System & architecture review
An assessment of what you run today: what is sound, what is fragile, what is costing more than it should, and the order in which to address it.
Services
Six things we are asked to do
Every line below sits under one of the four activities MPH Media & Tech Ltd registered at Companies House. Each can be bought on its own; the value of buying more than one from us is that they stop being separate projects.
01 · SIC 62012
Bespoke software development
Software written for one organisation's requirement. We are usually called when an off-the-shelf product has been stretched past what it can sensibly do, or when the process that makes a business distinctive is being carried by a spreadsheet.
The first thing we do is try to talk you out of it. Custom software is a long-term commitment, and a great deal of what people ask us to build can be solved with a configuration change or a product that already exists. When it genuinely cannot, we build it properly: conventional technology, plain architecture, tests around the parts that matter, and documentation written for whoever comes next.
Typically this means
- Web applications and internal tools used by staff every day
- Member, subscriber and client portals with sensible access control
- Integrations between systems that were never designed to speak to each other
- Data import, cleaning and reporting pipelines
- Small applications that replace a fragile spreadsheet or manual routine
- Incremental work on an existing codebase you already own
02 · SIC 62012 & 62090
Media & content platforms
The systems between a piece of work and its audience. Publishers, studios, agencies and membership bodies all end up needing the same machinery, and almost all of them end up with three products doing one job badly.
We build and consolidate that machinery: getting material in, moving it through review, scheduling it, putting it out to the channels that matter, and reporting honestly on what happened. Where a good product already exists we will integrate it rather than rebuild it — the aim is one coherent workflow, not a monument.
Where this earns its cost
- When the same asset is being uploaded by hand to several places
- When approvals happen in email and nobody can prove what was signed off
- When the archive exists but nothing can be found in it
- When reporting is assembled manually each month from three dashboards
- When a rights or embargo mistake would be genuinely expensive
03 · SIC 62020
Independent technology consultancy
Advice bought before the money is committed, from someone with no product to sell you and no commission riding on the answer.
Consultancy is delivered as a written document, not a workshop that leaves no trace. It states what we found, what we recommend, what it will cost, what could go wrong and what a reasonable alternative would be — in language a finance director can read without a translator. You are buying a position we are willing to put our name to, and you may act on it with us, with somebody else, or not at all.
Build, buy or leave alone
A costed comparison of the realistic options for a requirement, including the option of changing nothing — which is more often right than suppliers admit.
Supplier selection & scrutiny
Help writing a specification, reading proposals, and asking the questions that separate a capable supplier from a confident one.
Technical due diligence
An independent read on a system you are about to acquire, invest in, or inherit — code, infrastructure, licensing and key-person risk.
04 · SIC 62030
Managed platform operations
Computer facilities management: the running of the thing after it is built. This is the least glamorous service we offer and the one that decides whether the rest of it was worth doing.
Hosting & environments
Infrastructure specified, provisioned and documented — in your cloud accounts, under your billing, with our access grantable and revocable by you.
Deployment pipelines
Releases that are repeatable and reversible, so shipping a change is a routine act rather than an event requiring a quiet afternoon.
Monitoring & alerting
Alerts tuned to things that actually matter. A monitoring system nobody trusts is worse than none, because it teaches people to ignore it.
Backup & restore
Backups verified by restoring them on a schedule. An untested backup is a belief, not a control, and we will say so in writing.
Patching & dependency care
Security updates and dependency upgrades applied on a rhythm, before they accumulate into a project of their own.
Incident response
A named contact, an agreed response window, and a written account afterwards of what failed and what has changed so that it does not recur.
Operations run on a rolling monthly retainer with a short notice period on both sides. We do not ask clients to sign multi-year commitments to keep their own systems running.
05 · SIC 62090
Migrations & project rescues
Moving a system somewhere better, or picking up a project that has stopped moving. Both start the same way: with a paid assessment and an honest answer, including the answer you may not want.
A rescue is not a rewrite by default. We read what exists — the code, the infrastructure, the contract, the history — and give you a costed comparison between finishing it, replacing it, and renegotiating with whoever built it. In a meaningful number of cases the existing work is more salvageable than anyone has been willing to say out loud.
- Platform and hosting migrations, with a tested rollback
- Data migration with reconciliation both sides of the move
- Takeover of an unfinished or abandoned codebase
- Recovery of access, credentials and ownership from a departed supplier
- Untangling systems after a merger, split or change of provider
06 · SIC 62020 & 62030
Resilience & continuity
Most organisations discover the state of their backups on the worst possible day. This service exists to move that discovery forward to a calm afternoon of your choosing.
The work is deliberately unexciting. We establish what would actually happen if a key system, a key supplier or a key person became unavailable; we test the controls that are supposed to help; and we write down the plan in a form somebody can follow at six in the morning without needing to understand the architecture.
It is scaled to the organisation. A ten-person practice does not need an enterprise-grade continuity programme — it needs three things fixed and one page printed out.
Commercials
How the work is priced
We do not publish rate cards, because a number without a scope is not information. We do publish the shape of every arrangement we offer, so you know what kind of quotation to expect before you ask for one.
Scoping conversation
No fee. Around half an hour, by call or in person, to establish what the problem is and whether we are the right people for it. It sometimes ends with a recommendation to do nothing.
Written assessment
Fixed fee, quoted from the scope of the question. You receive a document you own outright and may act on however you wish, including with another supplier.
Development & project work
Fixed price per stage against an agreed written scope. Work outside that scope is re-quoted and agreed before it is done, never absorbed and invoiced later.
Managed operations
Rolling monthly retainer covering an agreed set of systems, a response window and a defined amount of change work. Short notice period on both sides.
Day rate
Available where a client's procurement genuinely prefers it — for example alongside an in-house team. We will usually suggest a fixed stage price instead.
Third-party costs
Hosting, licences and subscriptions are bought in your own accounts wherever possible. We take no reseller margin or vendor commission on anything we recommend.
Quotations state whether VAT is applicable at the time of quoting. Payment terms and the general basis of engagement are set out in our terms.
Service questions
Before you send the brief
No. Each service line stands alone and a good number of clients buy exactly one. Plenty of engagements are a single written assessment that we never build anything from. The chain is available, not compulsory.
Mainstream, well-supported web and cloud technology, chosen for how easily somebody else could pick it up rather than for how interesting it is. We are deliberately conservative here: a system that only one supplier in the country can maintain is a liability dressed as an advantage.
Where you already have a technology in-house, we will normally work within it rather than introduce a second stack for our own convenience.
Yes, and it is a common arrangement — they hold the desktops, network and day-to-day support; we hold the application and the platform. It works best when the boundary is written down at the start, which we will propose in the first assessment.
Assessments and reviews usually start within a fortnight. Build work depends on current commitments and we will give you a real date rather than an encouraging one. Where something is genuinely urgent — a live outage, a supplier who has disappeared — say so in your first email and we will tell you the same day whether we can help.
Regularly. We decline work that needs a larger team than we are, work where the requirement is genuinely unknowable at the price being offered, and work where a client wants an outcome the technology cannot honestly deliver. Saying so at the first conversation costs both parties far less than discovering it in month four.
Next step
Describe the problem, not the solution
The most useful first email tells us what is going wrong and what it is costing you. We will come back with what we think it involves and roughly what it costs to find out properly.