Search has become prolific in everyone’s day to day life, with 93% of online experiences starting with a search engine and 60% of all website traffic coming from search. I’ve personally found it fascinating to see search engines develop over the last two decades from basic navigation engines getting you to the right website to the platforms we see today with common questions being answered directly in the search engine results pages (SERPs). What will the future hold for this ubiquitous technology?
Important in attempting to answer this question is to interrogate the statement “Future of Search”. The classic definition of “to search” is defined as, “to examine thoroughly in order to find something or someone”. However, if you think about your last search online, it usually started with a problem that you were looking to solve. The search was ultimately aimed at solving that problem. So, when searching online we should think about how consumers seek resolution for a problem. Secondly, when looking at the concept of the future, the main challenge here is the concept of predicting future innovation. In his 2014 book “How we got to now” Steven Johnson wrote:
Bees and other insects evolved the sensory tools to see and be drawn to flowers, just as the flowers evolved the properties that attract bees. Meanwhile, the hummingbird evolved a special form of hovering flight that made the extraction of flower nectar easier.
He observed that the function of insects feeding on flowers (innovation #1) is the same function as that enabled by the evolved musculoskeletal structure of hummingbirds who eat on flowers in the same way (innovation #2). So, whilst we have reference points for innovations now, we don’t have reference points for those future innovations.
However, we can see how the business models and innovations have historically shaped search. Google became the most successful search engine by quickly connecting users with content across the internet that solved their problems. YouTube has provided us with a plethora of How-To videos, unboxing and review pieces that have aided people in solving their problems. Amazon has connected users with the products they are looking to purchase at ever increasing speeds, ultimately solving their problems. On this basis we can conclude that the future of search will likely be led by any products, technology and services that are able to resolve consumers problems quicker and easier than those around today.
On this basis below are my seven predictions of trends that we will see in the future of search:
1) Search will be experience led
When search first started the experience was very much a text in, text out affair. Users would input their search query in text and have 10 text results returned to them by the search engine. This experience has improved over time through the introduction of rich search results with outputs taking the form of images, products, maps, news or videos. These universal search results provided new forms of media that improved the search experience. Bringing us up to today there are also now a plethora of new input experiences (voice, image and video) and lots of new output experiences (360 images, Interactives and augmented reality).
In this new world we have voice search (whereby the primary input and/or output from search is voice). Voice searches are already happening at a rate of 1Bn+ per month, demonstrating their value, however this number is dwarfed by the current number of text searches that are happening. This is because whilst Voice Search has a 95% input accuracy, currently the output is still of questionable value a lot of the time. Once the usefulness of the output from voice search increases, we can expect this experience to aid users in solving their problems and be rapidly adopted.
There is also visual search (whereby the primary input and/or output from search is visual), with some recent examples including Google Lens (using imagery as an input) which is aiding users to identify objects or the inclusion of augmented reality results for certain animals within SERPs. Visual search is proving so useful to users that 62% of younger consumers want visual search more than any other technology in search.
As these experiences improve in the future, search will be more experience led.
2) Search will exist everywhere
As we start to think about the potential of these new experiences, we can start to mix inputs and outputs (and more importantly the devices on which these sit).
For example, there could be a foreseeable situation whereby you would ask a search engine to “Show me a selection of blue shirts I can buy on my TV”, whereby the search engine would respond “Sending a selection of blue shirts to your TV”. Browsing and refining your selections further on the TV.
Or there could be a situation where you could ask a search engine to “Send the directions to Tom’s house to my car” and the search engine would respond “Sending the driving directions to Tom’s house to your car”. Thereby saving valuable time when you get into your vehicle.
In fact, the role that these other connected devices will play in the search landscape could be dramatic. If we look at the units shipped combined with projected sales forecast for Desktops / Mobiles and other Connected Devices, we can see that whilst Mobile is the dominant device currently, it could be that by 2022 it is overtaken by the sheer volume of other connected devices.
As these devices become more prevalent, search will exist everywhere.
3) Search will be more personal
Search engines are currently undergoing a transformation, and this is related to their definition. At a basic level they are designed to return content from a database based on the search query the user has inputted. However, for them to develop any further in the future a transformation needs to occur. That is in additional layers of Contextual understanding and personalisation for the user. With this change we will see a move away from search engines and towards personal assistants. A move away from returning content from a database and towards seeking resolution for your problems. Technologies such as Google Assistant, Amazon Alexa, Apple’s Siri and Microsoft’s Cortana are all in their infancy but will eventually power everything around us. Think Google.com taking us straight to a conversation with Google Assistant or Alexa being served to you rather than Amazon.com and you are close to the possibilities of what may happen.
This level of deep personalisation will require users to be comfortable giving away this personal data for a better service. However, based on historical reference, consumers will see the value of doing this if it ultimately assists them.
4) Search will be more action based
If we now look at the inputs and outputs for personal assistants, we can see a challenge still. Ultimately the output of a personal assistant still requires users to take additional steps in order to solve their problem, whether this is onward browsing to a website or additional filtering of the search. Consider this instead, could a search engine solve the problem as an action rather than requiring the user to take other steps?
This is already happening within Google Search Engine Results Pages (SERPs). For search results such as “what is the time?” or “who were the Mayans?” Google is already answering the results directly in SERP. However, consider a situation where search engines can solve more complex problems, for example “Please order me a brown hair dye to match my usual colour”, “please order me an Uber”, “Please buy some toys for my daughter for holiday”. With access to personal data, calendars, address books and previous shopping habits a direct action from all these searches is possible.
As personal assistants can remove barriers to solving problems, search will become more action based.
5) Search will be more conversational
In the natural world, conversation between humans is the , the one-sided experience that we currently have with search whereby input results in output (or action) will ultimately be limiting. Conversation is required for additional context as well as refinement of the search.
Some level of conversation can already happen, when searching within the Google Assistant for example, you can ask about a celebrity. You can then ask who their partner is or who their children are. It uses context from previous searches in the conversation, rather than requiring you to repeat who the celebrity is that you are searching for.
However consider this, with levels of personalisation and contextualisation you could ask who a certain person is that you’ve just met in a meeting, the assistant would be able to tell you and then you could ask follow-up questions such as “when did I last meet them?”. With publicly available LinkedIn data and access to your calendar this is a possibility.
As personal assistants become better at refinement, search will become more conversational.
6) Search will be more pre-emptive
With the outputs of search replaced with actions, the last evolution of search could come from the removal of inputs to search. By pre-empting what a user might search for, it would remove the need for a user to search in the first place, surfacing useful solutions before the user even realised they needed them.
Pre-emptive search already exists within the Android platform, with push notifications giving you information on upcoming restaurant bookings, hotels, flights or travel. With access to calendars and emails this pre-emptive search becomes possible. Another example is Google’s Discover platform that surfaces content before a user searches for it, based on their previous search history.
In theory this pre-emptive methodology could be applied to more complex searches, for example “your hair colour is starting to fade, shall I order you another permanent dye in the same colour?” or “as it’s raining and you are running late shall I order you an Uber?” or “It’s your holiday coming up, shall I order some beach toys for your daughter?”.
As search understands what users will need before they themselves realise, search will become more pre-emptive.
7) Search will run your life
If all these potential innovations are stringed together backed by AI powered personal assistants, could the following be possible?
“I’ve organised your Christmas card list, ordered for you, they are being delivered for you to sign”
“I see that the new TV box set is out, would you like me to phone in sick to work for you?”
“I’ve written the marketing plan for next year, for the new product I recommended you bring to market, I’m awaiting your sign off…”
Could Search run your life?