to teachers, benchmarking reports against its community of colleges, and a Connect API that syncs users with a college's MIS or ePortfolio and draws results back.

That combination is why, in our view, CENTURY is hard to displace in a college. It is not just a tutor. It is an assessment front door, a content library and a reporting feed, and a replacement has to be weighed against all three, so a replacement plan should cover all three; our AI solutions page lists where DILR builds.

Why do colleges and training providers look for a CENTURY alternative?

A college or training provider usually looks past CENTURY Tech when the course it needs is not in CENTURY's library. CENTURY's strength is a teacher-written catalogue centred on English, maths, science and digital skills. A provider teaching a vocational, professional or multilingual programme, or wanting its own syllabus taught rather than practised, may find that work sits outside the library.

The gap is structural, not a quality judgement. A training provider running an accountancy apprenticeship, a care certificate, a software bootcamp or an English for speakers of other languages programme is likely to hold curriculum outside a core-subject library, and its own tutors already hold that syllabus. The question then becomes whether any AI tool can teach that syllabus well, which each buyer has to test on its own content, not whether a platform can recommend practice in the subjects it already covers.

A second trigger is the shape of learning. CENTURY's model is a sequence of short nuggets followed by questions, which suits practice and revision. Some programmes need a longer arc: a learner who has to reason through a case, explain a concept back, or work a simulation before moving on. That is closer to what we describe in our explainer on the Socratic AI tutor, where the tutor asks before it tells.

The third trigger is language. CENTURY's FE page describes its courses against English and Welsh qualification specifications and does not describe teaching in other languages. A provider whose learners study in another language, or who teaches international cohorts, needs to know whether a whole course can be rebuilt in that language rather than a menu relabelled. None of these triggers mean CENTURY is failing at its job. They mean the provider has a second job, and the decision is whether to add a tool, replace one, or leave both alone. If you are not sure which of those you face, an AI placement diagnostic is built to answer exactly that question before any licence is signed.

How does CENTURY decide what a learner studies next?

CENTURY Tech starts each learner with diagnostic assessments that identify strengths and gaps, then serves a Recommended Path of nuggets based on what the learner should study next. Teachers can override it by setting assignments, and a memory boost function brings previously completed nuggets back into the path to support retention, so AI and teacher choices run side by side.

The mechanics come from CENTURY's own descriptions. Its account of the Liverpool City Region pilot, published in March 2025, explains that the diagnostics kick-start the AI and that the Recommended Path then suggests nuggets, while teachers use the Markbook in the Teacher Dashboard to spot class-wide misconceptions and set assignments. Its guidance on revision strategies describes the memory boost function that returns completed nuggets to a learner's personal pathway.

For how the recommendations are computed, the most detailed public source is older. A June 2019 post on how CENTURY's AI works, written by a CENTURY data scientist, describes recommender systems trained on student behaviour, clustering to identify the guidance different student groups need, and natural language processing that feeds its predictive modelling. It also says the platform began with a beta used by 20 schools in a closed study for two years. Treat that as a description of the design intent at the time, not a specification of the current system, because seven years is a long time in this field.

CENTURY has also been explicit about where generative AI sits in its content process. A November 2025 post on how CENTURY crafts its content says generative AI now extracts exam specification points into a table, while the vast majority of questions are written by in-house subject specialists and every incorrect answer carries targeted feedback. CENTURY describes generative AI as able to suggest additional questions, "always as a support, never a replacement". That is a sensible position, and it is worth noting what it implies: CENTURY's AI chooses and sequences teacher-written content. It does not write a new course for you. For the general theory of how an engine estimates mastery, our adaptive learning guide covers knowledge tracing in detail, so we do not repeat it here.

What do CENTURY's published impact studies actually show?

CENTURY Tech's published studies report that learners who use CENTURY more tend to make more progress than learners who use it less, across several settings. CENTURY published all of them, the newest is explicitly an association study, and none describes a randomised design. A buyer should read them as evidence that engagement and progress move together, not as proof of how much CENTURY itself causes.

The newest and most carefully described is the international schools impact study, published on 21 October 2026. It analysed more than 3,300 student-test pairs from Inspired Education Group schools (1,741 in maths across eight schools, 1,571 in English across seven) over four academic years, from 2022/2023 to 2025/2026, against an independent, external progress test that CENTURY did not create, administer or score. CENTURY's headline is that the most engaged students were up to 58% more likely to exceed expected progress in maths and 29% more likely in English. The full technical report is clear that these are relative differences: in maths, on the coverage measure, 37.7% of the highest-engagement third exceeded expected progress against 23.7% of the lowest third, a gap of 14 percentage points.

