Using the System — A UI Walkthrough

This page is the literal click-by-click guide to running HCPI in the Odoo web interface: which menu, which form, which button label, in what order. If How the Calculations Work tells you what the system computes, this page tells you how to make it compute.

It assumes a fresh, empty HCPI instance. If you already have a running country instance, skip Phase 0 (setup) and jump to Phase 2 — Collecting prices (monthly).

The HCPI navbar at a glance

When you log into Odoo and open HCPI, the top navbar shows these root menus, left to right (in sequence order):

Menu	What it's for
Dashboard	Landing page with key index summaries
Data Collection	Field questionnaires (hcpi.data.collection) and validation batches (hcpi.data.validation)
Indices	Index results, the seven computation wizards (numbered 1️⃣–7️⃣), and computation history
COICOP	The product-classification ladder: Divisions → Groups → Classes → Sub Classes → Micro Classes → Elementary Aggregates
Items	The item catalogue and Units of Measure
Outlets	Outlets, outlet items, observations, consumption segments, and EA weights

The monthly working rhythm only really uses Data Collection and Indices. The other three menus are mostly one-time configuration that you barely touch after the first setup.

---

Phase 0 — One-time setup (a fresh instance)

You only do this once when standing up a new country instance. Order matters: each step depends on the one above.

0.1 Build the COICOP classification ladder

Menu: COICOP

Build it top-down, in this exact order — each level points to its parent:

1. COICOP → Divisions → click New → fill in COICOP code and name (e.g. "01 — Food and non-alcoholic beverages") → Save.
2. COICOP → Groups → New → pick the Division → Save.
3. COICOP → Classes → New → pick the Group → Save.
4. COICOP → Sub Classes → New → pick the Class → Save.
5. COICOP → Micro Classes → New → pick the Sub Class → Save.
6. COICOP → Elementary Aggregates → New → pick the Micro Class → Save. (This is the rung where the index actually starts being computed — see Stage 4 of the calculations page.)

0.2 Define units of measure and items

Menu: Items

1. Items → Units of Measure → New → for each unit you'll record prices in (kg, litre, piece, …).
2. Items → Items → New. For each item: name (e.g. "White bread loaf, 500 g"), link to its Elementary Aggregate and Micro Class, set the standard quantity and standard unit.

The link to the Elementary Aggregate is the chain that, weeks later, will route the item's price up the right COICOP ladder.

0.3 Define consumption segments (the "baskets")

Menu: Outlets → Consumption Segments

Click New, name the segment (e.g. "Kampala Urban", "Rural North"), save. You will create one segment per population stratum — each one has its own weights and produces its own basket-level index.

0.4 Define outlets

Menu: Outlets

1. Outlets → Configuration → Outlet Types → New → categories like "Supermarket", "Open market stall", "Pharmacy". (Reusable across all outlets.)
2. Outlets → Outlets → New → name, address, outlet type, and the consumption segment it belongs to.
3. Open the outlet form. In the header, click "Add Items" (header button, visible to users in the Outlet Manager group). This opens a wizard that lets you link multiple items at once to this outlet and record each item's base price (the reference-period price each long-term relative is measured against).

After this, each link is an hcpi.outlet.item record — visible under Outlets → Outlet Items.

0.5 Enter EA weights

Menu: Outlets → Elementary Aggregate Weights

Click New, pick an Elementary Aggregate and a Consumption Segment, type in the weight (the household expenditure share of that EA in that basket), save. Repeat for every (EA × segment) pair.

Weights are the only thing stored manually above the EA

Every higher-level weight (class, group, division, basket) is the sum of the EA weights beneath it — computed automatically. You never enter a class weight or a division weight by hand.

---

Phase 1 — Designing the monthly questionnaire (optional helpers)

Before enumerators go into the field, you may want to bulk-generate or assign their questionnaires instead of clicking "New" on every single one.

Menu: Data Collection

1.1 Bulk Questionnaires wizard

Menu: Data Collection → Bulk Questionnaires — opens a wizard (model bulk.questionnaire.wizard) in a popup.

What you do on the form:

1. Pick the Consumption Segment (field consumption_segment_id).
2. Pick the Collection Date for the month (field collection_date).
3. Click "Create Questionnaires" (the green highlighted button at the bottom of the popup; calls create_bulk_questionnaire). It creates one hcpi.data.collection draft per outlet in that segment for that date.
4. Or click "Discard" to close without creating anything.

