In preparation for our research for our State of AI Account Management 2025 report, we polled a sample of Kaizan clients and asked them to identify the clients they thought they were least efficient on.

Then we looked for the common patterns across these ‘inefficient’ clients.

Efficiency can be measured in many ways and each organisation is different.

It could be repeating work, spending time on tasks or workflows that could be automated, or too much team time spent on tasks that aren’t billable. Operations leaders know that inefficiency rarely shouts – it creeps in through too many meetings, duplicated tasks, manual handovers and hidden admin.

We isolated three operational patterns that cluster in the least efficient teams managing these clients:

1. Meetings where 50% of participants speak less than 10% of the time.

Too many attendees, too little contribution. This is one everyone can fully appreciate.

How many calls were you or teammates on where you didn’t need to be there? Imagine what that number is across your entire company… now imagine if that person’s time was put on something more productive, and you can see how inefficient this can be.

It turns out it’s trackable; if on your client calls you consistently see less than 10% talk time for over half the people on the call, then your account is inefficient.

You can direct individuals to rewatch calls later or ask questions of the transcript so they don’t miss key actions. You can also train them to be more impactful. But if they’re not required on the call, it’s ultimately time wasted.

Start tracking talk time and engagement on each account to uncover inefficiency.

2. More than 20% of the teams’ working day is spent on internal admin, updates and being a water carrier of information between people.

In lower-performing teams we saw hours consumed by updating Project Management systems, posting to Slack, adding notes to CRM that no one will read, routing briefs and chasing for updates.

This eats up capacity, slows delivery, and hides the real cost of “coordination tax”. Conversely, the most efficient teams are already leveraging AI heavily on all daily internal admin tasks with prescribed playbooks and automations;

>> AI Agents for pre-meeting prep and agenda drafting

>> AI Meeting Assistants that generate follow-up emails and update project management boards, CRM fields and Slack updates

>> AI Chatbots to search and retrieve knowledge from past calls, emails, chats, and docs you now need access to

3. Reporting errors dominate risk topics, signalling broken underlying processes.

When reporting mistakes are mentioned 3× more than any other risk topic this infers future lack of client engagement, but was also seen in the least efficient teams.

What does this tell us? That there’s operational work to do on reporting QA.

By evaluating the low sentiment/friction areas in all dialogue it shines a light on where inefficiency lies, and top of the list is friction relating to reporting accuracy.

The more reporting inaccuracy, the more friction and time spent fixing the data.

Operational Levers You Can Pull Immediately

>> Design attendance intentionally.

Use “meeting by invite only when you have to speak” rules. Reduce status-only calls if attendees won’t engage. Leverage a shared Meeting Assistant like Kaizan’s so that others can easily watch back meetings, ask questions with a chatbot, or have attendees share the moments they need to know about. Create mandates and best-practice playbooks for meeting capture so your team isn’t wasting hours on calls they’re not required on.

>> Automate the easy admin tasks.

Go AI first with playbooks and workflows specific to daily admin flows. Now, handoffs, system updates, and follow-ups can be generated. Free your people for high-leverage work. Start to build a plan on how you’ll save your team hours on their most common, redundant, least valuable tasks.

>> Reporting QA & Workflows.

In a sample of a confined dataset, there appears to be inefficiency, client and operational risk in reporting workflows. How streamlined are yours?

In operations, efficiency isn’t about doing more with the same, it’s about doing less of the wrong things.

The teams that do it will scale client-service without simply adding headcount or complexity.