With the tools your team uses
Interpret data to support decisions
Learn to frame a business question, check the data and explain a decision with its limits. We use sales and operational examples in a tool that fits your team.
Before we begin
Basic table handling and familiarity with a work process. Each edition uses one agreed tool, chosen around the team’s environment; CSVs can also be reviewed as tables. No administrator permissions, BI knowledge or access to real company data required.
We agree one tool for opening tables, doing simple calculations and, where appropriate, presenting results. We prepare the three fictional CSVs, definitions and a blank worksheet. If using the client’s tool, we check features and access beforehand; production information is neither connected nor imported.
Learn methods and criteria you can apply with your own tools. We adapt the exercises to your team’s environment and confirm the features, access and usage needed for the agreed practice.
See preparation and terms →By the end you will be able to
- 01
Define what an indicator measures, which records it covers and its reporting period.
- 02
Detect duplicates, missing values and comparisons that mix different situations.
- 03
Prepare a clear table or chart and explain which decision it supports and what still needs checking.
What we will work on
An isolated number is not enough to make a decision. We begin with a specific question, trace the calculation to its records and test a conclusion that sounds convincing. We then apply the same judgement to another process. Practice, the individual variation and feedback take place within the eight hours.
01Know what you are measuring
Question, entity, period, source and shared definitions before calculating.
You will work on
- Frame a question the available records can answer.
- Define the opportunity or request, reporting period and snapshot; locate the source and meaning of each field.
- Review a duplicate copy and a missing status without filling gaps by intuition.
- Practice
- Prepare the sales-opportunity extract: distinguish 21 rows from 20 unique opportunities and record why one copy is removed.
- How you will check it
- A reviewed table and decision log that retain the unknown status and make the cleanup reproducible.
02Compare in context
Counts and rates, denominators, incomplete data and differences between groups.
You will work on
- Calculate a count and a rate with an explicit numerator and denominator.
- Distinguish open opportunities, known closed outcomes and an unconfirmed status.
- Compare request times by request type and by the number with a recorded measurement.
- Practice
- Compare two sales channels and two request weeks; explain why a higher overall average can coexist with lower averages within each type.
- How you will check it
- Reproducible calculations, visible bases and an explanation of records omitted from each calculation.
03Explain a decision
Presentation, uncertainty and a recommendation someone else can review.
You will work on
- Correct a scale that exaggerates a difference and add units and bases.
- Separate an observation from a possible explanation or an unsupported causal claim.
- Write a short recommendation and complete a variation with incomplete records.
- Practice
- Deliver a decision note on one case and individually review a small on-time delivery dataset.
- How you will check it
- Someone else can reconstruct the calculation, locate missing information and understand the next step without reading unsupported certainty.
Five wins do not necessarily make a better channel.
Classroom example with fictional data
You will review two fictional cases: sales opportunities by channel and operational request times. You will correct a misleading comparison and complete an individual variation, retaining the sources and the information that remains unconfirmed.
Fictional classroom case. We review opportunities created between April and June 2026, with a 30 June snapshot. Referrals has eight opportunities and four wins; Outbound has 12 and five. The extract contains a duplicate copy, open opportunities and one unconfirmed status.
- Starting point
- A report compares only won opportunities: four versus five. Its chart starts at 3.5, making the difference look much larger.
- First reading
- The initial commentary claims that Outbound is better and should receive more budget, without explaining the number of opportunities per channel or which remain open.
- What we review
- We return to the rows and period, remove the proven copy and retain the unknown status. We compare four of six known closed outcomes with five of ten, and distinguish that rate from four of eight and five of 12 across all observed opportunities.
- Result
- The table retains counts, denominators and pending information; the count chart starts at zero again. The data show differences in this sample but do not prove that the channel caused them. Before reallocating budget, the missing status and follow-up of open opportunities need checking.
Check before drawing conclusions.
More wins, with a different comparison base
Twenty unique opportunities created in Q2 2026. Snapshot: 30 June 2026. Open and unknown statuses are retained.
| Channel | Unique total | Won | Lost | Open | Unknown status | Won / known closed |
|---|---|---|---|---|---|---|
| Referrals | 8 | 4 | 2 | 1 | 1 | 4/6 · 66.7% |
| Outbound | 12 | 5 | 5 | 2 | 0 | 5/10 · 50% |
Swipe to see all columns.
Truncated scale · for review
Axis starts at 3.5 · wins
Scale starts at zero
Axis starts at 0 · wins
How it is calculated
- Known closed outcomes = won + lost: six in Referrals and ten in Outbound.
- Across all observed opportunities, wins are 4/8 = 50% and 5/12 = 41.7%. That base answers a different question.
- The original CSV contains 21 rows: O03 occurs twice with identical values. Removing that copy leaves 20 opportunities, not 21.
