4.1Tables, filters, sorts & groups
The essentials in 30 seconds
- Tables define the displayed columns; filters show only what matters; sorts order the rows; groups categorize with subtotals: four complementary tools for querying a schedule.
- Combined, these tools form targeted analysis views — “late tasks, grouped by owner, sorted by size of delay” — far more telling than a raw schedule.
- These tools never change the data: they are reading angles (“glasses” you put on), the basis of all analysis and reporting.
Running mini-case — Meriem puts analytical glasses on her schedule
Meriem, project manager in an engineering & construction (BTP) company in Algiers, tracks several client projects; this morning, the steering meeting is approaching. Rather than scrolling through a Gantt chart of several hundred rows, she decides to query her schedule. She displays a table that highlights Tracking, applies a “Late tasks” filter, sorts by size of delay and then groups by owner to read subtotals per team lead. In a few seconds, she gets a telling analysis view: who is late, and by how much. She notes the reflex: these tools change nothing in the data, they are glasses; and if the display looks “broken”, it is often a filter left active.
Standalone case for this section: it returns in the course, the diagram, the script and the 48h challenge of section 4.1.
1. Four analysis tools
Project contains hundreds of fields; the art lies in showing only what serves the question of the moment. Tables define the visible columns (Entry, Cost, Tracking…). Filters show only the tasks or resources matching a criterion (“critical”, “late”). Sorts order the rows according to one or more fields. Groups gather rows by category with subtotals.
2. Targeted analysis views
Combined, these tools turn a raw schedule into analysis views: “all late tasks, grouped by owner, sorted by size of delay”. Meriem thus creates tailor-made reading angles, without ever touching a single piece of data — like glasses you put on as needed. This is the basis of all analysis and all reporting.
3. Filtering is not grouping
Two logics not to confuse: a filter reduces what you see (it hides rows outside the criterion); a group structures what you see (it reorganizes all rows into categories, without hiding any). They are often combined — filter “late” then group “by owner” — and useful combinations are saved as reusable custom views.
🔎 Concrete example — 🔎 Concrete example — Filter + group
Two tools combined for a telling view, as Meriem does:
Filter “Late tasks” -> reduces what you see.
Group “by owner” -> structures with subtotals.
-> combined: “late, by owner, by size”.Reading: the filter reduces the display, the group structures it into categories with subtotals.
Lesson: you move from “looking at” to “querying” your project, without changing a single piece of data.
Section vocabulary
- Table
- Set of columns displayed in the grid (Entry, Cost, Tracking…), customizable.
Simple example: Meriem displays the Tracking table for her steering meeting.
What it is not: a report exported to Excel.
Reflex: choose the columns → the table.
- Filter
- Criterion that hides the tasks or resources not concerned (critical, late, milestones…).
Simple example: Meriem applies the “Late tasks” filter.
What it is not: a group, which hides no row.
Reflex: show only what matters → the filter.
- Sort
- Ordering of rows according to one or more fields (date, cost, duration).
Simple example: Meriem sorts tasks by size of delay.
What it is not: a filter, which hides instead of ordering.
Reflex: rank the rows → the sort.
- Group
- Gathering of rows by category (owner, status, priority) with subtotals.
Simple example: Meriem groups tasks by owner.
What it is not: a filter, which removes rows from the view.
Reflex: categorize with subtotals → the group.
- Analysis view
- Combination of table + filter + sort + group offering a targeted reading angle on the project.
Simple example: Meriem gets “late, by owner, by size”.
What it is not: the raw schedule, unqueried.
Reflex: combine the tools → an analysis view.
In practice — In practice — Analyze the project with tables, filters and groups
- Change table (View > Tables) to display the relevant columns (e.g. Cost or Tracking table), as Meriem does.
- Apply a filter (View > Filter): “Critical tasks”, “Late tasks” or a custom filter.
- Add a sort (by finish date, by cost) to order the rows according to your analysis.
- Group tasks (View > Group by) by resource or status, then read the subtotals.
Step-by-step script — analyze like Meriem (4 min)
1) Open the project's Gantt view. 2) Change table (View > Tables > Tracking). 3) Apply the “Late tasks” filter. 4) Sort by size of delay (for example by finish variance). 5) Group by owner and read the subtotals per team lead. 6) Clear filter and group (View > No Filter / No Group) and note that the data is intact.
Save your useful combinations as custom views: they can be reused from one meeting to the next.
Copy-paste template — my analysis view recipe
Table : the columns that serve the question of the moment
Filter : keep only what matters (e.g. “Late tasks”)
Sort : order (e.g. by size of delay)
Group : categorize with subtotals (e.g. by owner)
Rule : these tools NEVER change the data (they are glasses).
Reflex: “broken” display -> often a filter left active.Useful columns
Meriem changes table to show only the Tracking columns before her meeting.
“Late” view
Meriem filters “Late tasks” to reduce the display to problematic tasks.
Subtotals by owner
Meriem groups by owner and reads the cumulative delay of each team lead.
Forgotten filter
Tasks are missing: Meriem clears the filter left active and recovers her whole schedule.
✓ Best practices
- Show only the useful columns for the question of the moment.
- Combine filter + group for telling views.
- Save useful combinations as custom views.
✗ Common pitfalls
- Leaving a filter active and believing the display is broken.
- Confusing filtering (reducing) and grouping (structuring).
- Drowning in too many columns.
Key takeaways
- Tables = columns; filters = show only what matters; sorts = order; groups = categorize.
- Combined, they create targeted analysis views (“late, by owner, by size”).
- These tools never change the data: they are reading angles.
- They are the basis of all effective analysis and reporting.
48h challenge — apply it tomorrow
Like Meriem, build an analysis view on your schedule: display a targeted table, apply a filter (“late” or “critical”), add a sort and a group by owner. Write in one sentence the question the view answers, then clear filter and group while checking that your data is intact.
- I create analysis views by combining table, filter, sort and group.
- I distinguish filtering (reducing) from grouping (structuring).
- I know that these tools never change the data.
Frequently asked questions
What is the difference between a filter and a group?
A filter hides rows outside the criterion: you only see the late tasks. A group hides nothing: it reorganizes all rows into categories with headers and subtotals (all tasks, by owner, with each one's total cost). The filter reduces what you see, the group structures what you see; they are often combined.
Are my view customizations lost when I change views?
No, they are remembered per view and per table. Filters, sorts and groups apply to the active view and can be cleared at any time (View > No Filter / No Group). You can also save a combination as a reusable custom view. Nothing alters the data: a “broken” display almost always comes from a filter left active.
What are tables, filters, sorts and groups used for?