An indicator tracking table that survives the quarter.
Fifteen columns, in the order a reviewer reads them, with the two most-forgotten ones included: per cent of programme time elapsed, and the deviation narrative you owe for anything more than ten points off.
Download it
| IPTT | The blank tracking table, fifteen columns, one row per indicator. |
| Example | Seven indicators from a smallholder programme, mid-implementation. |
| Guide | What each column is for, and the two that reviewers query most. |
Free to use, copy and modify, with or without attribution. This is Karagem's template, laid out to match what donors commonly ask for — it is not an official document from USAID, the EU, the World Bank or FCDO, so check your own award papers for the exact format your agreement requires.
Per cent achieved is half an answer
An indicator at 32% is fine in month four and a crisis in month thirty-two, and the number alone cannot tell you which. Every serious reviewer asks the same follow-up: how much of the programme has already gone?
So the table has a column for it. Thirty-two per cent achieved against sixty-seven per cent of time elapsed is thirty-five points behind pace, and that is the sentence the deviation narrative has to address. Without the elapsed column, every reader does that subtraction in their head and half of them get it wrong.
The ten per cent rule
PPR-style reporting expects a written explanation for any indicator more than ten per cent off its target. This is the column that gets filled in at midnight, and it is the one a reviewer actually reads.
A good deviation narrative names the cause, says what changes, and gives a revised expectation. "Delays in procurement" is not a cause, it is a category. "Seed arrived six weeks after the planting window, so adoption for the long rains is lost; the revised expectation is 2,100 by December" is a cause, a consequence and a number.
Disaggregation is a column, not an annex
Sex- and age-disaggregated data is required by most donors and produced by almost every survey, and it still routinely ends up in a separate tab nobody opens. Put the disaggregation the indicator is reported against in the row itself, so that a total which cannot be broken down is visibly a problem rather than a quiet omission.
"We reached 2,310" and "we reached 2,310, of whom 1,090 were women" are different claims, and only one of them survives the next question.
The other two
Logframe template
Four levels, seven columns, and a worked example from a real-shaped agriculture programme so the empty grid is not the first thing you meet. Free, no email wall, yours to modify.
Data quality assessment checklist
Fifteen questions across the five dimensions, each paired with the specific document or record an assessor will ask to see. Use it as a self-assessment a quarter before the real one, when there is still time to fix what it finds.