The methodology section deserves credit and close reading. It compares students within the same school using a linear probability model with school fixed effects, adjusts for gender, and splits students into thirds within each school, showing only the lowest and highest. It also excludes two groups: students with no CENTURY activity between their two tests, and students at the very highest and lowest prior attainment. Both exclusions are defensible for the question CENTURY asked, which was whether more active use is associated with more progress. Neither answers the question a buyer cares about, which is what would have happened to the same students without CENTURY. A motivated learner tends to do more of everything, including more nuggets, so some of the gap may be motivation showing up twice. Separating the tool from the learner is the question we would put to any AI result, and our approach to placing AI explains how we work with institutions.

The older studies follow the same pattern. CENTURY's Brighton Hill analysis, published in June 2025, reports a strong correlation between regular use and higher GCSE grades, with top users averaging 25 more Attainment 8 points. The Liverpool pilot's headline 10% gain is the average improvement between a learner's first and second attempt at the same nugget, which measures learning within the platform rather than attainment, and the approximately 1,800 teacher hours freed by over 1,200 assignments is CENTURY's own figure. The college evidence is the oldest: a Ufi VocTech project in three colleges during 2018/19, published in February 2020, compared resit improvements with the national average, without describing a comparison group.

The government's own standard sets the bar a buyer should apply to every vendor, us included, and it is a useful bar for any AI an institution brings in, whether bought directly or through DATS. The Department for Education's generative AI product safety standards say: "Suppliers should not exaggerate the impact or capabilities of their tools. Any claims should be supported by robust and transparent evidence." CENTURY's Inspired report is more transparent than most, because it publishes its method and its exclusions. The independent comparison point is sobering. A two-year randomised trial of Khanmigo in 18 Tennessee middle schools, published as NBER Working Paper 35620 in 2026, found gains of 1.3 national percentile ranks per term, similar to Khan Academy practice without AI, and the authors wrote that the binding constraint appears to be engagement: students tried the tutor but rarely used it for substantive dialogue. That is a different product, but the lesson transfers: a platform only helps the learners who use it, and a correlation with use is where evaluation starts, not where it ends.

How does Dilr Academy differ from CENTURY?

Dilr Academy differs from CENTURY Tech in where the course comes from. CENTURY sequences a teacher-written library of short lessons and questions. Dilr Academy builds a course from the institution's own curriculum, with live diagrams and simulations, then teaches each learner one to one, asking a question before it explains and tracking what the learner has mastered.

The table below compares stated design, taken from each vendor's own pages, not measured outcomes. Dilr Academy has not published learner attainment figures, so there is no outcome row to fill on our side.

Three points deserve a plain reading, and our Dilr Academy explainer covers the rest of the product. On initial assessment and MIS integration, CENTURY has mature, published tooling and the live Dilr Academy page does not describe an equivalent, so a college that depends on Smart IA or on the Connect API feed should not assume Academy replaces either. On content, the difference runs the other way. Academy's live page says the longest course it has built follows the England KS3 programme of study for Year 9 Mathematics, with others ranging from Class 11 CBSE Accountancy to Python, and it shows a course page with stated outcomes, then modules and topics.

The teaching model is where the two products feel most different in use, and it is the loop our Socratic AI tutor guide, linked earlier, shows a buyer how to test. Academy's tutor opens a topic with something to look at, drawn live with Mermaid or D3, or a step-by-step storyboard animation. It then poses a guided question in voice or text and waits for the learner to reason, fills the gap with a worked example or a Deep Dive simulation when an answer shows one, and moves on only when the idea has landed. The interactive pieces are checked templates rendered in a sandbox, and each institution's learners and progress are kept apart from every other institution's, which is how the page answers whether it is safe for students to learn from an AI tutor.

Ownership questions like these sit in our AI operating model consulting, which sets who owns an AI tool, who checks its output and how it is retired once it stops paying back.

What do the condition of funding and KCSIE mean for this choice?

The 16 to 19 maths and English condition of funding binds the college or provider, not the software vendor. Students on 16 to 19 study programmes without a grade 4 in GCSE maths or English must keep studying those subjects, and the rules decide which qualification they take, so any platform is judged on whether it supports that route.

The detail is in the Department for Education's maths and English condition of funding guidance, which now carries a version for the 2026 to 2027 academic year. It clarifies that students with a GCSE grade 3 must study towards a GCSE, while students with a grade 2 or below may study towards either a GCSE or Functional Skills level 2, and that non-compliance affects the institution's 16 to 19 funding allocation. CENTURY's FE page does not mention the condition of funding, but in our reading its content maps onto it: GCSE content across the major specifications, Functional Skills from Entry 1 to Level 2, and an initial assessment that places a learner on a level.

For that cohort, our view is that keeping a specialist platform is the safer course. Dilr Academy has built courses to the England KS3 programme of study and to the CBSE board, but a provider would be building and validating a resit course where CENTURY already ships GCSE and Functional Skills content. The case for Academy is strongest in the programmes around the condition of funding rather than inside it: the vocational and professional content where no off-the-shelf library exists.

