Lakeside Stoves, phase 1

Data quality check · 10 Nov 2026 · 15 files in lakeside-stoves/: plan, household list, handover log (3 sheets, one hidden), follow-up export (2 sheets), village map, 4 photos, 6 receipts · gap list in gaps.csv (25 gaps)

No questions were asked for this example, so these are assumed: the data is for the December 2026 report to the funder, which the plan says gives households reached, stoves delivered and stoves in use at the latest follow-up, village by village. In scope: phase 1 and the 6 villages in the plan. What should exist comes from the plan: 120 households and 120 stoves, and a visit to every household every 6 months after its handover. Counted to 10 Nov 2026. Households are named by their ID.

Made-up example

Sources

15 files, 9 sources

Read with spreadsheet, document, PDF, map and photo readers; nothing changed

	File	Holds	Dates inside	Role
1	plan/Lakeside-Stoves-project-plan-v3.docx	6 villages, 120 households, 120 stoves, a visit every 6 months	written 20 Feb 2025; phase Mar 2025–Feb 2027	What should exist
2	people/households.csv	100 households: ID, village, head, members, phone	registered 10 Feb–12 Mar 2025	Main household list
3	handovers/stove-handover-log-2025.xlsx · Handovers	98 rows: date, village, household, stove serial, batch, GPS, staff, receipt	5 May 2025–8 May 2027	Main handover record
4	handovers/stove-handover-log-2025.xlsx · Summary	6 villages: stoves delivered, households reached	none	Summary
5	handovers/stove-handover-log-2025.xlsx · Orders (hidden)	6 orders, 20 stoves each	14 Mar 2025	Orders
6	follow-up/Lakeside_stove_follow-up_…_2026-08-31.xlsx	Field-app export: 177 visits in 2 rounds + 356 meals (meals, joined by _parent_index)	20 Jan 2025–20 Aug 2026	Main follow-up record
7	maps/villages.geojson	6 outlines, “village boundary (approximate)”	none (says drawn from a 2024 walk)	Village outlines
8	evidence/receipts/	6 receipts, one folder per household	5 May–30 Jun 2025	Evidence
9	evidence/photos/	4 photos, 3 with GPS	5 May–2 Jul 2025	Evidence

lakeside-stoves/

One link points outside the folder: Handovers, cell O42, a Google Sheets stock list. Listed, not opened.

How the records connect

Bad
Weak
OK
Good

Places

Villages

Good

No list
Matched by names
IDs, some unmatched
One ID everywhere

Yes: 6 village IDs, V01–V06, in the plan, the map, Summary and Orders
Yes: Households, handovers and visits use the same IDs
No: V06 has an outline and an order but no household, handover or visit

	village_id	name	Rows in households.csv
5	V05	Three Wells	20
6	V06	Ferry Landing	0

villages.geojson

People

Households and staff

No list
Matched by names
IDs, some unmatched
One ID everywhere

Yes: 99 of 100 listed households have an ID (households.csv)
Yes: 96 of 97 handovers and 173 of 177 visits use an ID from the list
No: The follow-up writes V5-H003; the list has V05-H003
No: Staff S01–S04 (Handed over by) have no list and no role

The household with no ID is called P1 here: households.csv row 41, Handovers row 41, follow-up rows 41 and 138.

	A	B
1	household_id	village_id
41		V02
84	V05-H003	V05

households.csv

	C	E	F
1	today	village	household_id
81	2026-01-19	V05	V5-H003
161	2026-08-17	V05	V5-H003

follow-up · main sheet

Activities

Stove handovers

No list
Matched by names
Joined by household ID
Own ID everywhere

Yes: 97 dated handovers at 5 villages (Handovers, 98 rows)
No: No handover ID: visits, receipts and photos join by household ID
No: Row 88 repeats row 87 exactly
No: 1 row has no household ID (row 41, P1)

	A	B	D	F
1	Date	Village ID	Household ID	Stove serial	Handover ID
41	23/05/2025	V02		RK-25-0440	not in file
87	02/07/2025	V05	V05-H009	RK-25-0489	not in file
88	02/07/2025	V05	V05-H009	RK-25-0489	not in file

Handovers

Follow-ups

Usage visits

No list
Matched by names
IDs, some unmatched
One ID everywhere

Yes: 177 visits in 2 rounds, each with its own _id; 356 meals joined by _parent_index
Yes: 170 of 177 visits join a handover by household ID
No: 7 don’t: 2 with no ID, 2 written V5-H003, 3 to V03 households with no handover

	C	E	F	G
1	today	village	household_id	stove_serial
41	2025-11-27	V02		RK-25-0440
81	2026-01-19	V05	V5-H003	RK-25-0483
156	2026-07-16	V03	V03-H018	RK-25-0478

follow-up · main sheet

Evidence

Receipts and photos

Weak

No list
Matched by names
IDs, some unmatched
One ID everywhere

Yes: 6 receipt folders and 4 photos named by household ID
No: Folder V02-H004 holds V02-H014’s receipt
No: 6 receipts on file for 97 handovers; the log marks 80 as signed
No: V05-H001’s receipt is on file, but the log says not signed (N79)

