Eight Steps to Success When Designing Document-Centric Workflows

Document Services

For organizations such as banks, insurance providers, law firms, or government agencies looking to automate common business processes, there are two major technology choices to be made: a Business Process Management system (BPMs) and a Document Generation platform. The BPM handles data routing, rights management, stage approvals, credit checks, etc., while the document generation platform handles data-gathering and document production.

Because workflow requirements across organizations—even across divisions or business units within an organization—vary so dramatically, it’s difficult to offer advice on overall workflow design. However, the document generation requirements of workflows within organizations tend to be more standardized. With this in mind, here are eight steps to success when transforming your real-world, document-centric business processes into automated workflows:

1. Don’t settle for less than inline document preparation

Virtually none of the major BPM systems have document generation capability (at least, not adequate capability for complex documents, such as loan documents). Consequently, BPM vendors may just assume that you’ll be generating documents using some dark, offline process and then passing the finished documents into the workflow for routing and approval.

But the idea of document-centric workflows that don’t actually have the ability to generate documents seems . . . well, like somebody’s missing the point. At any rate, that’s why you need an enterprise-grade document generation platform to work in unison with your BPMs—to gather all necessary information for the documents themselves and to generate the documents from within the workflow. Anything less is tantamount to a duct-tape-and-super-glue kluge for what could otherwise be an elegant solution.

2. Choose an enterprise-grade document generation platform

There are lots of lightweight, template-building utilities on the market which may be an upgrade from the native document generation functionality in your BPM. Such lightweight applications are typically designed for merging simple variables into word processor documents and controlling the inclusion of language blocks with simple business logic. But such applications may not allow for server/cloud deployment (critical with a BPM integration) and, those that do, may not have the type of flexible APIs necessary to integrate with a robust BPM.

In contrast, enterprise-grade document generation platforms, such as HotDocs, allow for either on-premise server deployment or a cloud-based (hosted) deployment. Likewise, such systems will have powerful APIs, enabling you to integrate interviews (wizard-like, interactive data-gathering sequences) into a workflow wherever they’re needed.

3. Focus on data gathering

Virtually all BPMs have utilities for creating simple, data-gathering forms. And in many types of workflows, these simple forms may be adequate. However, in any workflow that includes complex document generation (such as loan origination workflows), BPM forms will not likely get the job done.

Automating the generation of complex documents requires ultra-sophisticated data-gathering forms, which can only be designed and created after the documents themselves have been automated. Put another way, you won’t know which questions need to be asked to generate the document(s) until you’ve merged variables and business logic into the documents themselves. The variables you merge into the document serve as question fields in the data-gathering forms.

Further, a sequence of data-gathering forms (an interview) for complex legal documents must mirror the internal structure and business logic of the underlying documents. Otherwise, system users would be presented with dozens—even hundreds—of irrelevant questions. (See “Building Interactive, Data-Gathering Forms for BPM-Defined Workflows”). To automate the documents and to mirror the structure and business logic in an interview, you’ll need an enterprise-grade document generation platform, such as HotDocs.

And here’s the key point—since you have to use the document generation platform to create interviews that are sophisticated enough to gather data for your complex documents, you might as well use the document generation platform to generate all data-gathering forms in all of your workflows.

4. Build safeguards into the data-gathering sequence (interview)

One of the key benefits of any type of business process automation is the codification of best practices. In the case of document-centric workflows, adherence to best practices doesn’t just improve efficiency; it also safeguards you from making costly mistakes.

As was already mentioned, elite document generation platforms have, arguably, the most powerful interviewing capability available, a byproduct of the need to generate data-gathering sequences for sophisticated legal instruments. Such platforms will allow you to design wizard-like interviews that lead workflow users through the process of gathering all the necessary data. Question fields can include validation logic, and help resources can be attached to individual questions. Workflow designers may even want to call outside web services to help loan officers get the correct answer in each field. One global bank tests each answer as it’s entered into a field against the bank’s own internal guidelines and policies. Answers are color coded based on how closely they comply with policy (For example, answers that don’t comply are colored in red. (See “The Royal Bank of Scotland Case Study”).

Given correct and complete business logic in document templates, the key to generating perfect transactional documents is correct data entry. Safeguards in the interviewing sequence are the key.

5. Account for both types of documents (text and PDF)

Depending on the types of documents involved in your workflows, documents generated out of your workflow may include text-based documents (word processor documents) as well as PDF forms (graphical forms having fields and checkboxes). When choosing a document generation platform, make sure it can handle the complex automation issues of both types of documents.

One key functionality to look out for is component sharing, which means variables used in the automation of, say, a text-based lending agreement can be used in any other document template as well, including PDF forms. This approach allows you to update a component (variable) in a single place and have it updated globally, throughout all documents and PDF forms. Furthermore, it allows for bank staff (loan officers, for example) to enter an answer a single time and have it appear throughout all documents and forms.

6. Account for deep data integration with enterprise storage systems

One of the early stages in your workflow should be a check of enterprise storage systems to see if you already have existing, relevant data relating to the transaction at hand. Given that you’ll likely want to use a document generation platform to create data-gathering interviews, make sure that the platform you choose can integrate with your existing data storage systems. You should be able to populate fields in your workflow interviews as well as write new data (entered during the interviewing process) back out to data storage systems.

7. Focus on data routing rather than document routing

Given that most BPMs don’t have sophisticated document generation capability, most BPM vendors assume that you’ll be generating documents outside the workflow. In this scenario, it follows that the finished documents will be passed into the workflow, where they can be routed for various stages of approval.

But gathering all the data necessary to generate documents within the workflow (as well as for any other data needs within the workflow) facilitates a much more efficient approach—routing data for approval, rather than documents. In the Safeguards section above, I mentioned a bank that uses an automated workflow for loan origination. The workflow color codes answers in an interview according to an answer’s compliance with bank policy. The complete answer set is then routed from relationship managers (the staff members that interface with customers and actually gather loan data) to a loan approval committee. The committee can then quickly review the color coded answer set, rather than spend hours reading complex loan documents. This approach has allowed the bank to reduce the origination process for commercial loans from three days to under an hour. And the best part is that the bank now makes fewer bad loans.

8. Plan for the future

The workflows that you’re envisioning now may be simple or complex, incorporating your current wish list of features and functionality. But there’s a good chance that what you’re envisioning today may be rudimentary compared to what you may want or need a year or two down the road.

Don’t get drawn in by the siren’s song of simplicity. Choose a BPMs and a Document Generation platform that have the features and functionality to handle virtually any level of technological sophistication. (See “Fixing the ‘Simple’ Problem”) That way, you won’t have to abandon a technology investment a couple years down the road and make yet another investment that will actually be adequate for your long-term needs.