Consulting
Advice with a price
attached to every option
Most consulting reports fail at the point where a recommendation meets a delivery estimate. Ours carry sizing, sequencing and a named engineer who will still be there when the build starts, because advice you cannot cost is not advice.
Overview
Short engagements with a written answer, a cost, and someone who will stay to build it.
We run short engagements with a defined question. Should this platform be rebuilt or refactored? Why does the release take nine days? What will it cost to serve four times the current load? Which of eleven vendor proposals survives contact with your data volumes? Each of those has a written answer with the working shown.
The method is unglamorous: read the code, query the production database, sit with the operations team during an incident, and check the claims in the vendor deck against a load test. We assume nothing in the documentation is current, because about a third of it never was, and another third stopped being true two releases ago.
Deliverables are short. A twenty-page assessment that an executive reads and an architect can act on beats a two-hundred-page appendix nobody opens. Where we recommend building something, the estimate carries a range and the assumptions that would break it, and we are happy for another firm to deliver against it.
- 19 days
- median engagement length from kickoff to written recommendation
- 42%
- cloud spend reduction identified at Aldridge Risk, 31% realised
- 2 of 11
- vendor proposals survived load testing on real data volumes
Capabilities
What this covers
Six areas we staff properly. If your problem sits outside them, the honest note at the foot of this page says so.
Architecture and code assessment
A read of the codebase, the deployment path and the production telemetry, producing a ranked findings list separating what is genuinely risky from what merely looks untidy in review.
Technology selection and vendor evaluation
Structured scoring against your scenarios, with proof-of-concept builds where the claim is load-bearing. Vendor benchmarks are re-run on your data volumes before anything is signed.
Cost and capacity modelling
Cloud and licence spend modelled against projected growth, broken down per workload, with the specific line items that will bite at three times current volume identified in advance.
Delivery and engineering effectiveness reviews
Lead time, change failure rate, review latency and on-call load measured from your own tooling, then traced to the specific process or architectural constraints causing them.
Security and resilience posture reviews
Threat modelling, dependency and secrets review, and a tested recovery position: we ask when the restore was last exercised, and the answer is usually more interesting than the backup policy.
Due diligence for acquisitions and investment
Technical diligence on a target's codebase, team, licensing and infrastructure commitments, delivered against transaction timelines with findings graded by remediation cost.
Deliverables
What you get
- Ranked findings report with severity, effort and cost of inaction
- Reference architecture options with a costed trade-off comparison
- Capacity and spend model projected across three growth scenarios
- Delivery metrics baseline drawn from your own pipelines and incident data
- Prioritised remediation roadmap with sequencing and staffing shape
Stack
What we build it with
- OpenTelemetry
- Grafana
- SonarQube
- Semgrep
- k6
- Terraform
- Kubernetes
- PostgreSQL
- AWS Cost Explorer
- Backstage
Process
How the engagement runs
Two-week increments against a written definition of done. You can stop at any increment boundary and keep everything built so far.
Frame the question
A half-day session to make the question specific and falsifiable, agree what evidence would answer it, and set the date the answer is needed by.
Gather evidence
Read-only access to repositories, telemetry and infrastructure, plus interviews with the engineers and operators who carry the pager.
Test the claims
Load tests, restore drills or proof-of-concept builds where a recommendation depends on something nobody has verified recently.
Write it down
A short report with findings, options and costed trade-offs, reviewed with the technical team before it reaches the executive audience.
Stay or step back
We hand the roadmap over, and whether we deliver against it is your decision. Engagements that end here are a success, not a failed sale.
When this is the wrong engagement
If you already know the answer and need capacity to build it, skip the assessment and buy delivery instead.
FAQ
Questions we get asked
- Will you recommend your own delivery services?
Sometimes, and we say so in the report where it happens. The assessment fee is fixed and does not depend on what we recommend, and roughly a third of our consulting engagements end with the client building it themselves or hiring another firm. We would rather keep the referral.
- How much access do you need?
Read access to repositories, CI history, production telemetry and cloud billing, plus time with four to six engineers. We do not need write access to anything, and we sign whatever data handling terms your security team requires before the first credential is issued.
- How long does an assessment take?
Two to four weeks for a focused technical question, six to eight for a full estate or diligence review. Anything advertised as a two-day audit is reading documentation rather than evidence, and documentation is the artefact least likely to reflect what is actually running.
- Can you challenge a decision that has already been made?
Yes, and that is often the most valuable version of the engagement. We will set out the evidence for and against, quantify the switching cost of reversing course now versus in a year, and leave the decision where it belongs. We do not manufacture agreement.
Related
More in Consulting
Consulting
Digital Transformation
Sequenced change programmes that ship working software, not slide decks.
Consulting
Dedicated Development Teams
A standing squad with its own lead, its own metrics and a memory of your codebase.
Consulting
Staff Augmentation
Named specialists into your existing team, under your management, on your board.