Outbound contributes more wins at this snapshot but also more opportunities. The count alone does not demonstrate a better channel. First clarify the unknown status and review open opportunities.
What we cannot conclude
- A small sample with no random allocation between channels.
- One unknown status and three open opportunities; results may change.
- The channel is not presented as the cause of the outcomes.
A higher average with a different work mix
Completed requests across two weeks, with recorded working time in hours. The second week has one missing measurement.
| Week | Type | Time available / completed | Observed mean |
|---|---|---|---|
| 1 | Simple | 4/4 | 1.5 h |
| 1 | Complex | 2/2 | 9 h |
| 2 | Simple | 2/2 | 1 h |
| 2 | Complex | 4/5 | 8 h |
Swipe to see all columns.
How it is calculated
- Week 1: (1+2+1+2+8+10)/6 = 4 h.
- Week 2: (1+1+7+8+8+9)/6 = 5.67 h across six available measurements from seven completed requests.
- Observed means fall within each type while the share of complex work rises. One complex request in week 2 has no recorded time.
The known overall mean is higher in week 2 even though means within each type are lower. The mix and missing time prevent concluding that the team deteriorated or that a new working method caused an improvement.
What we cannot conclude
- These are recorded working hours, not elapsed time or SLA compliance.
- An observational comparison of two weeks and a few requests.
- The week 2 mean omits one unknown measurement and must state that.
Before calling it complete
- The extract contains 20 unique opportunities after removing one identical copy.
- Referrals: 4/6 = 66.7% of known closed outcomes; Outbound: 5/10 = 50%. Each percentage states its base.
- The three open opportunities and the unconfirmed status are neither converted into losses nor hidden.
Four sessions to put it into practice.
Reference edition · 8 h live
Reference edition: eight live hours across four two-hour sessions. The sales case, operational case, individual variation and feedback take place during sessions. The tool and examples are adapted to the group without adding mandatory work between meetings.
- Session 12 h
Define the question and prepare the source
Agree the entity, period, snapshot and status definitions; review the sales rows and retain data-quality decisions.
By the endA focused question, short dictionary and 20 unique opportunities with the unknown status visible.
- Session 22 h
Compare channels without switching the base
Calculate counts and rates, distinguish their denominators and correct a display that exaggerates a difference.
By the endA verifiable sales comparison with bases and open or unknown outcomes identified.
- Session 32 h
Read the operational process
Calculate times only where measured, group by request type and review the changing work mix between weeks.
By the endAn explanation that distinguishes the observed average, work composition and missing time measurement.
- Session 42 h
Defend a decision and review another sample
Prepare the final note, complete the individual variation and check calculations and limits with trainer feedback.
By the endA traceable recommendation and reviewed individual practice with a defined next check.
If your team needs to go further
- Dashboards, BI models, SQL queries, forecasts and advanced statistical analysis require a different scope.
- Cleaning large volumes of real data, connecting systems or implementing company reporting is agreed separately.
An individual piece of work to close.
Review the individual delivery dataset, define the rate, show missing outcomes and write a decision with its next check.
Calculation and base
- Pass
- The calculation is reproducible from records and states its period, unit, numerator and denominator.
- Needs review
- Counts the copy, changes the base without stating it or treats unknown data as zero.
Presentation and context
- Pass
- Units, bases and missing values are visible; the comparison retains relevant group differences.
- Needs review
- Exaggerates the difference through its scale or interprets the overall average without reviewing the mix.
Decision and limits
- Pass
- Separates the observation from a possible explanation and leaves a specific check before widening the decision.
- Needs review
- Claims causation, a reliable forecast or an improvement not demonstrated by the sample.
The trainer reviews the task and explains the result within the agreed course time. Additional reassessment is agreed for each edition.
Put sales and operational information in context.
Juan Pedro Belmonte
GTM consultant
Over four years of experience developing sales strategies, CRM and lead generation. He has worked with companies across sizes and industries.
Juan Pedro is the planned trainer. Before an edition we confirm practice preparation, the tool, dates, language and his availability.
The data and a worksheet to check them.
Three fictional datasets, definitions, a blank worksheet and worked examples for practice and calculation checks.
- Working guide
- Sales opportunities
- Operational requests
- Individual variation
- Definitions and reading rules
- Blank analysis worksheet
- Checks and worked examples
- Individual practice and rubric
With guidance, practice and support.
Each delivery includes working materials, feedback on practical work and assessment against the agreed objectives.
Materials you can keep
Downloadable, versioned guides, exercises and review criteria for internal use.
To prepare and complete the delivery
A short recommendation, an editable communications kit and a group learning report.
TOF’s own certificate
A certificate of achievement after the agreed assessment. It is not a vendor certification.
A month of email support
On the content covered, up to two accumulated hours for the whole group.
Your team carries out the practice. Implementation services are agreed separately. See the scope →