The created records appear under Data Collection → Data Collection, all in draft state, ready for an enumerator.

1.2 Assign Questionnaires wizard

Menu: Data Collection → Assign Questionnaires — opens a wizard (model assign.questionnaire.wizard) that bulk-assigns the Data Collector and Supervisor users on a selection of questionnaires.

What you do on the form:

1. In the Data Collections field (many2many tags, field data_collection_ids), pick the questionnaires you want to assign.
2. Pick the Data Collector (field data_collector_id) and Data Supervisor (field data_supervisor_id).
3. Click "Assign Questionnaires" (the green button at the bottom of the popup; calls assign_bulk_questionnaire). It writes those two users onto every selected hcpi.data.collection record.
4. Or click "Discard" to cancel.

If your team creates questionnaires one at a time on phones (via the mobile app in hcpi_data_collection_mobile_app), you can skip both wizards.

---

Phase 2 — Collecting prices (monthly)

This is what enumerators (and supervisors) do at the start of every month — turning field visits into draft observations.

Menu: Data Collection → Data Collection

The record you're editing is an hcpi.data.collection — one questionnaire per outlet per month. It walks through five states, and the header of the form shows different buttons depending on which state it's in. Here is the complete journey, click-by-click.

2.1 Create the questionnaire (state: none → draft)

1. Click New in the list-view toolbar (top-left).
2. Fill in the form: Outlet, Collection Date, Consumption Segment, Data Collector (the enumerator), and Supervisor.
3. Click the cloud / save icon (top-left of the form) to save.

The record is now in draft state — visible at the top of the form's status bar.

2.2 Enter the price lines (still draft)

Open the "Observations" tab (the notebook tab near the bottom of the form). For each item priced at this outlet today, add a line:

• Outlet Item — picks one of the items linked to this outlet (and through it, the COICOP classification and the base price).
• Observed Price — what the enumerator saw on the shelf or shop counter.
• Observed Quantity and Unit — only set if different from the item's standard.

Each line is an hcpi.data.collection.line record. Save the form again after adding lines.

2.3 Approve the questionnaire (draft → survey)

Click the green "Approve" button in the form header (visible only while state = draft; calls approve()).

This marks the questionnaire as ready for collection / further processing. The status bar advances to survey. If you decided not to use it, click the red "Cancel" button (also visible only while state = draft; calls cancel()) — that moves it to cancel state, where the "Reset to draft" button can revive it.

2.4 Standardize the prices (survey → standardization)

Click the green "Standardize" button in the header (visible only while state = survey; calls standardize()).

This walks every collection line and computes the standard price — the observed price normalised to the item's standard quantity and unit. So if the enumerator priced rice as "4,200 UGX per 1.5 kg" but the item's standard is "per 1 kg", this step produces standard_price = 2,800 UGX/kg. That standard_price is what every later calculation uses.

The status bar advances to standardization.

2.5 Validate the prices (standardization → validation)

Click the green "Validate Prices" button in the header (visible only while state = standardization; calls validate()).

This is the handover from the collection workflow into the validation workflow. It groups every collection line for the same consumption segment × month into a single hcpi.data.validation batch — the record you'll open in Phase 3 to run the Tukey outlier check.

The status bar advances to validation, and a new stat button "Validations" appears at the top-right of the form (calls action_open_validation()) — click it to jump straight to the validation batch this questionnaire belongs to.

If something is wrong with the standardized prices, two recovery buttons exist:

• "Reject" (red, visible only while state = standardization; calls action_open_reject_view()) — opens a popup to enter a reject reason, and routes the questionnaire back for re-collection.
• "Reset to Standardization" (visible only while state = validation; calls reset_to_standardize()) — lets a supervisor walk it back to the standardization step to re-enter prices.

