Budgeting for Enterprise Search
It's time for enterprise search vendors to give up the numbers. 

For some years now senior IT managers have had to cope with significant pressure on capital and operating budgets. At one time there was a lot of debate about IT spend as a percentage of revenue but that has now been overtaken by something uncomfortably close to zero-based budgeting.

Every enterprise system purchase or upgrade has to be justified in some way, even if (in the case of SharePoint in particular) the business case should be filed under F for Fiction. A significant proportion of the total IT capital spend will be on keeping existing applications alive. Most of these will be compliance-support applications, such as finance, personnel, treasury and asset management as without these the chances are that the accounts will not get audit approval. IT managers will have a pretty good idea of how much these applications cost because they have been involved in managing them for much of their career.

There is then a set of more specialized applications which are not directly related to compliance requirements but which can make an impact on operational performance. Among these would be enterprise resource planning (ERP), product life cycle management (PLM) and customer relationship management (CRM) applications. These applications have been around for some time now and there is quite a substantial amount of published information available on the costs of purchasing and implementing ERP, PLM and CRM software. Organizations that subscribe to analyst services from Gartner, IDC and Forrester (as three examples) will also be able to obtain guidance from them on typical budgets.

License versus Professional Services

An important element of the budgeting decision is what to allow for in professional services costs for installation and implementation, even if those services are internal. For a Web CMS or document management application services costs could be as high as four times the license costs. SharePoint is a really good example of this as the development costs can be quite substantial. Even just understanding what these cost elements can be is a challenge.

A strong argument put forward by the open-source community is that using open-source solutions means that there is no licence cost but that does not mean to say that there will be no development costs.

The Cost of Enterprise Search

For most organizations search will be a novel investment. No one will be able to find out what was paid for the incumbent system and no one in the IT department will have implemented more than perhaps one other search product.

Going to the vendor websites for information is a total waste of time other than for dtSearch. Even Google is rather reticent on costs. One page of the website talks about a cost based on document numbers but you have to explore the site further to find a definition of "document" that mentions that a row in a database counts as a document!

I know of many organizations that have selected the Google appliance because it seems to have a very well defined pricing model. To be fair to Google it does (sort of!) define the model but that is not the problem. The problem is that the organization has no idea how much information it has to be indexed and also has no idea about the rate of growth until the appliance calls time out and asks for more money to be put in the cash slot.

The open-source community does not seem to be much better. The key element of an open-source solution is the support costs. LucidWorks indicates that its support programs are delivered through a complete selection of subscription packages but to get the prices you have to email Sales.

At the point of developing a budget, an IT manager needs to have some idea of the range of packages and what the multiplier is (i.e. per server, per user, per instance). In addition, as with all enterprise software, there will often be a charge for additional software instances for test and development purposes and these (and additional hardware and middleware costs) can quickly mount up but are rarely listed.

Time for a Change

The search industry sees everything in ROI terms but seems to be very unwilling to specify what the “Investment” is actually going to be. It also creates the impression that its products can be installed with the minimum of effort and with no need for any support team. I can quite understand why vendors want to keep some elements of their pricing under wraps but the industry needs to appreciate that IT managers under pressure from users to improve search have to not only put some sort of number in the Excel budget sheet but make a case for not spending the amount on some other application, and do so way before they have an initial meeting with a vendor sales team.

Title image courtesy of wrangler (Shutterstock)

Editor's Note: Martin has been sharing a lot of thoughts on Enterprise Search. To read more, see Is There a Future for Enterprise Search?