Safeguarding duties also sit with the institution. Keeping children safe in education is statutory guidance for schools and colleges, and the September 2026 version has been in force since 1 September 2026. The DfE product safety standards, by contrast, say they are mainly intended for edtech developers and suppliers to schools and colleges, which is why a college should ask every shortlisted supplier, CENTURY and Dilr Academy alike, to show how its product meets them. Our AI tutor supplier checklist sets out what to ask for, and much of its supplier-evidence list is still useful to an adult training provider where KCSIE does not apply.

What is the best Century Tech alternative for UK colleges in 2026?

There is no single best Century Tech alternative for UK colleges in 2026, because the right choice depends on the job. For Functional Skills, GCSE resits and initial assessment, CENTURY already ships the content and tooling, so it is a strong incumbent. For a provider's own vocational, professional or multilingual courses, an AI tutor that builds from that curriculum, such as Dilr Academy, is the closer fit.

We would judge any shortlist on five criteria:

1. Whose curriculum does it teach? A fixed library is faster to start; a course built from your syllabus fits programmes no library covers.
2. Does it teach or only practise? Practice and retrieval suit revision; a reasoning loop that asks before it tells suits new material.
3. What evidence does the supplier publish, and of what kind? Ask whether it is an association study or a trial, who ran it, and what was excluded.
4. What does it plug into? Initial assessment, MIS sync and funding reports matter more in a college than any lesson feature.
5. Which languages does the whole course run in? Menus are not enough for an ESOL or international cohort.

The concessions, stated plainly. A college whose AI spend is mainly about the condition of funding cohort should stay with CENTURY, or compare it with another resit specialist, rather than switch to a course builder. A school maths department wanting homework should read our Sparx comparison or our Third Space Learning comparison instead. A US learner studying alone is better served by Khanmigo, which we cover in the Khanmigo alternative guide. Dilr Academy is the right shortlist entry for a college or provider that owns a curriculum outside the core-subject library and wants it taught one to one, in the learner's language.

How should a provider pilot an alternative safely?

A provider should pilot a CENTURY Tech alternative on one cohort and one course that CENTURY does not already cover, so the two tools are not competing for the same learners. Name the cohort, agree the evidence you will accept before the pilot starts, run one course for a fixed period, then compare results against a matched group.

Choosing the cohort is the decision that protects the pilot. Pick a programme where CENTURY has no content, such as a vocational unit or a professional course, so you are not asking learners to split their attention between two tools in the same subject, and so a weak result does not damage condition of funding outcomes.

Reading the evidence, in a pilot, means agreeing it in advance, and the Inspired report's method shows why. Decide before the pilot how you will measure progress, which learners count, and what you will compare against. If the comparison group is simply the learners who used the tool least, you will reproduce the association problem described above. A matched group on the same programme, taught the usual way, is a better baseline.

Piloting one course starts with the curriculum. The live page asks providers to send their curriculum and says the first call shows Academy teaching one of your own topics rather than a demo course, which is a reasonable first gate: if what you see does not match how your tutors teach the unit, stop there. Comparing the results should rest on measures the provider collects itself, such as completion, assessed work and tutor time, rather than on any vendor dashboard. If the pilot works, the governance questions come next, and an AI execution office is one way to run that once the tool is live. If you would rather talk the pilot design through first, contact our team.

Frequently asked questions

Can Dilr Academy replace CENTURY's initial assessment?

Dilr Academy should not be treated as a replacement for CENTURY's Smart IA initial assessment. CENTURY publishes an adaptive initial assessment for Functional Skills English and maths from Entry 1 to Level 2, while Dilr Academy's live page describes knowledge tracing inside a course but no standalone initial assessment product. A college relying on Smart IA should keep it.

Where the two meet is inside a course. Academy's knowledge tracing estimates what a learner has mastered and sets the difficulty of the next step, so two learners on the same course can take different paths. That is placement within a course, not placement onto a qualification level, and the distinction matters for funding paperwork. Our overview of what Dilr Academy is explains the course model in more depth.

Can a school or multi-academy trust license Dilr Academy?

Dilr Academy's live page says it is licensed to the institution that owns the curriculum, names universities, colleges and training providers, lists colleges and schools among its audiences, and says it is not sold to individual learners. It has built a school curriculum course, Year 9 Mathematics on the England KS3 programme of study, so a school or trust with a specific curriculum need should ask about it on a call.

For most schools, though, the question behind a Century Tech alternative search is about practice and homework platforms, and that comparison lives in our Sparx guide. Our sector view of where AI pays in UK education sets out which DILR products suit which kind of institution, and the AI teacher buyer's guide linked above covers the full market. More posts in this series sit in our strategy category.

Want to see this on your own syllabus? Explore Dilr Academy for colleges, read our comparison of AI education tools, see our DATS methodology, or read about our approach to placing AI inside institutions.

Written by the Dilr.ai engineering team, practitioners who ship enterprise AI in production. Follow us on LinkedIn for shipping notes, or subscribe via the RSS feed.