When you add a deadline, the category looks like the least important field: a drop-down to fill in so you can move on. In practice it is the choice that decides what you will be able to see a year from now, when there are two hundred rows and the question is no longer “what expires tomorrow” but “where does my recurring work actually sit”.
Bad categories do not bother you straight away. They bother you when you try to add things up to work out what hosting is really worth to you, and discover that half the entries sit under “miscellaneous” and the other half are split across six labels that all mean the same thing.
It is worth deciding them properly once, because redoing them later means touching every row.
Categorise by what breaks, not by what you call it
The most useful criterion is not commercial, it is operational: two deadlines belong to the same type if, when they lapse, the same thing happens and your reaction is the same. An expired domain takes down the site and the email together, and has to be recovered within precise windows. Expired hosting takes down the site but not the email, and is sorted out by talking to a provider. An expired plugin licence takes nothing down for weeks, and then leaves a security hole.
Three different levels of urgency, three different procedures: they deserve three types. Conversely, “WordPress hosting” and “Laravel hosting” follow exactly the same procedure, and the distinction only helps you remember how the site is built — which makes it a note, not a type.
The test is simple: if two deadlines would have you making the same phone call to the same person, they are the same type.
A short starting set
Five categories cover most of a web agency’s work:
- Domain — the cheapest item and the one that does the most damage. Its own recovery rules, tight timescales.
- Hosting — where the recurring side starts to carry real weight, and where the provider matters.
- Annual maintenance — the only item with no provider to buy from: the cost is your own hours.
- Antispam — and with it everything that keeps email running. It fails quietly, and you find out from clients who “aren’t getting emails any more”.
- Privacy policy — the deadline that is not technical: updates, cookie banners, revisions. No provider ever warns you.
The list is deliberately short, and it is not fixed: it is a starting point to strip down where you do not need it and extend where something is missing. If you do not sell antispam, that category is noise. If you resell licences to twenty clients, “licences” deserves a row of its own.
When is it worth adding a service type, and when is it not?
It is worth it when the new category would change a decision: if seeing the annual total for that item might lead you to raise its price, switch provider or stop offering it, then separating it makes sense. A type exists to be added up.
It is not worth it when the category only describes an individual row more precisely. At that point you are using the category as a title, and you already have a title. The warning sign is a category holding one or two deadlines in total: almost always it came out of a one-off situation that will never come round again.
In our experience the healthy number sits between four and eight. Below four you cannot read anything; above eight the drop-down becomes slow to use and people start picking at random, which is the quickest way to make every total meaningless.
Colour is not decoration
Once the categories exist, colour is what lets you read the list without reading it. It works on one condition: that it is consistent and that it means something. Colours assigned at random make the list more colourful, not clearer.
Two practical rules are enough. First: keep colour families close together for services that are close together — technical services living on the same server in similar shades, administrative items in a different family. Second: do not use red for a category. Red belongs to statuses, where it signals that something needs doing; give it to a type as well and the list shouts constantly, until you stop looking at it.
The reverse holds too: if you use too many muted shades that look alike, you may as well use none. Five colours you can tell apart at a glance beat twelve shades you have to interpret.
Type and provider are two different things
This is the most common structural mistake: using the provider’s name as the category. There is a logic to it — when you need to renew, that is where you log in — but it breaks everything else. If the type is the provider, you can no longer tell what hosting is worth in total, because it is spread across three different names. And the day you change provider, you have to reclassify the lot.
They are two independent pieces of information and you need both: the type says what something is, the provider says where it lives. Keeping them apart lets you answer both “what is my hosting worth” and “what would I have to move if I left this provider”, which is exactly the question you ask yourself when a price increase lands.
What are the most common mistakes when categorising services?
- Using the provider as the category. Fine for renewals, fatal for any total.
- Keeping a “miscellaneous” entry. It always grows, and it ends up holding precisely the things nobody looks at.
- Creating a category for every one-off case, until you have twenty types with one row each.
- Categorising by client instead of by service: that is the view you already have, and it does not tell you where the recurring work sits.
- Changing categories without tidying up the history. Half the list on the old scheme and half on the new one is worse than an imperfect but consistent list.
Where do you start categorising recurring services?
- List everything you renew in a single column, with no categories, for a dozen real clients.
- Group by reaction: put together the items you would handle the same way if they lapsed. The groups that emerge are your types.
- Drop the one-row groups and check that none of them holds more than a third of the total.
- Assign colours by family, keeping red out of the types.
- Add up by category. It is the first useful number you get, and it usually contradicts something: the comparison with a proper deadline tracker explains why this total falls apart in a spreadsheet, and how to organise a web agency’s client deadline tracker puts the whole structure together.
How do we categorise services ourselves?
The five categories above are the ones Dotify starts with, the deadline tracker we built for ourselves at EnneStudio and still run on. They are a starting point, not a fixed list: they can be renamed, deleted and added to, because the right set depends on what you sell.
The decision we cared about most is the one in the previous section: type and provider are two separate fields, each with its own colour. That is why you can look at the same list in two ways — by service when you need to see where the margin is, by provider when you need to decide what to migrate.
How to set up types, colours and providers is covered in the documentation on deadline management. The free plan holds five clients with three deadlines each: enough to try the structure out, not to move your whole client base across.