2.6 The full button reference for the data collection form

Button	Visible when state =	Method called	What it does
"Approve"	draft	approve()	draft → survey
"Cancel"	draft	cancel()	draft → cancel
"Standardize"	survey	standardize()	Computes standard_price on every line; survey → standardization
"Validate Prices"	standardization	validate()	Groups lines into a hcpi.data.validation; standardization → validation
"Reject"	standardization	action_open_reject_view()	Opens reject-reason popup; routes back to standardization
"Reset to Standardization"	validation	reset_to_standardize()	validation → standardization
"Reset to draft"	cancel	reset_to_draft()	cancel → draft
Stat button: "Validations"	always (if any exist)	action_open_validation()	Jumps to the related hcpi.data.validation batch

flowchart LR
    D["draft"] -->|"Approve"| S["survey"]
    S -->|"Standardize"| ST["standardization"]
    ST -->|"Validate Prices"| V["validation"]
    V --> DN["done"]
    D -.->|"Cancel"| C["cancel"]
    C -.->|"Reset to draft"| D
    ST -.->|"Reject"| ST2["(reject popup)"]
    V -.->|"Reset to Standardization"| ST

---

Phase 3 — Validating prices (the Tukey check)

This is what supervisors do to clean the month's prices before they become permanent observations. One hcpi.data.validation record covers one consumption segment × month and walks through four states. Each header button drives one transition.

Menu: Data Collection → Data Validations

Open a validation record (the one auto-created when an enumerator clicked "Validate Prices" in Phase 2). State is draft.

3.1 Run the Tukey outlier check (draft → validation)

Click the green "Validate" button in the form header (visible only while state = draft; calls action_validate()).

Behind the scenes this runs run_turkey() — the Tukey outlier algorithm described in Stage 3 of the calculations page. It computes a price relative for every line, trims the top and bottom 5%, then flags anything outside gen_avg ± 2.5·(side_avg − gen_avg) as is_outlier.

The status bar advances to validation. Now you can see what got flagged:

• Stat button "Outliers" at the top-right of the form (calls action_open_outliers()) — opens a filtered list of every line where is_outlier = True. Walk this list and untick is_outlier on any price you believe is real (e.g. a genuine market shock).

3.2 Run the inlier check (validation → validate_inliers)

Click the green "Validate Inliers" button in the header (visible only while state = validation; calls action_validate_inliers()).

This runs validate_inliers() — it flags any line where the price has been identical for the last 12 months. Persistent non-movement is just as suspicious as a wild jump (often it's a copy-pasted or forgotten price). The flag goes onto is_inlier.

The status bar advances to validate_inliers. Now review:

• Stat button "Inliers" at the top-right (calls action_open_inliers()) — opens a filtered list of every line where is_inlier = True. Walk it and untick is_inlier on any price you believe is genuinely unchanged (e.g. a regulated tariff).

3.3 Fill the gaps with imputation (validate_inliers → imputation → done)

Click the green "Impute" button in the header (visible only while state = validate_inliers; calls action_impute()).

Before it runs, Odoo pops up a confirmation dialog with the exact text: "Please note that this action confirms the current inliers and outliers before proceeding with imputation." Click OK to proceed. This is the point of no return — after this, the outlier/inlier flags are locked in.

Imputation then climbs the COICOP ladder to fill any missing prices, calling in turn:

1. impute_with_prices() — same product in another outlet
2. impute_with_ea() — other items in the same elementary aggregate
3. impute_with_micro_class() → _sub_class() → _class() → … — keeps climbing until it finds usable data

Each imputed line is marked is_imputed with the method recorded for audit.

The status bar advances through imputation to done. The surviving prices are now permanent hcpi.outlet.item.observation records — visible under Outlets → Price Data.

Review what was imputed via the stat button "Imputed Prices" at the top-right (calls action_open_imputed()).

3.4 Recovery — "Reset to Draft"

If you discover a problem after starting validation, the "Reset to Draft" button (visible whenever state ≠ draft; calls reset_to_draft()) walks the record all the way back to draft so you can re-run "Validate" from scratch.

3.5 The full button reference for the data validation form

Button	Visible when state =	Method called	What it does
"Validate"	draft	action_validate() → runs run_turkey()	Tukey outlier check; flags is_outlier; draft → validation
"Validate Inliers"	validation	action_validate_inliers() → runs validate_inliers()	Flags is_inlier on 12-month-flat prices; validation → validate_inliers
"Impute"	validate_inliers	action_impute() → runs the impute_with_*() ladder	Fills gaps; confirms outliers/inliers; validate_inliers → imputation → done
"Reset to Draft"	any non-draft state	reset_to_draft()	Restarts the validation workflow
Stat button: "Inliers"	always	action_open_inliers()	Filtered list of flagged inliers
Stat button: "Outliers"	always	action_open_outliers()	Filtered list of flagged outliers
Stat button: "Imputed Prices"	always	action_open_imputed()	Filtered list of imputed lines

flowchart LR
    D["draft"] -->|"Validate"<br/>(run_turkey)| V["validation"]
    V -->|"Validate Inliers"<br/>(validate_inliers)| VI["validate_inliers"]
    VI -->|"Impute"<br/>(impute_with_*)| I["imputation"]
    I --> DN["done"]
    V -.->|"Reset to Draft"| D
    VI -.->|"Reset to Draft"| D
    I -.->|"Reset to Draft"| D

