Reaching the top three on Google Maps and Yandex Maps in Tashkent
The three-result local pack collects more clicks than the first organic position. Here is what you can control about those three slots and what you cannot.
How the three-pack is assembled
Google describes local ranking through three factor groups: relevance (how well the profile matches the query), distance (from the user's location or the district named in the query) and prominence (mentions, reviews, links, the site's organic visibility).
You do not control distance — it depends on where the user physically stands. The same business appears in the pack for someone in Chilanzar and not for someone in Sergeli. Therefore 'top three in Tashkent' does not exist as a single metric.
The controllable levers are relevance and prominence. The first is tuned through categories, description and services in the profile; the second accrues through reviews, local directory mentions and site authority.
Categories: the main relevance switch
The primary category in a Google Business Profile is the strongest controllable signal. 'Furniture store' and 'Custom furniture manufacturer' produce fundamentally different query sets.
Choose the primary category by what earns money, not by what sounds impressive. Add secondary categories only for services you genuinely provide: irrelevant categories dilute the profile and are routinely stripped by moderation.
Yandex.Business follows similar logic but uses its own taxonomy, noticeably more granular in consumer services. Fill categories in each service separately rather than copying mechanically — matching names do not guarantee matching meaning.
NAP consistency in this market
NAP means name, address, phone. The requirement is simple: identical, character for character, everywhere. In practice one Uzbek company's address appears as 'Tashkent, Amir Temur st. 15', 'Toshkent, Amir Temur ko'chasi 15' and a landmark-based variant.
Pick one canonical wording per language and align the Google profile, the Yandex listing, the site footer, the schema.org/LocalBusiness markup and every local directory to it.
Write the phone in international format, +998 XX XXX XX XX. Different formats of the same number across sources can be read as different phones, weakening the link between mentions and the organisation.
Reviews: volume, recency and replies
No algorithmic detail is confirmed, but the correlation between review volume plus recency and local pack position is consistently observed. Regularity beats raw count: ten reviews over six months outperform forty in one week followed by silence.
Reply to every review, negative ones included, within a couple of days. A reply is a public activity signal and an entry point for the user reading a dispute thread before calling.
What not to do: buy review packages. A burst of identical five-star ratings from empty accounts is filtered by both platforms, and the profile takes months to recover. It is especially conspicuous in Uzbekistan given the low overall volume of local reviews.
Photos, hours, services and posts
Profiles with regularly refreshed photos get more views and contacts — visible in the profile's own Performance statistics. Shoot the real interior, the facade with a landmark, the team and the products rather than stock imagery.
Fill in hours including holidays. Mark special hours for Navruz and other public holidays: a mismatch between real and stated hours is a common source of negative reviews, which then feed back into ranking.
The services and products section acts as an extra text corpus for query matching. List services in the same wording people search with, including Uzbek Latin variants.
Handling multiple locations
A chain needs a separate profile per physical location with a unique address and ideally a distinct phone number. One profile for the whole chain loses to competitors with a branch near the user.
On the site, build a dedicated page per location with address, map, hours and LocalBusiness markup, and point the profile at that page rather than the homepage. This is one of the few places where a profile-to-specific-URL link shows a visible effect.
Never create virtual locations at addresses where you are not physically present: both platforms verify addresses, and a detected fake location gets every linked profile suspended.