MMOOZZGG.com

AI assistant

In short

AI does not make decisions for the owner, but it can make sure each decision rests on current figures and a tested hypothesis. The article gives a three-step method: collect the numbers, test the hypothesis on one process with a metric, then decide from a before-and-after report. The main takeaway: a small business needs no data warehouse and no analyst on staff.

Key ideas

  • Before a decision, an agent collects leads, conversion and the current cost of the process from your systems.
  • A hypothesis is tested on one process with a metric, instead of an argument.
  • At the end of a pilot, the owner gets before and after numbers and three options: stop, keep on support, or scale.
  • Pilots suit decisions about a process with a flow and a number, such as hiring a person or using an agent.
  • An agent can misread a question or take the wrong number, so its output needs checking.

Prepared by AI from the article

Data driven decisions with AI: how to decide by numbers and pilots

AnalysisConsulting

Medium4 min read

AI does not make decisions for the owner, and it should not. Its job is to make sure the owner makes data driven decisions from current figures and a tested hypothesis, not from a feeling. Mike Mozg’s method has three steps.

On this page · 8 sections

Step 1. Numbers before the decision

Before any decision, an agent collects numbers from your systems: how many leads there were, what the conversion is and what the process costs now. If the numbers do not exist, that is the first finding: the process is not measured, and it is too early to decide.

Step 2. A pilot instead of an argument

A hypothesis is tested on one process with a metric. For example: answering a lead in one minute instead of one hour will raise the share of leads that reach a deal. The pilot runs 14 days, the metric is written in advance, and the result does not depend on who argued louder.

Step 3. A decision from the report

At the end of the pilot you get a report with “before” and “after” numbers and three options: stop, keep on support, or scale. If there is no payback, that is also a result, and the report records it.

Which decisions suit a pilot

Pilots suit decisions about processes that have a flow and a number: hire a person or use an agent, change the order of answering leads, open a new request channel. Pilots do not suit decisions that cannot be tested in 14 days on one process, such as a strategy change, entry into a new region or a large purchase. For these, AI prepares the numbers, and the test needs another method.

How to phrase the question for a pilot

The decision must read as a choice: hire a manager or use an agent, keep the process or change it. The choice needs a number that shows after 14 days which option won. If you cannot name such a number, the question is not ready to test.

What data you need for data driven decisions

Data driven decisions in a small business need no data warehouse and no analyst on staff. Three sources that you already have are enough: your accounting system (1C), your CRM and your bank. The agent reads them with a separate account and collects numbers on leads, deals, cash and debts. You do not need to change systems.

An example of a decision from a pilot

The question: hire a second manager for leads, or use an agent? Numbers before: leads per day, leads with no answer for over an hour, and the cost of a manager with taxes. Pilot: an agent takes leads for 14 days, and the metrics are reply time and the share of leads in the CRM. Decision: from the report, with a clear price for each option.

Where AI makes mistakes and how to allow for them

An agent can misread a question or take the wrong number. So every report names its data source, disputed cases go to a person, and in the first weeks the numbers are checked by hand. The responsibility for the decision stays with the executive.

Questions and answers

Can AI make management decisions?

It can prepare them: collect numbers, compare options and show risks. The executive makes the decision.

How do you test an idea without a large investment?

With a pilot on one process: 14 days, with a fixed price and the metric in the contract.

What if the company has no numbers?

The first step is to set up their collection: an agent reads the accounting system, the CRM and the bank and builds a morning report.

How does AI help decisions every day, not only in a pilot?

A morning report shows deviations from the plan and from the previous period, with the source next to each number. The executive asks about a specific line and gets an answer from data, not from memory.

Who runs the pilot in the company?

One responsible employee who gives half an hour a day to the agent’s disputed cases. The executive takes part twice: approving the metric before the start and deciding from the report at the end.

What happens to the money after the pilot?

If you build a department, the pilot cost counts toward its price. If you keep one agent, you pay a monthly fee for support. If the metric is not reached, the pilot fee is returned under the contract.

Can you test several ideas at once?

It is better to test them one by one. When several processes change at once, you cannot tell what worked. A pilot takes 14 days, so you can test several ideas in a row within a quarter.

Publications

Contact us