Sneak Peek: Spending Updates


A welcomed change. When will we be able to see longer higher date ranges than one month? I want to look at how much I have spent this year and how much I’ve earned…


I think a few people have mentioned it earlier in the thread but custom date ranges would be useful. For instance I get paid the last working day of the month, this isn’t alway the last day so my spending can go a bit a skew when I’ve spent a small fortune over millionaires weekend which still falls within the previous month.


I agree… Entirely! In fact, there are actually quite a few more categories I would like to see, but I know that they are particular to me and my way of budgeting and my lifestyle. But this isn’t one of them. My thinking here was that I believe the implementation of custom categories could be quite an overhall and, I assume, difficult and time consuming. Whereas they have just demonstrated that changing the categories is not so much. I only put “Fuel” forward, as opposed to any of my other requirements, because I imagined it would be more common and therefore less meaningless. For a HUGE number of people, fuel is a specific expense that there are normally attempts to control and budget for. Pretty much every single person that I know who drives can tell me IMMEDIATELY how much they spend on petrol or diesel for a particular time period. Like, “Oh, I normally put thirty quid in and try and get through a fortnight.” Whereas parking tends to change depending on what you’re doing and things like servicing and changing tyres, etc are infrequent.

In my particular case, for example, I try to keep my fuel costs less than £200 a month. During random months recently I’ve got over a grand showing in my “Transport” category because my car needed stuff. It makes that number useless to me.

So yes… Customisable categories gets my vote EVERY SINGLE TIME … And yes… Asking for a category called “Snake food for Hector” is meaningless to everyone else (but not to Hector :snake: ) … But, in the meantime, I honestly feel that if someone suggests something, like fuel, or any other idea, because they genuinely feel it could be a more common and more needed suggestion, and enough people agree with it to make Starling take notice and consider it, then its not really meaningless at all… It’s just trying to help everything be that little bit better.

As it is only one person said yes (Thank you Andy) so maybe it won’t get noticed at all :joy:. But I did it with good intention and so will only cry at the rejection a little bit. :grimacing::grimacing::joy::joy::joy:.

Goodnight everyone. Xxxx


I agree Fuel is an important category (as often reclaimable from employers whereas screenwash or car wash may not be), along with Charitable Gifts (for annual self assessment returns), and prescriptions (for England, as free in other countries) as some reclaimable from private insurance cover and if NHS ones the figurework useful to calculate if the cost of a PPC is cheaper than Rx charges.


And just to back up my point, I’ll add that while I’ve never claimed Charitable Gifts on a self assessment form and I honestly can’t remember the last time I took medicine, or paid a prescription (maybe 15 years ago?), they’re worth suggesting because, if enough people agree, it could make it happen.

Just because we can’t have customisation yet doesn’t mean, to me, that we might as well not bother with anything!


Personally I have about 10 Rx items a month so having categorized data enabled me to realise a PPC was cheaper. But I don’t just want what is right for me, I want what will help a lot of people.

Some existing categories are useful. Now while I don’t have a pet I still think a category like that makes sense, as even if you can put some things like dog food under groceries, vet bills for example may be covered by an insurance scheme or a pay monthly budgeting plan.

How people use the categories may vary. Some may put airport parking, flights and hotel as Holidays but put the food as Eating Out. Others may put the parking and flights under Transport and the hotel and eating out under Holidays. What matters is it is right for us, the customer.

Monzo’s reasoning for having less more rigid categories was so they can use anonymized data for statistical uses. However, I feel categories should be there to assist the customer, even though that means less useful amalgamated data.


Agree this should be for the customer rather than the bank as the Monzo’s and Starling’s want to be financial hubs. Custom categories shouldn’t be difficult but would probably have less meaning to the banks but that’s fine by me.


I can see the benefit of having custom categories, there’s some very good examples in this thread. However, the biggest arguement against this IMO is automation.

With a pre-defined set of categories Starling can easily build up a central database of merchants and categories to ensure that any purchases from a merchant are automatically assigned to a suitable category. As the category list does not change per user, any machine learning of categories could be done centrally and benefit all users.

With custom categories, this would either require a manual approach in assigning all custom categories, or per-user machine learning of categories. This would be a lot more resource intensive than the central approach, and any learning would not benefit other users, as there categories could be different.


Machine learning of the larger customer base doesn’t actually help us as individuals hugely. Of course it helps with new transacations at locations we’ve not spent money at but really what matters is being able to categorise the regular spending in order to help budgeting. By be being able to create custom categories and simply apply those to all previous spend at that merchant, just this spend or all future spends at that merchant it would cover a much larger number of use cases for the wider customer base. Generally someone may like Marks and Spencer to be categorised as Groceries, another may prefer this to be Working Lunches, another Clothes… This is just a very simple example but the need for customised categories exists and it’s a very small change to make the solution fit a very large number of customers use cases.


Purchases with a card all show an MMC four digit code indicating merchant type, e.g. grocer, petrol, car hire, hotel, automated fuel dispenser (pay at pump), fast food, restaurant, etc and this code should enable correct categorization in 99% of the cases, except when a merchant is wrongly registered or a huge supermarket chain has a mini supermarket that sells multiple retail types but the one POS rather than separate ones in different departments.


I didn’t know about the MMC code, but Starling still need to map those merchant types to their categories.

My point was that I imagine this is currently done centrally, and to be able to learn to automatically categorise transactions to custom categories this would need to be done per user (as their categories may be different).

This is definitely possible, but I suspect would need a significant change in how Starling automatically assign categories. I may well be wrong, and Starling could already be storing this information on a user-level, in which case this would be a relatively easy implementation.


Now there’s a cool feature - one click automatic claims for Charitable Gifts.


with one account abroad they have a feature to enable you to prepare a tax return from your internet banking and transmit direct to their local tax authorities which is quite useful.

Some integration with HMRC may be good even if it just a secure messaging function to send messages that the HMRC know comes from a verified person!


Just thinking as well it would be good to split a transaction into different categories as well. I’ll often buy other things along with fuel like milk, bread or whatever and I don’t want to include that in my fuel categorisation as that doesn’t give me accurate information - a bit trickier to implement probably but I personally would find it very useful


I second this. :+1:t3: