AllSet.solutions was created by the team behind AllSet.chat, a service used by thousands of solo service providers to run their businesses through messaging. AllSet.solutions brings that experience into the enterprise through professional services, technology and managed services.

CEO Robin Lacey has 18+ years in software, AI and cloud, including large-scale transformation programs across energy, financial services and government.

CTO Mike Hogan has 30+ years building operational software systems and leading engineering teams.

CMO Michael Darragh brings 25+ years of B2B and B2C marketing leadership across EMEA, APAC and the Americas.

The team combines that enterprise experience with two years of building and operating AllSet.chat.

Every engagement starts with professional services. We identify the problem, define the operating model and scope what needs to change.

From there, we may build custom technology, integrate with your existing systems, provide parts of the capability as a managed service, or deploy the capability within your own environment and hand it over to your internal teams.

The delivery model depends on your requirements, technical capability, data sensitivity and governance model.

If there is a meaningful problem to solve, the next step is usually a fixed-price, fixed-outcome Conversation Audit. The cost depends on the scale and complexity of the area being examined.

Implementation work is scoped separately and may include professional services, fixed-price delivery or a managed service. Managed services are typically charged at an agreed recurring rate based on scope and volume.

Where third-party messaging or AI services are required, those costs are passed through at the underlying rate rather than marked up.

If there is a good fit, we agree the scope of a Conversation Audit.

The audit identifies the highest-value opportunities and gives us the evidence needed to define the first phase of work.

From there, we agree a Statement of Work, implement the highest-priority opportunity and measure the return before moving on to the next.

It defines:

• Who should be contacted, why and on what basis.
• What an automated agent is authorized to say and do.
• How conversations are supervised, assessed, retained and improved.

We measure it on two dimensions:

Control: Can every conversation be justified, supervised and evidenced?

Return: Does the accumulated conversational record improve outcomes, targeting and cost?

Conversation Ops turns conversations from an informal layer of the business into something the organization can observe, govern and rely on.

Common signs include operational processes being coordinated through channels such as text messages, Teams, Slack or email; people manually moving information between conversations and systems; AI and conversational tools operating in isolation; limited visibility over why conversations happen or when they go wrong; and valuable knowledge disappearing into inboxes, channels or people's heads.

It is also relevant when your organization wants to move beyond mass messaging and use AI to identify the right customer, member or employee, understand why a conversation should happen, and safely drive an outcome.

A mature Conversation Ops capability makes those conversations observable, governed and measurable, with clear accountability, intervention paths and organizational learning.

Common problems include:

• Operational processes held together by messaging.
• Excessive manual work around conversations.
• Fragmented conversational data.
• Poor visibility or governance.
• Valuable knowledge trapped in people or message histories.
• AI systems that can talk, but cannot safely drive meaningful outcomes.
• Outbound systems that can broadcast, but cannot identify who genuinely needs a conversation.

That typically means high volumes of customer, member or employee interactions, complex service operations and enough scale for inefficiency or poor conversational processes to create a material cost.

We have experience across sectors including Aesthetics & Dentistry, Commodities & Global Trade, Fitness & Wellness, Home Trades, Luxury & Advisory, Solopreneurs, Training & Credentialing and Workforce Coordination.

Our sponsors are typically senior leaders in operations, transformation, AI, technology or customer experience.

Conversation Ops is generally not designed for small businesses. It works best where the problem is strategically important and solving it can produce a meaningful return.

In practice, that can mean:

• Turning noisy messaging channels into observable operational processes.
• Connecting conversations to systems of record.
• Automatically updating systems based on conversational activity.
• Helping staff identify the right customer, member or colleague for a particular conversation.
• Capturing what is learned so the next decision is better informed.

For a workforce team, that might mean managing shift changes through conversation while keeping operational systems up to date.

For a health and fitness operator, it might mean identifying the members who would genuinely benefit from speaking to a particular trainer.

For a premium retailer, it might mean identifying the handful of clients most likely to value a unique product based on behavior, preferences and previous conversations.

AllSet.solutions is technology-agnostic. We can work with Microsoft, Salesforce, contact-center platforms, specialist agents and your existing systems and data.

We are not trying to replace your systems, dictate which AI models you use or own your prompts.

We focus on whether the right conversations can happen: what data informs them, what triggers them, what an agent is authorized to do, how outcomes are measured and how the organization learns from them.

The goal is not another AI tool. It is better conversations across the tools you already have.

Conversation Ops can use operational data, customer behavior and previous conversations to make better-informed decisions about who to engage.

It also creates a learning loop. The outcome of each conversation becomes part of the organization's data and improves future decisions.

The aim is not to send more messages. It is to have fewer, better-informed conversations that drive measurable outcomes.