---

Phase 4 — Computing the indices (monthly, strictly in order)

This is the heart of the monthly cycle. Each wizard is its own menu entry under Indices → Computations, numbered 1️⃣ to 7️⃣. Each one reads from the level below — do not skip ahead.

Menu: Indices → Computations

flowchart TD
    A["1️⃣ Update EA Indices<br/><i>geometric mean (Jevons) per EA × segment × month</i>"] --> B["2️⃣ Update Basket COICOP Indices<br/><i>weighted mean: micro → sub → class → group → division</i>"]
    B --> C["3️⃣ Update Basket Indices<br/><i>headline per consumption segment</i>"]
    C --> D["4️⃣ Update National Indices<br/><i>weighted across all baskets</i>"]
    A -.-> E["5️⃣ Update Special Indices<br/><i>custom EA groups: Energy, Core…</i>"]
    B -.-> F["6️⃣ Update Classification Indices<br/><i>custom class groups: Food/Non-food…</i>"]
    C -.-> G["7️⃣ Update Domestic Indices<br/><i>regional subsets</i>"]

4.1 What every "Update X Indices" wizard looks like

All seven wizards share the same form pattern. Knowing it once means you know all of them. When you click any of the seven menu items, Odoo opens a fresh wizard record (a new draft) with this layout:

• Form title at the top (e.g. "Update Elementary Aggregate Indices", "Update Basket COICOP Indices", "Basket Index Update", "National Index Update").
• Status bar at the top of the form: draft → processing → done (the EA wizard also shows cancelling, cancelled, failed).
• No filter fields. You will not see a date-range picker or a segment selector — these wizards process all data the database has. The only field shown is "Last Update" (create_date).
• A blue info alert in the form body while state = draft: "Ready to update X Indices — Click 'Start Update' button above to begin processing."
• An orange warning alert if the zero-price gate has flagged contaminated months — see 4.4 below.
• Two header buttons (depending on state):
  • "Start Update" — green, highlighted, visible only while state = draft and the zero-price gate is satisfied.
  • "Cancel" — grey, visible only while state = processing (and only on the EA, Special, and Classification wizards — see 4.3).

The flow on each form is therefore identical:

1. Open the menu entry → fresh wizard record appears in draft.
2. Read the "Ready to update…" alert. Verify there is no zero-price warning.
3. Click "Start Update" (top of the form).
4. The form stays open and the alert turns orange: "⏳ Update in Progress…" with a live progress bar, status message, "Records: X / Y" counter, and "Estimated Time Remaining". The EA wizard also notes: "💡 Detailed nested progress is available in the system tray (top-right corner). Click the notification icon to see real-time updates."
5. When the computation finishes, the alert turns green: "✓ Update Completed Successfully! All X indices have been updated." with a "Total Records Processed" count. State is now done.
6. If something goes wrong, the alert turns red: "✗ Update Failed" with a progress_message and the line "Please check the logs for more information." State is failed.

4.2 The seven wizards — menu, model, button method

Step	Menu (under Indices → Computations)	Wizard model	"Start Update" calls	Form title
1️⃣	Update EA Indices	hcpi.ea.index.update	action_update_ea_indices()	"Update Elementary Aggregate Indices"
2️⃣	Update Basket COICOP Indices	hcpi.basket.coicop.index.update	action_update_basket_coicop_indices()	"Update Basket COICOP Indices"
3️⃣	Update Basket Indices	hcpi.basket.index.update	action_update_basket_indices()	"Basket Index Update"
4️⃣	Update National Indices	hcpi.national.index.update	action_update_national_indices()	"National Index Update"
5️⃣	Update Special Indices	hcpi.special.index.update	action_update_special_indices()	"Special Index Update"
6️⃣	Update Classification Indices	hcpi.classification.index.update	action_update_classification_indices()	"Classification Index Update"
7️⃣	Update Domestic Indices	hcpi.domestic.index.update	action_update_domestic_indices()	"Domestic Index Update"

Steps 1–4 are mandatory and must run in that order. Steps 5–7 are optional side cuts that can run any time after their dependency:

• 5️⃣ Special needs EA indices (step 1).
• 6️⃣ Classification needs Basket COICOP indices (step 2).
• 7️⃣ Domestic needs Basket indices (step 3).

