What’s Next for Mobile Search? Exploring Future Input, Data, and Output Innovations

Recently, the mobile search arena has seen intense competition among major players: Baidu is intensifying its mobile efforts, Alibaba merged UC to launch "Shenma Search," Tencent partnered with Sogou to release an independent Sogou Search app, 360 Search also launched its own app, and Wandoujia is transforming its product toward search. These moves indicate an upcoming fierce battle in mobile search.

Since Wandoujia introduced the concept of "in‑app search" and adjusted its product, Sogou and 360 have followed suit, launching independent mobile search apps. This shift shows mobile search moving away from PC‑centric thinking toward greater innovation. The article explores future innovation points in mobile search across three aspects: input methods, data processing, and output methods.

Input Methods

Precise User Portrait

In the mobile internet era, smartphones act as extensions of humans, even likened to new organs. As the primary medium connecting users to the internet, phones carry extensive data that can help pinpoint user identities.

Mobile search can leverage a user's social accounts, installed apps, usage habits, and other signals to gradually build a precise user portrait. Accurate user profiling is the fundamental input for future mobile search, enabling more personalized services.

Crowdsourced Data

When a user interacts with mobile search, they implicitly contribute behavior data, which, combined with the user portrait, is sent to the cloud for mining, further refining the portrait. The more a user engages, the more accurate the portrait becomes, allowing the search engine to better understand preferences and deliver precise information.

Search as Exploration

Mobile search differs from PC search in usage scenarios; while PC search is confined to a desk, mobile search occurs in diverse contexts, tightly linked to user needs. Thus, mobile search can be viewed as "mobile exploration."

Future mobile search will go beyond keyword entry, accepting varied inputs such as GPS location, QR codes on posters, nearby street‑view scans, or iBeacon‑pushed product information from nearby merchants.

Data Processing

Natural Language Understanding

Voice‑driven mobile apps have cultivated user habits, making voice input increasingly accepted. 360's recent "360 Search" app emphasizes voice search, and future development will see voice gradually replace keyboard input.

The challenge lies not in speech recognition but in understanding natural language. Precise answers require deep natural language comprehension, a core technical hurdle for mobile search.

Deep Linking

Traditional search engines were built for the PC web, whereas the mobile era consists of numerous isolated apps. Links and data cannot easily traverse between apps as they do on the web.

Deep linking breaks these app silos, similar to Wandoujia's "in‑app search" protocol, enabling data exchange across apps.

As deep linking becomes a unified standard promoted by app stores, mobile search will be able to retrieve content across apps and launch specific app pages, turning search into a major distribution channel.

Machine Learning

One forward‑looking idea for mobile search is "no input"—the search engine would infer user intent from profiles, behavior logs, sensor data, and other signals, relying on extensive machine‑learning models.

Although "no input" may seem fanciful now, reducing user effort and delivering precise results both depend on machine learning, implying that mobile search back‑ends will involve complex data‑processing clusters.

Output Methods

Structured Data

Mobile search inherits challenges from PC search, where information overload creates noise. Mobile users need rapid access to relevant information, so presenting popular queries in structured formats—e.g., displaying a World Cup schedule table—reduces noise.

Intelligent Aggregated Data

Structured displays give clear information, but users also desire comprehensive insights. Search engines should intelligently analyze keywords to present related data, such as encyclopedic entries, images, nearby maps, and even route planning based on the user's current location.

Answers, Not Results

Unlike PC search, mobile search is highly purpose‑driven, context‑specific, and fragmented in time. Users expect swift, direct answers rather than long lists of keyword‑matched results or intrusive ads.