I have served skyr at multiple feasts, including Ymir 2025, and expect to make it again. This post is the canonical recipe for the dish across versions.
What, Where, When
- Home-cultured Icelandic skyr — a Norse fresh cheese, distinct from yogurt, made with a true skyr culture and a small amount of rennet.
- Recipe sourced from a mid-20th-century Icelandic cookbook via an online forum; technique substantially preserves the traditional method.
Discussion
Skyr is one of the best-attested Norse foods. It appears in Egil’s Saga and Grettir’s Saga, and is described in contemporary overviews of Viking-age diet as a staple of the Icelandic table — a kind of cultured dairy that travelers and settlers carried with them from Norway and developed further after Iceland was settled. The technique was, by some accounts, lost in mainland Scandinavia after the medieval period, surviving primarily in Iceland where it remains a defining national food. Modern commercial skyr production has revived it as an export product, and Icelandic-style skyr is now widely available in North American grocery stores.
One important clarification: skyr is not yogurt. It looks like thick yogurt and the taste is similar, but it is technically a fresh cheese — made with a small amount of rennet (which sets the curd) in addition to the bacterial culture (thus, while vegetarian, it is not technically vegan unless vegetable-based rennet substitute is used.) The texture is denser and the protein content is higher than even strained Greek yogurt. The bacterial cultures specific to skyr are also distinct from those used in yogurt, and produce the characteristic flavor — which is why the most reliable approach, if unable to source skyr culture directly, is to use commercial Icelandic skyr (Siggi’s is widely available in the US) as the starter culture for a fresh batch.
The recipe below is a half-gallon home-scale adaptation of a method published in Nýja Matreiðslubókin by Halldóra Eggertsdóttir and Sólveig Benediktsdóttir (Reykjavík, 1961), which I encountered through an English translation posted on the Icelandic discussion forum bland.is. The historical material in that translation is drawn in turn from Súrt og Sætt, a teaching leaflet by Sigríður Sigurðardóttir published by the Skagafjörður Folk Museum in 1998. While not a peer-reviewed historical reconstruction; it is a mid-20th-century Icelandic household recipe that preserves traditional technique, with historical context. More importantly, it contained important details regarding what the cook should look for and how to check the curd, precautions to take when straining, and exact temperatures needed.
I made one substantive deviation from the source recipe: it specifies skim milk, while I used whole milk. Traditional skyr was made from skim milk because the cream was separated for butter — that is, skyr was the leftover from a different dairy operation. Using whole milk produces a richer skyr with a softer curd. If you want the traditional product, use skim. If you want the richer version, use whole. The technique is the same either way.
Home-Cultured Skyr
Half-gallon home batch of Icelandic-style skyr, cultured with commercial skyr and set with a small amount of rennet. Yields about 1.5-2 cups of finished skyr after straining.
Ingredients
½ gallon (1.9 L) whole or skim milk (preferably non-ultra-pasteurized)
3 tbsp commercial Icelandic skyr (Siggi’s plain works well), at room temperature
1-2 drops liquid rennet, or ¼ rennet tablet dissolved in 1 tbsp cool water
Directions
- Heat the milk in a heavy pot over medium heat, stirring frequently to prevent scorching, until it reaches 86-90°C (185-195°F). Hold at that temperature for about 10 minutes — this denatures the milk proteins and produces a firmer, more skyr-like curd.
- Remove from heat and let cool slowly to 39°C (102°F). This takes about 2 hours at room temperature; do not rush it with an ice bath, as the slow cool helps the bacteria culture establish properly.
- In a small bowl, stir the commercial skyr with a few tablespoons of the warm milk to make a smooth slurry. Add the dissolved rennet to the slurry, then stir the slurry into the pot of warm milk. Mix gently but thoroughly.
- Cover the pot tightly and wrap in towels or a thick blanket to insulate it. Leave undisturbed for about 5 hours. The milk should curdle in this window. Curdling in less than 4½ hours produces coarse curds; curdling in more than 5 hours is reputed to produce curds that are difficult to strain, although I’ve allowed it to sit longer without ill effect..
- Test the curd: cut into the surface with a knife. If the cut stays open and does not close back up immediately, the curd is set and ready. If the cut closes, give it more time and check again in 30 minutes.
- Line a sieve or colander with cheesecloth or a fine linen cloth, set over a bowl to catch the whey. Pour the curds into the lined sieve. A successful skyr will release very little whey, and what releases will pool around the edges and in the cuts on the surface — the curds themselves will crack and fall apart in pieces. Do not pile the curds more than 7-9 cm (3-4 inches) thick.
- Tie the corners of the cloth together over the curds and refrigerate the whole sieve-and-bowl setup for 12-24 hours, allowing the whey to drain. The skyr should end up firm and dry-looking when ready.
- Transfer the finished skyr to a covered container and refrigerate. It will keep for 4-5 days. The whey can be saved as a drink, used to pickle vegetables, or used as a replacement for white wine in cooking.
- Serve plain, or stir in a little milk and sweetener and serve with cream and fresh berries. Blueberries are traditional in Iceland; raspberries, strawberries, or any soft fruit work well. Honey, maple syrup, or brown sugar are all good additions. Reserve a few tablespoons of the finished skyr to use as starter for the next batch.
Troubleshooting
If the whey does not separate cleanly, or if it floats over the curds rather than draining around them, or if the curds do not pull away from the edges of the sieve — something has gone wrong. The most common cause is that the milk was not heated to a high enough temperature, or was cooled too quickly, so the rennet did not have time to act fully. The second most common cause is using ultra-pasteurized milk, in which the proteins are too denatured to set properly even with rennet; standard pasteurized milk works much better than UHT milk.
One personal note: at one feast, I lost track of the milk while it was cooling, and the cultures got added significantly later than the recipe specifies — well past the 39°C window. I expected the batch to fail. The next morning, the curd had set perfectly. The technique has more tolerance than the precise temperature spec suggests, especially with a rich modern starter culture, but I would not bank on getting away with it twice.