ICT projects: from design to the day after go-live

We manage the whole project —design, pilot, implementation, testing and documentation— and we stay on for the first few weeks, which are the ones that hurt.

Tell us about your project

A project does not end the day it is switched on

Most infrastructure projects are delivered three times. The first, on go-live day, when everything seems to work. The second, two weeks later, when whatever nobody had tested shows up. And the third, a year later, when someone has to touch it and discovers there is no documentation, or that what there is does not match what is installed.

We work so that there is only one delivery. That means designing before buying, testing in a pilot before touching production, leaving a test plan that gets signed, and handing over the real map of what has been installed. And then staying on top of it for a few weeks, until the system has been through a month-end close, a full backup and its first scare.

The whole project is run by one named person. You will not have to explain your case three times to three different contacts.

Projects: the system is built piece by piece

The kind of work we are usually asked to do

Data centre renewal. Replacing servers, storage or network that can no longer cope, without stopping the business along the way.

Virtualisation and consolidation. Moving from many under-used physical machines to a virtualised platform managed as one.

Migration to the cloud or to a hybrid model. First deciding what makes sense to move and what does not, and then moving it with a rollback plan.

Backup site and disaster recovery. Replication, a recovery plan and real tests, not a document nobody has ever run.

Network segmentation and security. Separating what should not see each other, which is the measure that most reduces the damage when something gets in.

Opening or moving offices. Network, workstations, telephony and connectivity working on the day the doors open.

The six phases of an Algoritmio ICT project: design, pilot and proof of concept, implementation, testing and acceptance, documentation and post-launch support, with project management from start to finish.

1. Design

Before asking anyone for a quote, you need to know what is going to be built. We prepare a high-level proposal —what is done, why, what alternatives there are and what each one involves— and, once the path is chosen, the low-level proposal with the technical detail: equipment, versions, addressing, integrations and downtime windows. It is the document the work is then carried out from, and against which the delivery is checked.

High-level proposal · Low-level proposal

2. Pilot and proof of concept

When the risk justifies it, we first build a mock-up: a pilot with the real equipment on a limited part of the installation. It is used to check that what the manufacturer promises holds true in your specific environment, which is where surprises usually appear. If the proof of concept fails, it has failed somewhere where nothing happens.

Pilot installation · Proof of concept

3. Implementation

The go-live, in the agreed window and with the rollback plan written before starting. It includes the basic configuration of the equipment and the advanced one —high availability, policies, integration with what is already there— and, when needed, customising the applications so they fit the new platform.

Basic configuration · Advanced configuration · Application customisation

4. Testing and acceptance

A test plan written before implementation, not improvised afterwards. Each item is carried out in front of whoever is responsible and signed off. Until it is signed, the project has not been delivered. This is what avoids the "I thought that was included too" conversation.

Test plan · Acceptance test

5. Documentation

We hand over the real map of what has been built: diagrams, addressing, properly managed credentials, operating and recovery procedures. Written so that someone who was not on the project can understand it, because two years from now that person will be the one reading it.

Installation mapping · Handover documentation

6. Post-launch support

For the first few weeks we stay on top of the system: we monitor, adjust whatever is needed and answer the questions of the team that is going to live with it. It is the part almost nobody includes and the one that prevents the most problems, because the faults of a new deployment do not show up on day one, they show up at the first month-end close.

Babysitting after go-live

And above all six, project management: planning, monitoring, coordination with your suppliers and a weekly status report. From the first meeting to handover.

Tell us what you need to do

With one conversation and a look at what is already there, we can tell whether the project will take one month or six, and where it is best to start. If it turns out you do not need the whole project, we will tell you that too.

+34 91 005 28 58 info@algoritmio.com

Write to us See Consulting See Management & renovation