Before You Sign the Lease, Search the Block

You are choosing between two shop units. One is inside a mall, one is a shophouse a few minutes' walk from the MRT, and the rents are different. The agent has footfall estimates for both. Nobody has told you which of the two addresses people actually type into Google, and you are about to commit for years.

There is a ten-minute check for that, and it works at building level, not town level. On 26 September 2026 we ran three Singapore estate names through Google Autocomplete and read all 108 suggestions that came back. Two of the three estates showed buyers naming specific malls and neighbourhood centres inside the town. The third showed something else entirely: 13 of its 39 suggestions were about a place in another country.

Which of those two patterns your address sits in changes what your website should say, and whether search marketing is worth buying for that unit at all.

Why the town name is the wrong unit of measurement

Most owners, and most agencies, plan location pages at town level: a Tampines page, a Jurong page, a Woodlands page. That is one level too coarse, because it is not how people ask.

The reason to trust autocomplete on this is what it is built from. Google's own explanation of how autocomplete predictions work states that "Once you start to type, autocomplete predictions reflect real searches that have been done on Google", and that the system also weighs "the location a query is coming from" and "trending interest in a query".

So when Google offers to finish "pilates tampines" with "tampines hub", that is not a guess about language. It is a report that people have typed it.

What 108 suggestions across three estates showed

Three seeds, pulled through our own keyword tool on 26 September 2026 with the market set to Singapore. We counted two things per estate: how many suggestions named a place inside the town, and how many named a place outside Singapore.

Seed	Suggestions	Named a place inside the town	Named a place outside Singapore
pilates jurong	38	15	0
pilates tampines	31	11	0
gym woodlands	39	3	13

Three estates is not a survey of Singapore. It is enough to show that the answer differs by address, which is the entire point of running the check on yours.

Tampines and Jurong: buyers name the building

In both of these estates, the sub-town naming was the dominant pattern, and the places named were specific enough to plan around.

For "pilates jurong", 15 of 38 suggestions named somewhere inside Jurong: Jurong East, Jurong West, Jurong Point, Jurong Gateway, Jurong Lake, Jurong Kechil, and even the CPF Building. For "pilates tampines", 11 of 31 did the same: Tampines Hub, Tampines Mart, Tampines Plaza, Tampines Junction, Tampines East, Tampines West, and Tampines East CC.

Read what that means commercially. A buyer typing "pilates tampines hub" is not shopping the town. They have decided where they are willing to travel to, probably because they already go there for something else, and they want to know what is in that building. If your unit is in or beside that building, you are the answer to a search that already exists.

The practical consequence is a page, not a campaign. One page that names your building, the mall or centre it sits in, the nearest MRT exit, and your opening hours will do more for that search than a generic town page. Our local SEO guide for Singapore covers the Maps half of the same job.

There is a warning in this data too. If buyers name buildings, your competitor in the same building is being searched alongside you, and often the estate's search is held by a publisher rather than by either of you. We looked at that pattern in what to do when your competitor is in the same building.

Woodlands: 13 of 39 suggestions were about another country

The third estate behaved completely differently, and if your unit is there, this is the more useful half of the article.

Of the 39 suggestions for "gym woodlands", 13 named a place outside Singapore. The Woodlands in Texas appeared repeatedly. So did Doncaster, Grenada, Waterford, Limerick, Lusaka, Woodland Hills and Woodland Park. Only 2 of the 39 said "Singapore" outright. Only 3 named a place inside Singapore's Woodlands, and two of those were the community centre.

We have hit this before from the other direction. In our study of 109 local pages across five regions, 31 of the 120 results returned for five Woodlands searches were about somewhere else called Woodlands, mostly in Texas. Now we can see it in the suggestions too. The collision runs in both directions: it shapes what Google offers to type, and it shapes what comes back.

What that costs an owner in Woodlands is clarity. Every phrase you publish has to carry "Singapore" in it, in the page title, the opening paragraph and the address block, or you are competing with three continents for a word that was supposed to describe your neighbourhood. That is not a reason to reject the unit. It is a reason to write "Woodlands, Singapore" every single time, and to lean harder on the community centre and the shopping-centre names that do return local suggestions.

The check I would run before signing

Ten minutes, per unit, before the lease. Open an incognito window, set your service as the seed, and type it with each address's name.

