AI & Automation
AI advice from people with nothing to sell you.
A lot of firms are selling “AI development” right now. We’d rather tell you whether you need it. We look at how your team works, explain what’s available, and help you put the right pieces in place, or leave things as they are.
Straight answers.
What exists
The tools and approaches available today, in plain language, without the jargon.
What works
What has a realistic chance of working with your systems, your information and your team.
What’s hype
Claims and demos that sound impressive but won’t change much in practice.
What’s just expensive
Options that could work, but cost more to build and maintain than they save.
What an engagement involves
Observe the work.
Meet the people involved, map the existing process, and understand its exceptions.
Choose a starting point.
Prioritise a useful, manageable workflow and agree what improvement would mean.
Build and test.
Implement a pilot around the relevant systems and test it with representative information.
Put it into practice.
Train users, document responsibilities, and agree how exceptions will be handled.
Review and improve.
Assess adoption, reliability, and the results before extending the work.
A working day, with fewer things in the way.
Choose a process to see an example of how the work could move.
New enquiry
- StartsEnquiry arrives
- AutomatedSummary prepared
- AutomatedCRM draft created
- AutomatedOwner assigned
- Your teamResponse reviewed
The team receives the context and a clear next step.
Internal knowledge
- StartsQuestion asked
- AutomatedApproved sources searched
- AutomatedAnswer drafted with references
- Your teamUncertainty flagged for a person
People can check where the answer came from and escalate unclear questions.
Weekly reporting
- StartsAgreed inputs collected
- AutomatedMissing data flagged
- AutomatedReport drafted
- Your teamOwner reviews
Preparation becomes more consistent while interpretation stays with the team.
Content approval
- StartsApproved brief supplied
- AutomatedDraft prepared
- Your teamEditor reviews
- AutomatedPublishing handoff
The workflow supports the editorial process and keeps approval explicit.
Department examples
Illustrative applications. What is possible depends on your systems and information, and a person stays responsible for every judgment that matters.
| Team | Illustrative application | Human responsibility |
|---|---|---|
| Sales | Summarise enquiries, prepare CRM entries, and route follow-up. | Review sensitive or ambiguous enquiries and approve responses where required. |
| Marketing | Turn approved source material into drafts and route content for review. | Own editorial accuracy, brand judgment, and publication approval. |
| Operations | Extract information from documents and flag missing fields. | Check exceptions and approve consequential changes. |
| Internal teams | Search approved policies and return answers with source links. | Maintain source documents and resolve unclear cases. |
| Management | Prepare recurring reports from agreed data sources. | Verify significant figures and interpret the implications. |
Questions
Do we even need AI?
Not always. Sometimes a simpler change to a process, or a tool you already pay for, does the job. If AI isn’t worth it for you, we’ll say so.
Do we need to replace our existing software?
We start with what you already use. Where it supports the workflow, we build around it. Any new tool or change is discussed as part of the scope.
What does in-house consulting mean?
Working directly with the people doing the work, understanding the process in context, and supporting implementation and adoption. The mix of on-site and remote work is agreed for each engagement.
Will our team need technical skills?
The aim is a process people can use in their working day. Training and documentation are tailored to the team and the tools involved.
How is company information handled?
We agree what information the workflow needs, who can access it, and which tools will process it before implementation. Tool selection and configuration are part of that discussion.
How do we decide whether it is working?
We agree measures before the pilot. Depending on the process, those might include handling time, manual steps, error rates, or whether the team is using the new workflow consistently.
What happens after launch?
Ownership, documentation, maintenance, and any ongoing support are agreed as part of the engagement.
Show us the process everyone complains about.
We’ll start by understanding why it works that way and what could improve.
Let’s look at it together