Raspberry jam cookies, the Swedish way
"Kärt barn har många namn" – [a beloved treat that goes by many names], but always begins with butter, flour, and a spoonful of jam.
Welcome to The Bakers’ Cottage, a newsletter about baking, stories and the rhythm of life in the north. If you’re new here, thank you for joining! And if you’re already a subscriber – I’m so glad to have you back.
Today’s post is about syltkakor – Swedish jam cookies that belong on every fika table. Also known as hallongrottor [raspberry caves], rosenmunnar [rose mouths] or syltgrottor [jam caves], I’ve always made mine from the base my friend taught me: flour, potato starch, salted butter, icing and vanilla sugar. The dough comes together in minutes and keeps well in the fridge – or the freezer, if you like to bake just a few at a time.
My preferred filling? Cloudberry jam, if I have it. Otherwise, a tart raspberry or redcurrant is perfect.
I roll the dough into small balls, then press a little dip in each with the end of a wooden spoon. A teaspoon of jam goes in just before baking, where it bubbles softly, then sets.
Could it be even better?
Although I’ve tweaked and baked this recipe for years now – adjusting the sugars, switching flours, testing different techniques – I’ve always had the sense it could be even better.
So I set out on a little quest. Not to reinvent the Swedish jam cookie, but to refine it – to write a formula that holds together, scales easily, and delivers exactly the texture and flavour I want.
The research: looking for patterns
As often, I started by comparing recipes. Some I’ve baked. Others I’ve only read – and a few are my own, past versions I once thought were finished. I lined them up in a spreadsheet: five Swedish classics, my adapted Hallongrottor recipe, and three French butter biscuits.
What surprised me wasn’t the range of ingredients – it was the consistency in how they were used. Most recipes chose a single sugar – either caster, icing or demerara – and kept to it. Eggs were used sparingly, either as yolk, white or whole. Only Sju sorters kakor and my original recipe included potato starch.
Butter ratios sat high across the board – usually between 60 and 85 percent (in bakers’ %).
That’s where the real decisions start – not in the ingredient list itself, but in how each balances the next.
Sketching a formula
From there, I wrote a test formula, assigning percentages to each ingredient – sometimes using baker’s percentages, other times composition percentages, depending on which felt more intuitive for that component. These are things I’ve learned through practice – by making, testing, adjusting and making again.
100% flour (baker’s %) – I chose sifted spelt flour, like My Feldt, for its nutty flavour.
20% sugar (composition %) – a fairly standard sweetness, on the higher end of the range.
2% baking powder & 1% salt (baker’s %) – small but functional amounts to lift and balance.
75% butter (baker’s %) – a touch above average; aiming for richness.
8% whole egg (composition %) – for a less friable crumb.
The 1 kg test batch
I decided to scale the total batch size to 1 kg – a nice, round number that’s easy to work with.
From the composition percentages, I knew that:
Sugar = 20% of the total → 0.20 x 1000 = 200 g
Eggs = 8% → 0.08 x 1000 = 80 gThese added up to 280 g of the total recipe.
That left 720 g for the remaining ingredients: flour, butter, baking powder, and salt – all set in baker’s percentages, which are relative to the weight of the flour. In the formula, those added up to:
Flour: 100%
Baking powder: 2%
Salt: 1%
Butter: 75%
Total = 178%So, I calculated the actual flour weight using proportion:
If 178% = 720 g
Then 100% flour = (100 x 720) / 178 ≈ 405 g
Once I had the flour weight, the rest followed:
Baking powder = 2% x 405 = 8 g
Salt = 1% x 405 = 4 g
Salted butter = 75% x 405 = 304 gAnd there it was – a formula yielding 1 kg of dough. Perfect for the restaurant (I might even need to scale it up there), but too much for my kind of home baking.
Raspberry jam cookies, the Swedish way
Scaling it down from the egg
When scaling down, I often start with the eggs and work backwards. They're awkward to divide neatly, and I rarely feel like fussing with half an egg. The 1 kg batch used 80 g egg – about 1½ medium eggs. So I scaled the recipe by dividing everything by 80 and multiplying by 50.
That gave me a dough weight of around 630 g – just right for 30-ish small biscuits at 20 g each. Perfect for the kind of afternoon when you just want a tray of something warm and golden on the table, with the rest tucked in the freezer.
Here’s the scaled-down version:
My home version: syltkakor 1.0
And my home version, with vanilla sugar and rounded quantities. I added vanilla sugar to the rounded version (5 g, part of the 125 g total sugar) because I cannot not to.
The dough rolled easily and held its shape well. I tried three different moulds – two smaller ones and one larger. A mini-muffin tin would work nicely too, or small paper cups.






Tasting notes
Flavour is out of this world. Deep, buttery, perfectly balanced.
But… a little too sweet.
Next step
A small adjustment in the caster sugar ratio – maybe 40% instead of 48% (bakers’ %). But otherwise, this one’s very close.
Syltkakor 2.0
A slightly less sweet version of the original test – same structure, better balance.
After tasting the first batch, I knew I was close. This second version lowers the sugar slightly (from 50% (caster sugar + vanilla sugar) to 42% in baker’s %), while keeping the rest of the structure intact. The result? A more balanced cookie – still rich, still crumbly, but better suited to the jam.






Tasting notes
The sweetness felt right, but the texture didn’t. A little too sandy.
Next step
For the next test, I’ll aim for a sugar level somewhere between version 1.0 and 2.0 – maybe around 45% (in baker’s %).
Syltkakor 3.0



Really happy with these. Just right: buttery, crumbly… A cookie to keep!
For now, this is the version I’ve written down in my notebook but of course, there’s always room for another batch! TBC :)
Syltkakor, the final-for-now recipe
MAKES 30 to 50 cookies (12 to 20 g each)
For the dough
250 g light spelt flour (or plain flour)
110 g caster sugar
1 tsp vanilla sugar (read note above)
1 tsp baking powder
½ tsp flaky sea salt
180 g salted butter, cubed
50 g egg (1 medium)
For the filling
About 150 g tart jam (raspberry, redcurrant or cloudberry)
Make the dough
Preheat the oven to 200°C / fan 180°C. Line a baking tray with baking paper and arrange small metal forms or paper cups. Alternatively, use a mini muffin pan. Grease and set aside until needed.
In a stand mixer fitted with the paddle attachment, or in a food processor, pulse the flour, sugars, baking powder, salt, and butter until the mixture looks sandy.
Add the egg and mix just until incorporated.
Divide the dough into two and shape into logs. Wrap in clingfilm and chill for 20–30 minutes, or up to 3 days. At this stage, the dough can also be frozen for up to 3 months.
Shape the cookies
Take one log of dough and divide into 15–25 pieces, depending on how large you want your cookies (anywhere between 12 and 20 g per piece).
Roll into balls and place in the prepared forms.
Use the end of a wooden spoon (or your thumb) to press a dip in the centre of each ball.
Pipe a little jam into each dip, being careful not to overfill, as the jam will bubble up as it bakes and the cookies may stick to the pan.
Bake the cookies
Bake for 12–16 minutes, until just set and lightly golden at the edges.
Cool on the tray for 5 minutes, then unmould and transfer to a wire rack.












Spelt flour? Never really thought about it before, but I’ll give it a try - hallongrottor are such a lovely treat, älskar dem!”