1. Does the estate name return places inside it? Type your trade plus the town and read the suggestions. If you see malls, centres, MRT stations or sub-districts, buyers are searching at building level. Note every named place within walking distance of the unit you are considering. Each one is a page you can own.
2. Does the estate name collide with somewhere overseas? Look for foreign towns, states or countries in the same list. If more than a handful appear, every phrase you publish needs "Singapore" attached, and you should expect your reporting to carry noise from other countries.
3. Then compare the two units on what you found, not on the town. A unit next to a building people name by name is worth more in search terms than a unit in a better-known town whose name nobody qualifies.

If both units come back thin, with no sub-locations and no obvious demand, you have learned something worth knowing: this address will not be found through search, whichever one you pick. Sign on footfall and rent, and do not spend on search marketing for it yet.

What this check cannot tell you

Four limits, and they matter.

Autocomplete tells you which phrasings exist. It does not tell you how many people use each one, and we do not publish those estimates because our tool derives them partly from the Search Console data of businesses we work for.

Our pull ran from a server with the market set to Singapore, not from a phone standing in Tampines. Google says location is one of the factors in what it suggests, so a searcher physically in the estate may see a different list. Run the check on your own phone, in the area, as well.

Three estates and two trades is a small sample, pulled on one morning. Treat the table as worked examples, not a ranking of Singapore towns.

And the biggest one: search demand is not the same as business viability. A unit can be a good lease with no search demand, and a bad lease with plenty. This check tells you whether search can route customers to that address, nothing more. It is marketing guidance, not property, legal or financial advice, and it is no substitute for your own commercial and professional advice on a tenancy.

What it costs

The check itself costs ten minutes and nothing else, which is why we would rather you just ran it.

Acting on it costs more. Building-level pages mean you commit to maintaining them: opening hours, the right MRT exit, a real timetable. Three pages you keep accurate beat twelve that go stale, and stale local pages actively lose you trust. If you name a mall you are not in, or a station you are not near, you will be found out.

The other cost is the honest one. If both units read thin, the correct decision is to spend nothing with an agency, ours included. Search cannot create demand for an address. It can only route demand that is already being typed. Anyone offering to build that demand from nothing for a retainer is selling you a story, and we would rather lose the sale than tell it.

If you have not yet committed to the business at all, the earlier version of this check is the one to run first: before you incorporate, check if anyone is searching.

Want us to run it on both addresses?

Send us the two units you are choosing between and the service you will sell. We will pull the suggestion sets for both and tell you which address search can actually help, including if the answer is neither. It starts as a free visibility check through our contact page.

For a fitness studio specifically, the worked example above continues in our guide to SEO for gyms and fitness studios in Singapore.

Counts above come from Google Autocomplete pulled through SingRank's keyword tool on 26 September 2026, market Singapore. No monthly search volumes are published, and no business is named.

FAQ

Can search demand really tell me anything about a shop unit?

It tells you whether people already type the name of that building or estate when looking for your trade. In our pull, 11 of 31 Tampines suggestions and 15 of 38 Jurong suggestions named a specific place inside the town. That is routable demand. It says nothing about footfall or rent.

What if my town name exists in another country?

Write it with "Singapore" attached every time, in your page title, your opening paragraph and your address block. For "gym woodlands", 13 of 39 suggestions named places outside Singapore and only 2 said Singapore. Expect noise in your reporting and check that your traffic is local.

How many location pages should I build?

Only as many as you can keep accurate. Start with one page for the building or centre your unit sits in, plus the nearest named landmark that appeared in your suggestions. A dozen near-identical town pages is the pattern Google treats as spam, and it is also more than most owners maintain.

Is a thin result a reason not to sign the lease?

No, and we would not read it that way. Plenty of good businesses run on walk-past trade, referrals or delivery platforms. A thin search read means one thing only: do not buy search marketing for that address yet. Decide the lease on rent, footfall and your own numbers.

Should I run this before or after choosing the unit?

Before, because it is free and takes ten minutes per address. Once the lease is signed, the check becomes a content plan rather than a decision. Either is useful, but only the first version can still change which unit you take.

Does this replace a proper feasibility study?

No. It measures one narrow thing: whether people type your trade alongside that place name. A feasibility study weighs rent, competition, staffing, margins and licensing. Use this as one input among several, and take professional advice on the tenancy itself.