If you’re responsible for your organization’s search technology then it’s probably not new to you that Google’s Search Appliance (GSA) and its associated services are shutting down in March 2019. If you’re still looking for a replacement search solution for your website, intranet or app, chances are you might be feeling a bit stressed at the moment.
You shouldn’t be.
Replacing hardware in your infrastructure can seem like a daunting task, especially if you’re replacing it with more hardware. Unless you’re fortunate enough to already have an expert familiar with the new piece of technology in your ranks then you’re likely to be facing a lot of consulting hours, time and ultimately capital expenditure to get up and running. When you add the risk of a potential project time and cost blowout, any hardware-based solution to replace the GSA should probably be ruled out from the get-go.
Thankfully, moving to a cloud based solution for your search needs can be much quicker and cheaper.
So, what key factors should you be considering when assessing cloud-based search-as-a-service providers?
Flexibility & Customization: Learn from the GSA’s drawbacks
The big yellow box had a tendency to feel more like a black box.
While there were benefits to running Google’s search technology from within your infrastructure they were often outweighed by the drawbacks. If you’re an IT Manager, CTO or CIO you probably experienced a fair bit of pain from your digital marketing and customer experience colleagues complaining to you that the GSA was hard to tune for result relevance. Figuring out why certain results were shown for certain queries was very difficult, and figuring out how to make adjustments to adapt results to your business needs was just as challenging.
Thankfully most Enterprise grade cloud-based options are far simpler to customize than the GSA. You should look for:
- Easy to manage and flexible rules to add and exclude content.
- Easy to manage and flexible rules to boost content and add synonyms.
- Detailed analytics that will allow your business stakeholders to analyze the success of your search functionality.
- Fast ingestion of your content whether via a crawler or API. Waiting for periodic crawls or slow data ingestion can leave your site content hopelessly out of date.
Cloud & AI: Future proofing your business
Cloud & AI. Software-as-a-Service and machine-learning. These are trends that you’ve probably become a bit weary of hearing about but which aren’t going away. Cloud technology should give you the ability to scale up and down quickly, control costs and make rapid adjustments to business requirements. When applied to search technology, AI should give your results greater relevance as the search system learns from customer interactions.
We know where the future is going but a lot of the cloud-based search technology available today is based on software originally written 20 years ago, carrying the constraints imposed by the hardware of 20 years ago.
So what should you consider when making a decision?:
- Does the system truly use machine learning and AI to optimize results based on your data and your users or does it have to be constantly manually tuned as your content changes?
- Is the system built specifically for distributed computing environments? Or, does it require its own special dedicated hardware in a data centre to run properly? If the answer to the latter question is yes, then make sure you check how this impacts your implementation budget, as well as ongoing support and maintenance costs.
Straightforward setup, maintenance and support
Your new GSA replacement should easily integrate into your existing tech stack.
When you move to a SaaS product you should be leveraging the product development the supplier had invested in the platform. This reduces your exposure to delivery risks and maintenance overheads.
A key part of this product development should be a straightforward integration process - the more configuration that can be handled through the platform’s administration tools, the less work for development and engineering teams.
Some situations you'll want to avoid include:
- Lengthy integration timelines for standard intranet or website search should be a cause for concern. You should be able to index your site(s) and integrate the new solution in hours or days, not weeks or months.
- Vendors that try to steer you towards letting them handle the integration and interface build instead of building it yourself. If you can negotiate an agreement that sees your vendor handle some of the work for you, great. But be wary of services that don’t seem to have a straightforward standardized integration process.
If the system requires expert knowledge and custom development for the initial integration, where will that leave you when you need to make changes or launch new search applications?
- Support costs that are not factored into a predictable monthly fee. The GSA wasn’t cheap and moving to a SaaS solution should allow you to save some money. Make sure it doesn’t come back to bite you later!
In summary, to choose the best GSA replacement in the short time you have left, you should focus your decision making on these three factors:
- Ease of integration. You shouldn't require expensive development, and it should be quick to implement.
- Flexibility of configuration. You should be able to configure it in-house.
- Cloud and AI suitability. It should be easily scalable, and intelligently order results.
Content, Marketing, and IT can all work better together with an easily configurable cloud search solution.
Final thoughts: Grasp the chance to rethink search responsibilities
We all know that Google’s Search Appliance wasn’t the easiest piece of kit to work with.
It’s unlikely your technical teams have to help your digital marketers when they are creating content experiences in your CMS, but you did have to support them in optimizing search experiences if you were using the GSA. It put technical teams in the uncomfortable position of having to administer a tool that was directly delivering end user experiences, which meant that search remained an IT responsibility.
That should no longer be the case. A modern search-as-a-service solution should allow your marketing and customer experience team to configure and optimize your search solution to meet your customer and business needs, while your engineering teams handle the interface integration.
Take the migration from your GSA as a chance to reconsider where the responsibility for search lies within your organization so your respective teams can focus on where they deliver the most value for your business.