We can connect conversational processes across existing CRM, messaging, AI, data and operational systems.

Our aim is not to replace your technology. It is to make the conversations happening across it more connected, observable, governable and useful.

Where a system is closed and provides no usable access to its data, that may limit what can be integrated.

The scope is agreed upfront based on the outcome you want to achieve.

We examine the relevant people, processes, technology, data and costs to understand:

• What conversations are happening today.
• How they are being handled.
• Where they create cost, friction or risk.
• What conversations the organization wants to have but cannot manage effectively today.
• Where better data, automation or governance could create measurable value.

The output is a clear view of the current landscape, the highest-value opportunities, the likely return and what we recommend addressing first.

We work with the sponsor to identify the area where action is most likely to create value and focus the audit there.

The objective is not to understand every conversational process in a complex enterprise within a few weeks. It is to generate a strong enough signal to identify where action will create the greatest value.

The audit gives us a clear view of where the strongest opportunities are.

We define the first phase of work and deliver in stages, starting with the opportunity that can create the clearest measurable value.

Each phase should demonstrate a return before moving on to the next.

There are already enough AI pilots that never reach production.

Our preference is to identify a tightly scoped problem, solve it properly, put it into use and measure the return.

Start small, but build something that matters.

A Conversation Audit typically involves around two weeks of focused work, although elapsed time may be closer to a month depending on access to people, systems and data.

Once priorities are agreed, a first implementation would typically take around two to three months.

The aim is to prove value in the highest-priority area and build from there.

During the audit, we map relevant processes end to end and establish how long the work takes, how much human effort is involved, where it fails or gets repeated and where conversations create bottlenecks.

Depending on the use case, we may measure:

• Employee time and operational cost.
• Cost per interaction.
• Failure demand.
• Response and resolution times.
• Revenue opportunity.
• Conversion.
• Customer, member or employee outcomes.

We calculate ROI from the process being changed, not from the technology being installed.

The measures depend on the use case and may include conversation outcomes, escalation rates, failure rates, cost, response times, conversion, human intervention or data quality.

Once live, those measures are monitored so we can see whether the capability is delivering the expected value and where it needs to improve.

We need to be able to explain why a person was selected, which data informed the decision and how it shaped the conversation.

That involves two types of information:

Facts: deterministic data from systems of record, such as transactions, visits and account activity.

Learnings: qualitative information derived from previous conversations, such as preferences, interests and context.

Both need to be structured, reliable and traceable.

Data validation is fundamental to both the quality and governance of Conversation Ops.

During the audit, we identify the specific data points each process depends on.

Where possible, we avoid broad database access. We prefer narrow, controlled interfaces that retrieve only the information required for a particular decision or conversation.

Conversation Ops can also write learnings and outcomes back into existing systems so future conversations benefit from what the organization has already learned.

The principle is to start with the minimum access required and expand only where there is a clear reason to do so.

We can work with partial system access, deploy components inside your environment or build capabilities that your own teams operate.

Where required, sensitive queries can take place entirely within your own environment.

The objective is to give Conversation Ops access to the information it needs while keeping sensitive data where it belongs.

Regulated organizations already have requirements around data access, processing, security and governance. Our role is to work within those requirements, not replace them.

That may mean keeping sensitive processing within your own environment, using anonymized data or integrating with existing governance and data platforms.

These requirements are agreed as part of the Statement of Work.

We build Conversation Ops around your governance model, not the other way around.

Where your policies require it, data and processing can remain within your own infrastructure.

If we provide a managed service, access and data-handling requirements are defined during scoping and incorporated into the Statement of Work.

Where your security requirements mean a third party should not process particular data, we design the solution so that processing remains within your environment.

We do not ask you to weaken your security model to use Conversation Ops.

Where systems need to run within your own environment, particularly in regulated organizations, your existing technology teams will typically manage the underlying infrastructure and runtime.

If you want to take full ownership of the software and continue developing it internally, you will need the appropriate engineering capability.

The delivery model is designed around your technical capabilities, governance requirements and appetite for ownership.

We typically need:

• A senior stakeholder with a clear mandate.
• Access to key people who understand the operational processes being examined.
• A technical stakeholder where systems or data are involved.

During implementation, we aim to work through the governance, technology and data structures you already have rather than creating unnecessary new roles or processes.

Messages can be checked against organizational rules before they are sent, and live conversations can be monitored for signs that they have moved beyond what the agent is authorized to handle.

When necessary, the conversation can be stopped and escalated to a human.

The organization also retains the evidence needed to understand what happened: the data that informed the interaction, the instructions given to the agent, the conversation itself and the point at which it failed.

The objective is not to pretend AI never fails. It is to make failure observable, recoverable and improvable.