Improving Intranet ROI by improving your intranet search

Prescient's consulting methodology always begins with an assessment phase. We need to know 'where you are now' before we work on 'where you want to be' and then carry out the gap assessment between the two states. If there is one thing which is a common thread between all intranet assessments it is search. In fact if there is one statement which is echoed in stakeholder interviews or focus groups all over North America it is "search sucks" !

So if you want to improve the Return on Investment (ROI) of your intranet, a low hanging fruit to pick off is search. If you improve the ability of your staff to find the information they need, then you will gain improvements in efficiency. To illustrate this, let’s examine one of many studies undertaken in this field.

In 2006 Accenture surveyed 1500 managers across many and varied organisations in the UK and USA.

Highlights from the results of this survey included:

  • "Managers spend more than a quarter of their time searching for information necessary to their jobs, and when they do find it, it is often wrong."
  • "Managers spend up to two hours a day searching for information, and more than 50 percent of the information they obtain has no value to them."

(Source Press release 4th Jan 2007)

Obviously these examples are generic, but let us build on these statements and develop a similarly generic scenario based on them. Our imaginary global organization will have 15,000 staff worldwide, and so have 5,000 managers of various grades. An average salary figure for these managers is $100,000 per annum, divided by 235 working days of 8 hours duration that gives us a rough average figure of $55 dollars per hour (including a notional amount for overheads).

So 5,000 managers searching for two hours a day, roughly equals a cost of $550,000, so multiply this by our 235 working days gives us a total of $129,250,000 – nearly 130 million dollars a year ! To stick with our simple example and rough math, let’s say an improvement of 50% is feasible, reducing the searching to 1 hour per manager per day; we are instantly halving this cost and thus saving $65 million.

How could we achieve this saving? Well I am not going to suggest that you go and ‘speculate to accumulate’ by spending half a million or more on the latest search engine technology from Autonomy or FAST; at least not as the first step.

Often, when we delve into the detail behind the numerous “search sucks” statements, we generally find two areas of deficiency:

  1. Metadata – basically the complete or almost complete lack of it attached to the organizations content
  2. Irrelevant search results – this is not normally due to any deficiencies in the search engine technology but in the management of search in general, and the configuration of the search tool in particular.

Let us consider the second of these elements first, largely because the whole of Prescient’s methodology is based on the importance of strategy, planning and governance to the success of an intranet (see Prescient’s Approach to Intranets). It is very rare that we find an intranet strategy fully aligned with the organizational strategy, and even less common to find a lower level strategy (or sub-strategy) dealing with search specifically. However if your users feel they are getting largely irrelevant search results this can be directly associated to a lack of strategy, which in turn will have impacted on any planning – if any has taken place at all, and the governance and eventually ongoing day to day operational management of your search technology.

It is the strategy piece that brings in the link to metadata. Does your organization have a separate metadata strategy? Perhaps it is lumped together with taxonomies, thesauri and other elements in an information management strategy? Hopefully you have a full set of standards, that set down what metadata elements should be captured for each type of content or piece of information, be it in an Oracle database, or Word documents. Metadata is of course useful for many things, such as records management or long term archiving, but in this context it helps to improve the ‘findability’ of content or information. It may be surprising to you, but quite often when Prescient consultants are undertaking their assessments of intranets, we find that there are no metadata standards at all, and that even if content is being managed in a good CMS, there will still be no metadata appended to it. Hardly surprising that staff cannot find what they need to.

So if you successfully created and published your strategies, have you moved onto the planning stage and created an operational plan for managing your search engine? Quite often the search results will be labeled as inaccurate and irrelevant, and even the statement “completely useless” is heard again and again. This even happens when modern technology has been deployed. So why is that? Well normally because inadequate planning has taken place for the deployment of the search technology, or because the organization has fallen for the vendors marketing ‘fluff’ and reduced the ‘search team’ to half a FTE System Administrator.

We often see results sets that are full of content that the user does not want to see because the features in the search engine software that allow ‘scope’ to be set have not been configured, or the search engine has been set up to create one single enormous index that is not split into different ‘collections’ of content. So even though you want to search the intranet, your results are hidden beneath thousands of ‘hits’ brought back from your sixteen external customer facing web sites.

Similarly the interfaces have not been configured to use all the features possible, the deal with the vendor has not been levered to provide a test server, the ‘best bets’ have been configured at setup, and never touched since, etc. All of this is quite often because the governance aspect of search becomes a black hole. There is no clear ‘business owner’ of search, and thus it is left to IT. IT quite rightly are happy that its running, crawling what it should be every night, and that they can still back up the index (“what do you mean the search engine index is so big it’s making the backups fail ?”).

How do we achieve our 65 million dollar per year saving then? By focusing on a strategy, planning and governance model for your search facility, and by examining the improvements that can be made to your existing search engine by providing a multi-disciplinary ‘team’ to manage search. You might only need 2 or 3 people, and they might work on other tasks, perhaps they would be part of the wider intranet team. However they should examine the license deals with the vendor, ensure IT is selecting the right configuration options, be a focal point for ownership of search issues and problems, and undertake analysis of what people are searching for, and more to the point, whether they find it.

Of course in an economic downturn, it may seem a little naïve to suggest increasing head count, after all staff costs are usually the first thing to come under the microscope, but if you could make a saving of $65m by raising the wages bill by less than $200K per year, and focusing and configuring and managing the technology you already have, does that not make more sense than spending an awful lot more on a brand new search engine, that is still going to be ‘managed’ by a part time Sys Admin ?

Obviously as I have stated, the math in our examples is both simplistic and generic, but Prescient can help by advising you on how to get the most out of your current search technology and how to ensure the strategy, planning and governance issues are managed in a way that will improve your intranet ROI.

To get a good grasp of how search fits in the bigger picture, AIIM (the ECM association) provides Information Organization and Access training courses that cover metadata, taxonomy, classification schemas and search technology. See:

Additional Reading:

Find more Prescient articles on improving ROI

More articles on Intranet specific subjects