✳ IskraAI

Shopping list: how to properly prepare data and check the result

The 'Shopping list' tool combines ingredients from multiple dishes into one grouped list with quantities where possible. Input is a textual description of dishes and a number of people or portions, output is a plain text list. The quality of the result directly depends on how carefully the source material is prepared. This guide covers practical techniques: how to formulate ingredients, what to do with units of measurement, how to specify portions and how to check the ready list. All examples are illustrative and are not real data.

Open tool: Shopping list maker ↗

Preparing source text: structure and completeness

The tool has no culinary logic: it processes only what is explicitly written in the 'Source material' field. If an ingredient is mentioned in passing, in parentheses or after a comma, it may be skipped. Therefore list each ingredient separately, better on a new line or via a semicolon, without nested constructions. For example, the phrase 'pasta (tomatoes, basil, garlic)' risks turning into one position or losing some components, while 'pasta: tomatoes 30 g; basil 20 g; garlic 10 g' gives a predictable result.

Avoid synonyms in one text: 'tomatoes' and 'tomatoes' may be perceived as different products, and two lines will appear in the list. The same applies to 'salt' and 'sea salt'. If you want positions to be combined, call them the same. Also do not overload the description with unnecessary adjectives: 'cherry tomatoes sweet 250 g' is better written as 'cherry tomatoes — 250 g', otherwise the whole phrase will end up in the list.

Remember the limits: the first text field accepts up to 8000 characters, the second up to 1000. If you insert too long a list of recipes, part of the data may be cut off and the shopping list will be incomplete. Better split a large request into several or shorten descriptions, leaving only ingredients and quantities.

Bad: 'Salad: cucumber, tomato (optional), greens'. Good: 'Salad: cucumber 100 g; tomato 150 g; greens 30 g'.

Units of measurement and quantities: how to avoid confusion

The most common reason for an inaccurate list is inconsistent units. Writing 'tomatoes 3' is ambiguous: three pieces, three grams or three packages? The tool will try to guess, but may err. If in one dish '200 g of tomatoes', and in another '2 tomatoes', combining may give meaningless '200 g + 2 pcs'. Bring all ingredients to a single system: grams, milliliters or pieces. If precision is not important, write 'to taste' or 'about 1 cup', but then the list will have an approximate number.

The tool does not convert units of measurement. '1 cup of rice' and '200 g of rice' will remain two separate lines, not summed up. If you want to get an exact amount, pre-translate volumes into weight or milliliters. For dry products this is especially important, as density may differ. Do not expect the tool to guess the correspondence.

After generation review the list for weird combinations, e.g. '3 pcs + 150 g'. If such exist, go back to the source text and unify units. The tool is not a culinary handbook, it merely combines what you gave.

Bad: 'pasta with tomatoes and basil; salad with tomatoes and cucumber'. Good: 'pasta: tomatoes 300 g, basil 20 g; salad: lentils 200 g, tomatoes 150 g, cucumber 100 g'.

Portions and people: how not to double the unnecessary

The 'People or portions' field expects a number. If you specify '2', the tool may multiply all ingredients by two, but only if in the source text the quantity is specified for one portion. If the recipe is already calculated for two, and you put '2', double multiplication will occur. Always check how many portions the original recipe is calculated for, and indicate the final number of people. Phrases like '2 adults and 1 child' or a range '2-3' are not interpreted: enter just '3'.

If a child eats less, it is better to manually adjust the quantity after generation. The tool does not consider age or appetite. Also do not count on it automatically recalculating spices 'to taste' — they will remain unchanged.

Check: after generation make sure quantities look reasonable. For example, for pasta for 4 people usually 400 g of dry pasta is needed, not 800 g. If proportions seem inflated, review the source data.

Source text: 'Pasta: macaroni 100 g, tomatoes 200 g (for 2 portions)'. Field 'People': 4. Result should double: macaroni 200 g, tomatoes 400 g. If the source does not say 'for 2 portions', but just 'macaroni 100 g', then with '4' it becomes 400 g — this is an error.

Grouping and expectations about the result

The tool groups ingredients by categories or simply merges identical positions. If you expect grouping by dishes, but get grouping by products, this is not a tool error, but mismatch of expectations. The description does not say how exactly to group, so the result may be arbitrary. To get a predictable list, you can try changing the order of ingredients in the source text: the tool may preserve the order of appearance. For example, list all vegetables first, then all grains. This does not guarantee, but may help.

Do not expect the tool to prioritize or highlight main ingredients. It outputs simple text without complex annotations. If you need comments like 'for pasta' or 'for salad', you will have to add them manually after generation. The tool also does not check product compatibility and does not offer substitutions.

If grouping is inconvenient, copy the result to a text editor and sort manually. The tool is not designed for perfect organization, it only helps collect all positions in one place.

Input: 'Pasta: tomatoes, basil; Salad: tomatoes, cucumber'. Result may be: 'Tomatoes — 2 mentions, Basil, Cucumber'. Or 'Vegetables: tomatoes, basil, cucumber'. Cannot predict in advance.

How to fix an unsuccessful result

If the generated list does not suit you, first analyze the source data: are there unclear formulations, mixed units, missing quantities? Then check if the number of people is indicated correctly. Often it is enough to rephrase the source text more strictly: each ingredient on a new line, with quantity and unit. Then generate the list again.

If the problem is grouping, try changing the order of ingredients in the source text. Also you can split one large request into several small ones: first generate a list for one dish, then for another, and then manually combine. This will help avoid confusion and data loss.

Remember that the tool does not learn from your edits and does not remember preferences. Each time it processes the input from scratch. Therefore it is important to prepare the input data carefully each time. If the result is still unsatisfactory, perhaps the task is beyond the capabilities of the tool, and it is worth using another method.

After generation you saw 'tomatoes 300 g' and 'tomatoes 2 pcs'. Fix the source text: 'tomatoes 300 g' and 'tomatoes 150 g'. Then generate again — you will get 'tomatoes 450 g'.

Checking the result and tool limitations

The ready shopping list is a draft that requires review. Review each line: does the quantity match your expectations, are there duplicates, are all ingredients accounted for. Especially carefully check positions that were specified without quantity — the tool may have just copied them unchanged. If in the list there is 'salt', 'pepper' and other spices without numbers, this is normal, but make sure they are not lost.

The tool does not guarantee accuracy and may err. It does not know that tomatoes for pasta and for salad are the same product if you named them differently. It cannot add up '1 bunch of basil' and '10 g of basil'. Therefore for complex recipes it is better to use specialized applications or tables. This tool is suitable for quickly combining simple lists, where all ingredients are already brought to a common view.

Take into account that the tool does not save history and does not export to DOCX or PDF. You can copy text or download it to TXT. If you need additional processing, you will have to do it manually. Do not rely on the tool for critical purchases, always double-check the list.

After generation you received a list of 10 positions. Check that all of them are really needed, and that quantities correspond to the number of guests. If something seems strange, edit the text manually.

The 'Shopping list' tool can be useful if you approach data entry carefully. Main errors are associated with uneven units, missing ingredients, incorrect number of portions and expecting too smart grouping. To get a good result, unify units, list ingredients explicitly, indicate exact number of people and do not exceed limits. After generation always check the list. Remember that the tool does not replace culinary logic and does not guarantee ideal order. Use it as an auxiliary draft, and make the final decision yourself.

Open tool: Shopping list maker ↗

Interface translations are machine-assisted and may need refinement.