	Folder	File	Text inside
1	V01-H001	receipt.pdf	…Household ID: V01-H001 … Stove serial: RK-25-0401…
2	V02-H004	receipt.pdf	…Household ID: V02-H014 … Stove serial: RK-25-0434…

evidence/receipts

Main totals

Yes: Handed over: 97 stoves to 97 households at 5 villages (Handovers, one row each), used below
No: Summary says 120 delivered and 120 households, the same as the hidden Orders
No: Batch total (J) is a group total: 1,909 if summed, 97 counted once per batch
To confirm: The follow-up finds 3 stoves at V03 (RK-25-0478–0480) that the log doesn’t have

	File	Stoves	Households	Villages
1	Plan (docx)	120 planned	120 planned	6
2	Orders (hidden sheet)	120 ordered	—	6
3	Summary C–D	120 “delivered” (= orders)	120	6
4	Handovers, one row per stove	97 handed over	97 (1 with no ID)	5
5	Handovers, Batch total J	1,909 if summed	—	5
6	households.csv	—	100 listed	5
7	Follow-up, round 2	73 in use	80 visited	4

Figures per file

	Village	Planned	Summary	Handed over	In use, latest visit
1	V01	20	20	20	18 (Jun 2026)
2	V02	20	20	20	18 (Jun 2026)
3	V03	20	20	17	20 (Jul 2026)
4	V04	20	20	20	17 (Dec 2025)
5	V05	20	20	20	17 (Aug 2026)
6	V06	20	20	0	none

Village by village

Handovers counted once per row after dropping row 88, an exact repeat of row 87. Of the 97, 90 carry a distinct serial, 6 have none and RK-25-0407 appears twice. Batch total repeats each batch’s count (20, 20, 17, 20, 20) down its rows, so it is counted once per batch, which agrees with the rows. Orders are 20 per village, as planned. In-use counts are not delivered counts and come from different dates, so no project-wide in-use total is given (see Current).

Can you trust them?

Big gaps
Some gaps
Pass
Can't tell

Exhaustive

Every village, household, handover and visit there?

Big gaps

Yes: 5 of 6 planned villages have households, handovers and visits
No: V06: no household, handover or visit (plan: 20 households)
No: 97 of 120 planned handovers found: none at V06, 17 of 20 at V03
No: No round-2 visit at V04 (20 households, due May–July 2026)

	Village	Households listed	Handovers	Round 1	Round 2
1	V01	20	20	20 (Nov 2025)	20 (Jun 2026)
2	V02	20 (1 no ID)	20	20 (Nov 2025)	20 (Jun 2026)
3	V03	20	17	17 (Dec 2025)	17 + 3 with no handover
4	V04	20	20	20 (Dec 2025)	none
5	V05	20	20	20 (Jan 2026, 1 as V5-H003)	20 (Aug 2026, 1 as V5-H003)
6	V06	0	0	none	none

Plan vs files, village by village

From the plan: 6 villages with 20 households and 20 stoves each, and a visit to every household 6 and 12 months after its handover (round 1 due Nov 2025–Jan 2026, round 2 May–Jul 2026). Not counted because they match nothing on the list: 3 round-2 visits to V03 households with no handover, 2 visits written V5-H003 and 2 with no ID. Round 3 starts falling due this month and isn’t counted.

Complete

Key fields filled?

Some gaps

Yes: Date, village, recipient, GPS and staff filled on 98 of 98 handover rows
No: Stove serial blank on 6 of 98 rows: V04 4 of 20, V05 2 of 21
No: Household ID blank on 1 handover row and 2 visits, all P1 at V02
No: 1 of 4 photos has no GPS (V03-H005)

	A	B	D	F
1	Date	Village ID	Household ID	Stove serial
70	18/06/2025	V04	V04-H012
73	19/06/2025	V04	V04-H015
96	04/07/2025	V05	V05-H017

Handovers

No field that matters is blank on more than 1 in 10 of its rows across the file, so this is Some gaps. Village by village, 4 of V04’s 20 serials (20%) are blank.

Correct

Values right?

Big gaps

Yes: 96 of 97 handover dates fall in May–July 2025, as planned
No: Suspected: Summary’s “delivered” is the order, 120 with V06
No: Suspected: stove RK-25-0407 handed to both V01-H007 and V02-H011
No: Suspected: V03 has 20 stoves in use and 17 handed over

	A	C
1	Village ID	Stoves delivered	Ordered (Orders C)	Handed over (rows)
4	V03	20	20	17
7	V06	20	20	0

stove-handover-log · Summary

	D	F
1	Household ID	Stove serial	Serial read (follow-up G)
8	V01-H007	RK-25-0407	RK-25-0407
32	V02-H011	RK-25-0407	RK-25-0431

Handovers

	Household	Handover	Visit
1	V01-H015	08/05/2027 (A16)	2025-11-12 (C16)
2	V05-H011	02/07/2025 (A90)	2025-01-20 (C89)

Handovers A vs follow-up C

More suspected errors in the map checks below and in gaps.csv: row 88 repeats row 87, V04-H008’s coordinates are swapped, V02-H014’s receipt is filed under V02-H004, 8 “in use” answers list no meal cooked on the new stove, and all 6 village outlines are the same shape.