4.3 Cancellation — which wizards can be stopped?

Only three of the seven wizards expose a "Cancel" button (visible only while state = processing; calls action_cancel() with the confirmation prompt "Cancel this update? Partial results may remain."):

• ✓ EA Index Update
• ✓ Special Index Update
• ✓ Classification Index Update

The other four (Basket COICOP, Basket all-items, National, Domestic) do not currently expose a Cancel button — once you click "Start Update", you have to wait for the run to finish or fail.

4.4 The zero-price safety gate

If any active observation has observed_price = 0 in a month the wizard would process, two things happen on the wizard form:

1. An orange warning alert appears in the form body, listing the contaminated months and sample items (rendered from the zero_price_warning_html field).
2. The "Start Update" button is hidden (invisible="zero_price_has_issues == True and zero_price_has_safe_months == False").

Fix the offending prices in Phase 2 / Phase 3, then return to the wizard — the button will re-appear. See The zero-price safety gate for why this rule exists.

4.5 Reviewing past runs — Computation History

Each wizard also has a sibling history menu entry under Indices → Computation History (e.g. "EA Index Update History", "Basket COICOP Update History"). These open a list view of every past wizard run, showing:

• Name — the wizard record's auto-generated name
• Created By — which user clicked "Start Update"
• Started At — create_date
• Progress % — progress_percent (as a progress bar)
• Processed Records / Total Records
• Progress message — last status message
• State — coloured badge (draft blue, processing orange, done green, failed red, cancelled muted)

Click any row to open that historical run's form.

4.6 What happens if you click a wizard out of order?

Each wizard checks that no other computation is already running before starting (mutual exclusion via a PostgreSQL advisory lock pg_try_advisory_xact_lock(31818, segment_id) on the EA wizard). But it also reads whatever indices are currently in the database at the level below — so if you skip step 2 and run step 3, the basket all-items index will be computed from stale COICOP indices, not freshly-recomputed ones. Technically valid, but reflects last month's roll-up.

Rule of thumb: always run the chain end-to-end (1️⃣ → 2️⃣ → 3️⃣ → 4️⃣), then the side cuts.

---

Phase 5 — Viewing the results

Once a computation finishes, the results are visible as read-only list views. The list-view toolbar gives you Odoo's standard generic buttons — Filters, Group By, Favorites, the search bar, and the export icon (top-right cog menu → "Export All"). There are no per-row workflow buttons on index list views; the indices are pure data records.

5.1 Where to view each index type

What you want to see	Menu path	Model
The headline national index by COICOP level	Indices → National Indices → National Indices	hcpi.national.index
Regional / sub-population indices	Indices → National Indices → Domestic Indices	hcpi.domestic.index
All-items index for one basket (e.g. "Kampala Urban")	Indices → Basket Indices	hcpi.basket.index
Drill-down by COICOP level for a basket	Indices → Basket COICOP Indices → Division / Group / Class / Sub Class / Micro Class / Elementary Aggregate Indices	hcpi.basket.{level}.index
Energy, Core, etc. (national level)	Indices → Special Indices → National Special Indices	hcpi.national.special.index
Energy, Core, etc. (per basket)	Indices → Special Indices → Basket Special Indices	hcpi.basket.special.index
Food/Non-food, Goods/Services (national level)	Indices → Classification Indices → National Classification Indices	hcpi.national.classification.index
Food/Non-food, Goods/Services (per basket)	Indices → Classification Indices → Basket Classification Indices	hcpi.basket.classification.index
Computation logs / history	Indices → Computation History → pick the index type	hcpi.{type}.index.update (list view)

The default grouping in most index list views is by consumption segment or COICOP level, so you can collapse/expand to see a single segment's trajectory month-by-month. Visible columns include date, readable_date, index, weight, and the parent COICOP / segment.

5.2 The Excel Export wizard — buttons and fields

Menu: Indices → Export Excel Workbook

Opens a wizard (model hcpi.index.export.wizard) in a popup titled "Export HCPI Excel Workbook". This is the only "doing" UI in Phase 5; everything else is just looking at data.

