MapBees logoMapBees

Blog

Share of its category: what fraction of the whole, explained simply

Ten cafés in an area is one story if they are the only ten food places, and another if they sit among two hundred restaurants. 'Share of its category' tells you which - here worked through with cafés as a share of everything to eat and drink in London.

Formula seriesBeginner friendlyLondon example
Open full map appBack to blog
ByTinkal Gogoi(LinkedIn profile, opens in a new tab)September 3, 20266 min read

Ten cafés, two very different streets

On one street, ten cafés are the only places to eat or drink at all. On another, the same ten sit among two hundred restaurants, pubs and takeaways. A count says both streets have ten cafés. One is a café street; the other barely notices them.

Share of its category tells you which. It divides one kind of place by the count of its whole category in the same area - cafés by everything to eat and drink - and gives a percentage: 100% on the first street, 5% on the second. The field just calls it a share, and it is the ratio to reach for whenever the question is “what kind of place is this?” rather than “how much is here?”.

Share of its categorythe field's term: Share

Cafés as a share of everything to eat and drink. Brighter areas lean toward cafés; darker ones have their food scene made of other things.

Where the category comes from

You never type the denominator. The place data MapBees uses is a tree: every kind of place knows its parent, and its parent's parent. “Cafe” sits under “Eat & Drink (all)”, “Italian” under “Mediterranean”, “Gym” under “Active Life & Fitness (all)”. The control names the parent it will divide by, so you always know what the share is of.

Why some metrics grey it out

Total population and a transit-stop count belong to no wider category, so there is nothing to take a share of. The option is disabled and says exactly that.

Share versus rate

Per 1 000 residents compares a count with the people. Share compares it with its peers. 'Are there many cafés?' is a rate question; 'is this a café area?' is a share question.

Switching it on

Where to click

  1. Map type
  2. Amount
  3. Cafe
  4. Count as: Share of Eat & Drink (all)
The 'Count as' option is labelled with the parent it uses - 'Share of Eat & Drink (all)' - so the denominator is never a mystery.
Pick a different kind of place and the label changes with it: a share of Mediterranean for Italian restaurants, of Health & Medical for pharmacies.

Reading the map

A share is bounded: it cannot go below 0% or above 100%, which keeps the legend readable without any ranking.
One area in south-east London: cafés are 16.7% of its places to eat and drink - and the card admits the share was adjusted, because the category count there is small.

The maths, honestly

Cafés as a share of the food scene in one area
Cafés counted
10
Everything to eat and drinkthe parent category's count in the same area
60
Share of its category
10 ÷ 60 = 16.7%

The same ten cafés among two hundred food places would be 5%; as the only ten, 100%. The count never changes; the meaning does.

Click an area and the inspector writes the share out: this area's cafés, divided by this area's whole category.
Like a rate, a share is adjusted where the base is tiny: one café among two food places would read as 50% and mean nothing.

One café among two is not 'half'

The base of a share is the category count, and in a quiet residential area that can be two or three places. One café among two would read as 50%, which says nothing about the street. So, exactly as for per 1 000 residents, a tiny base is pulled toward the city's share and the area is flagged as adjusted. An area with no places to eat or drink at all has no share, rather than a share of zero.

Open cafés as a share of London's food scene

The whole idea, in four lines

  • Share of its category = one kind ÷ its whole category, in the same area, as a percentage.
  • The category comes from the place hierarchy the data already carries, and the control names it.
  • It answers 'what kind of place is this?' - ten cafés are a café street at 100% and background noise at 5%.
  • A share is bounded at 0-100%, so the legend reads straight off with no ranking needed.
  • Tiny category counts are adjusted and flagged; an area with none of the category has no share at all.

Quick questions

What does 'Share of its category' calculate?

The count of one kind divided by the count of its wider category in the same area - cafés divided by everything to eat and drink, for example - shown as a percentage.

Where does the category come from?

From the place hierarchy the data already carries. Every kind of place knows its parent, so 'Cafe' knows it belongs to 'Eat & Drink (all)', and the app names that parent in the control.

Why is it greyed out for some metrics?

Some fields have no wider category to take a share of - total population, or a transit-stop count - so the option is disabled and says why.

Is a share adjusted for small numbers?

Yes. One café among two food places would read as 50%, which says little. Where the category count is tiny, the share is pulled toward the city's and the area is flagged as adjusted.

See it on a real map

The best way to understand the score is to play with it. The full app is on a free tier, with no sales call.

Open the live mapBack to all articles

About the author

Tinkal Gogoi(LinkedIn profile, opens in a new tab)Founder & Location Data Engineer

Builds MapBees's location-intelligence pipeline end to end — census and Overture Maps ingestion, H3 hex scoring, and the map interface — and writes the methodology behind every number on the site.