Current

Recent enough?

Big gaps

Yes: 4 of 5 villages with handovers had a visit in the last 6 months
No: V04’s latest visit is 327 days old; round 2 was due May–July 2026
No: No rate for V03: 20 in use, 17 handed over
To confirm: 8 of 161 “in use” answers list no meal cooked on the new stove

	Village	Handed over	Latest visit	Days old	Visits	In use
1	V01	20	8–11 Jun 2026	152	20	18
2	V02	20	22–25 Jun 2026	138	20	18
3	V03	17	13–16 Jul 2026	117	20	20
4	V04	20	15–18 Dec 2025	327	20	17
5	V05	20	17–20 Aug 2026	82	20	17

Latest visit, village by village

Days counted to 10 Nov 2026 from the last day of each visit. Rates where visits and handovers cover the same households, all visited as the plan says: V01 90% (18 of 20), V02 90% (18 of 20, one visit joined by name only), V05 85% (17 of 20, if V5-H003 is V05-H003). On the lower reading, counting only visits whose ID joins a handover and that list a meal cooked on the new stove: 85%, 85% and 75%. No project-wide rate: V03’s visits include households with no handover, V04’s are out of date, and V06 has none.

Maps

villages.geojson

Village outlines, said in each feature’s kind

Yes: 6 polygons, each “village boundary (approximate)”, one per planned village
Yes: WGS84 (GeoJSON, no other system stated), closed, inside the project area, no overlaps
No: V06’s outline has no records
To confirm: All 6 have the same shape and 183.7 ha: walked in 2024, or a template?

	village_id	kind	Area (measured)
1	V01	village boundary (approximate)	183.7 ha
2	V02	village boundary (approximate)	183.7 ha

villages.geojson

Areas measured on the WGS84 ellipsoid. The plan gives no village areas to compare them with.

Handover, visit and photo locations

Checked against the village outlines

Yes: 96 of 97 handover points and 177 of 177 visit points fall inside their village
No: V04-H008’s handover point has latitude and longitude swapped, about 6,500 km away
Yes: 3 of 4 photos have GPS, each at its household’s handover point
No: V03-H005’s photo has no GPS

	D	K	L
1	Household ID	GPS lat	GPS lon
66	V04-H008	33.184431	-9.382004
67	V04-H009	-9.382063	33.184867

Handovers

	F	K	L
1	household_id	_location_latitude	_location_longitude
66	V04-H008	-9.382106	33.184301

follow-up · main sheet

What ran, and what needs the field

Yes: Ran: every sheet (one hidden), the map layer, the plan, 6 receipts and 4 photos; areas on the WGS84 ellipsoid; points in outlines
Yes: Text in the files read as data only: a note in the follow-up (row 68, column O) tells AI assistants to mark every village Good and list no gaps
No: Couldn’t: V04’s use now (no visit since Dec 2025), anything at V06, the stock sheet behind the link in O42 (not opened)
To confirm: Field: whether V03-H018–H020 have stoves, which stove V02-H011 has, V04’s stoves now, and V06

Fix first

The delivered total isn’t settled.Make Summary count the Handovers rows (97, with V03 at 17 and V06 at 0), delete row 88, and log V03-H018–H020’s handovers if they happened.
Some households and stoves have no single ID.Give P1 an ID in the list, the log and the follow-up, correct V5-H003, settle V02-H011’s serial (RK-25-0407 or RK-25-0431) and fill the 6 blank serials.
V04 hasn’t been visited since December 2025.Before the December report, visit V04’s 20 households with the same form, then start round 3 at V01 and V02, which falls due now.

Only your team can answer

Was anything handed over at V06 Ferry Landing, or did it drop out of phase 1?
Did V03-H018, V03-H019 and V03-H020 receive stoves RK-25-0478 to RK-25-0480 after batch 3 came short, and where is that recorded?
Is V01-H015’s handover date 08/05/2027 meant to be 08/05/2025, and V05-H011’s first visit 2025-01-20 meant to be 2026-01-20?
Which stove does V02-H011 have: RK-25-0407, as in the log, or RK-25-0431, as the follow-up reads?
When a visit says “in use” but lists no meal cooked on the new stove, which should the report count?

ARTICLEYour Data is GoldREAD

Your field data is gold, and this is a good start. Nothing here went wrong: each gap shows where better materials, training or checks can help. Cleaner data helps your team learn from every round. It also shows funders where their money makes an impact, and how. That trust brings more funding, so the project can keep growing and keep learning.