Filters — what to include (left column):

For each COICOP level you have a All <level>s checkbox and, if you uncheck it, a many2many tag selector to pick specific ones:

• All Divisions + Divisions picker (many2many)
• All Groups + Groups picker (filtered to chosen Divisions)
• All Classes + Classes picker
• All Sub Classes + Sub Classes picker
• All Micro Classes + Micro Classes picker
• All Elementary Aggregates + Elementary Aggregates picker
• Consumption Segments picker (many2many tags — always shown)

Filters — outlets and outlet items (right column):

• Outlets picker (many2many)
• Outlet Item Scope (field outlet_item_mode) — controls whether you select specific outlet items
• Outlet Items picker (only visible when scope = selected)
• Month Mode (field month_mode) — all exports every month present; selected exposes a checkbox list of months to pick from

Sheets — what tabs to include in the workbook:

• Include Prices Sheet (include_prices_sheet) — raw observed prices
• Include Price Relatives Sheet (include_pr_sheet) — with a sub-toggle Price Relative Type (long-term / short-term, visible only when this is checked)
• Include EAI Sheet (include_eai_sheet) — Elementary Aggregate Indices — with sub-toggle EAI Index Type (long-term / short-term)
• Include Hierarchy Indices Sheets (include_hierarchy_indices_sheets) — adds one tab per COICOP level (micro-class, sub-class, class, group, division) plus basket all-items

Footer buttons (bottom of the popup):

Button	Method called	What it does
"Generate Excel"	action_generate_workbook	Builds the .xlsx workbook with the chosen sheets / filters / months and writes the filename into the output_filename field. After generation the form re-shows with a download link.
"Close"	(special: cancel)	Closes the popup without generating anything.

After clicking "Generate Excel", the popup re-renders showing the Output Filename at the bottom — click it to download the file.

---

The monthly cycle, in one screen

Print this and stick it on the wall:

┌──────────────────────────────────────────────────────────────────┐
│  MONTHLY HCPI WORKFLOW                                           │
├──────────────────────────────────────────────────────────────────┤
│                                                                  │
│  1. Enumerators                                                  │
│     Data Collection → Data Collection                            │
│     New → fill lines → Approve → Standardize → Validate Prices   │
│                                                                  │
│  2. Supervisors                                                  │
│     Data Collection → Data Validations                           │
│     Open → Validate (Tukey) → review Outliers/Inliers            │
│     → Validate Inliers → Impute                                  │
│                                                                  │
│  3. Statistician (in strict order)                               │
│     Indices → Computations →                                     │
│       1️⃣ Update EA Indices       → Start Update                  │
│       2️⃣ Update Basket COICOP    → Start Update                  │
│       3️⃣ Update Basket Indices   → Start Update                  │
│       4️⃣ Update National         → Start Update                  │
│       (5️⃣ Special, 6️⃣ Classification, 7️⃣ Domestic as needed)     │
│                                                                  │
│  4. View results                                                 │
│     Indices → National Indices → National Indices                │
│     Indices → Export Excel Workbook                              │
│                                                                  │
└──────────────────────────────────────────────────────────────────┘

---

When something goes wrong

Symptom	Where to look	What to do
"Start Update" button is greyed out on a computation wizard	The zero-price banner at the top of the wizard form	Find the contaminated observations, fix them in Phase 2/3, then return
Tukey threw out a price you think is real	Data Collection → Data Validations → open the validation → Outliers stat button → review	Untick is_outlier, then re-run "Impute" so the price is included
Imputation filled a gap you'd rather leave empty	Same place → Imputed Prices stat button	Untick is_imputed, then re-run the affected index update
Basket index looks wrong	Indices → Basket COICOP Indices → Elementary Aggregate Indices	Inspect the underlying EA indices and their diagnostics (recorded by _job_compute_indices_for_segment — see Stage 4)
A computation hangs in processing	Indices → Computation History	Check the history record for error messages; if EA/Special/Classification, click "Cancel" in the wizard form
Wizard won't start: "another computation is running"	Indices → Computation History	Wait for the running one to finish, or check whether a stale job needs to be cancelled

---

Where to go next

• How the Calculations Work — the maths behind every button you just clicked.
• Data Flow Walkthrough — a narrative trace of one price observation through the full system, with model names.
• CPI Concepts — the building blocks (basket, weights, base period) with the system's vocabulary.