How does it work? A step-by-step guide

There are two options for how you can start using Logframe Lab:

With document: You can upload your proposal, and our machine learning models will get to work. The upload has to be a MS Word file (.doc/.docx). You can upload it in the middle of the screen (see purple arrow below) by dragging the file there, or clicking in the area and selecting the document from your computer.

If you do not want to see all the indicators that could be relevant to you, you can use the filters to limit the results of indicators presented to you.

Just make sure to select the filters before you upload a document, as the upload will directly move you to the next step.

Without document: You can also start the whole process and use Logframe Lab without uploading a document. Just click on NEXT on the first screen (see screenshot above). Or select some filters first, and then click NEXT. If you do not select any filters, you will get all indicators in our database.

We cannot check the vertical logic of your statements – meaning if there is a logical link between your output, outcome, and impact statements. We only check each statement by itself.

In this section, what happens really depends whether you have uploaded a document or not.

With document: Our machine learning model scans your document and searches for your results chain statements. These are listed and categorised in three different levels: outputs, outcomes, or impact.

What the different columns mean:

Level: This is the level of your statement. It's important to get this right, as you will later be able to connect indicators to statements, and visualise your statements. All this is based on the level of each statement.

Statement: Here is what is written in your actual statement. We are constantly working on our machine learning model, and hope it has found the correct statements. If not, you can edit the statements here as you wish.

Status: The score shows you the quality of your statement in a bit more detail. To make things easy, you only have two options, good and bad. You can filter here, and only see your bad statements for example (if you want to improve them).

Score: The score shows you a bit more detailed the quality of your statement. Statements get the badge "bad" in the previous column, when they have a score of 30% or less.

Action: You have three different action options here. The first one (purple circle with a +) lets you add a new statement. This will always be added to the end of the list, no matter which level you give it. Don't worry about that. The next one (green circle with a tick) checks the quality of your statement. It activates the machine learning model that checks the quality, and you can let it run as many times as you want. When you change something in the column "Statement", you need to activate the the check through this. The last option is the red circle with an x; this means you want to delete the statement.

Once you are happy with your statements (or you delete all, because you do not want to use this feature) click NEXT.

Without document: If you have not uploaded a document in the first step, the second step will look like this.

You can click on "Add Statement" to start this process, and then it will look like in the previous screenshot (and the rules for the columns are the very same as above). You can also just skip this step, and move to the indicators directly.

Outputs are the direct and tangible good or services a project/intervention provides. They are under the direct control of the project/intervention (e.g., deliver literacy lessons, build water wells, or distribute hygiene articles);

Outcomes are the effects and changes produced by your outputs, visible in the short-term but under the control of the target group (e.g., fewer illiterate persons, access to clean water, or more hygiene);

Impact is the long-term change the project/intervention contributes to, e.g. more welfare or less health problems and deaths).

As before, it depends on if you have uploaded a document or not.

With document: This step is per se independent from the previous step, but you will have a chance to link them (if you have statements from the previous step). The indicators suggested to you here are based on your uploaded document.

Let's go through the different columns first (most columns, you can filter and sort in any way you want through the respective little icons next to the column titles).

First (unnamed): This will open a pop-up, where you can choose the baseline value for World Bank indicators, and can link outcome and output indicators to their relevant results statements (if you have any). For some indicators, we also have additional information that is shown here.

Second (unnamed): This allows you to select indicators, and shows you which ones are selected. You can also just click anywhere in the row to select indicators, and all selected indicators are coloured. Furthermore, we have provided you support with the semi-transparent row at the bottom of the screen. It only appears once you selected the first indicator, and then it shows you how many indicators you have selected. While you can select as many indicators as you want, more is not always better. So we have built in some warnings, for when we think you might have selected too many indicators. Clicking on "show all" in the bottom right corner shows you all indicators you have selected.

Level: This is the level of the indicator. We present all impact level ones first, then outcome, and then output. While this is also the alphabetical order, we have ordered them this way because this is how the logic of your logframe progresses from the top level.

Indicator: This is the text of the actual indicator, so for you, this is probably the most important column.

Source: The column tells you where we got the indicator from (i.e., its source). This might help if you want to check and see if your donor or organisation has relevant indicators.

Disaggregation: Here we tell you if we think the indicator needs to be disaggregated. As a rule of thumb, whenever you count people, you need to have at least a sex/gender disaggregation.

Sector: As you might not want to see all indicators, this is the sector we gave your indicators. Sometimes, an indicator may be in a "related" sector. For example, some indicators for nutrition programming might be found under the health sector.

OECD DAC/CRS Code: This is a code/system often used by donors. It classifies the indicators, similar to what the sector does. If you do not know what this is, or have never heard about it, don’t worry; just ignore it.

SDG Code: More and more projects/interventions want to support/work towards the achievement of the Sustainable Development Goals. This column lets you filter by SDG. While we have given every indicator only one sector, we have often linked it to more than one SDG.

Score: The score is based on our machine learning model, and tells you how good the indicator fits your project/intervention. The score goes up to 100, and the higher the score, the better. You can use the slider in the heading of this column to only see indicators with the score within a certain range. Indicators with a very low score are most likely not very relevant for your project/intervention.

You can check the indicators (search for indicators at the top) and select as many as you want. Once done, click on next.

If you have result statements, make sure you linked them accordingly (as this will make your logframe more organised in the download tab), and add baselines to World Bank indicators (not all indicators have data for all countries/years, but we let you know if we do not find anything).

Without document: If you have not uploaded a document, and you have not used any filters, it might take a minute to load the indicators, as we have several thousand in our database. The rules and columns as described above are the same, with the difference being that you won’t see the "Score" column at the end (as there is nothing we could base this on).

Other than the score, this works exactly the same way as if you have uploaded a document. Feel free to browse the indicators for inspirations, or build a logframe by choosing the ones relevant for you.

Once you are done with the indicators, you move to the visualisation page. This deals with the results statements (if you have any), and allows you to build a very basic theory of change of your project. It might be helpful, especially if you have several outcomes and outputs, to make sense of how they logically fit together.

Last but not least, you can download all the work you have during the last steps. You can choose from different download formats.

Here is a quick overview of the options:

EU format: This is the usual European Union Delegation logframe that most EU (not ECHO) funded projects use.

PRM format: This is the PRM file format (of the US Government).

Indicators Excel: These are just the indicators and data in a basic Excel format.

DFID format: This is the format used by DFID/FCO.

SVG/PNG: These are picture formats for the visualisation.

No matter which of the first 4 formats you choose, the data you selected/entered will be in here, in a good format, that allows you to finalise your logframe offline. Currently, targets and assumptions need to be entered "offline".

Download section: In the download section, you can choose which file format you want for your created logframe. Available formats include editable ones (Word & Excel) and non-editable image files (SVG & PNG). After you have downloaded the file format you want, you can click “Done”, which will delete all your information, and start a new logframe.

If you need more technical details you can go to the FAQ What happens with my proposal?