EEEK! My Business Has Workflows!


Is it time to replace your system workflows with social business tools?



‘Subway’ by George Tooker, 1950


When pundits assemble the unwashed to preach the Gospel of Social Business, they often warn against the antisocial  forces within the enterprise. They’ll talk about diffident executives, hostile middle managers, unaligned business case, the folly of tool evangelism and weak community strategy. These are all valid roadblocks. We’ve certainly heard about them enough.

Yet there are more basic challenges to enterprise collaboration  These are the ‘data processing’ assumptions still coursing in the brains of business people. They cling to the old ways. Some examples:

  • Yes, they see the value of social business networking, but they still want a web site.
  • “Wikis? Well, sure, but how to fit one in the division portal?”
  • “We must account for exception processing; can we fit this in a Lotus Connections screen?” 
  • Communities? Yes, they seem to be better than our department Notes database, but can we move our project approval system to SharePoint?”
  • “Can I get my e-mails auto-posted to a blog somewhere?”
  • “How do I replicate these things so I can get them on my hard drive?”

Workflow systems are the migraines of social business deployment. Organizations want automated workflows because they enforce a chain-of-command. If a company succumbed to Six Sigma, it may have hundreds of stillborn, FMEA-spawned, automated swimlanes.

The enterprise still wants workflows because they promise the illusion of control.

Networked collaboration is the business antipode of workflow-based systems. Individuals make their own decisions in collaboration tools. Information pathways aren’t designed; they appear in an organic fashion, from the need of humans to connect and share. A social network pulses with information in an efficient way; it’s an amalgam of publication and alerts.  But the most important distinction between a designed, enforced workflow and a social business network is trust. A workflow doesn’t have it; a social network requires it. If trust is essential for good collaboration, then we can say a workflow is collaboration-free.

Reasons why workflows don’t flow or work:

  • They are fragile. Anything new (people, responsibilities, organization changes) shatters a workflow. They can’t adapt when the organization must change. They never last.
  • Workflows support the status quo. They are hierarchical.
  • Workflows require training. This means they are not intuitive.
  • They are expensive. In addition to the software costs, there are the additional costs of training.
  • Workflows are antisocial. People aren’t supposed to engage each other. An automated workflow is really forwarded annotation.
  • Directional flow is the opposite of collaboration. Work flow is a one-way, recursive information shunt.
  • Workflows lengthen tasks. People don’t know when to respond. Workflows add inefficiencies.
  • Workflows require design. They annihilate fluid interaction.
  • Workflows are optimized for the organization, not the individuals in it. Humans will always seek to optimize for themselves and sub-optimize the workflow. They do this all the time.
  • A workflow design is by its nature a compromise between the stresses of time and task. They dumb-down the personality of the individual, turning them into automatons, treating them like process equipment in a manufacturing line or items to be measured.
  • Workflows’ highest end-game is compliance. Staying in the lines is not the path towards growth or creativity.

Workflows impede collaboration. It’s time for the enterprise to rethink the value of automated workflow systems. There are places for automated workflows, especially when the enterprise has lazy, sneaky employees it can’t trust. But if it is blessed with hard-working, informed and smart people, perhaps it is time for the enterprise to throw those old workflows away.

Think about it. Are there inefficient workflows in your enterprise? Could they be done another way?







  1. Going to have to disagree with you on this one but I suppose it depends on the definition of ‘workflow’. I agree that workflow cannot deal successfully with complex evolving areas but it can give you consistency where it is required. The formal side of things where compliance is required and an audit trail needs to be kept. I think another dimension to add to this is the size of the enterprise you are talking about. I can see in a small organisation where someone is interested in putting in a tender for a job that they can socially distribute that decision. But in a large enterprise where you are bidding for a multi billion dollar contract that can have a lot of risk and multiple areas of expertise required just to make the decision to bid, then consistent principles and checkpoints that allow for exceptions, can minimise that risk and give a better outcome.

  2. Cory, these are all good points and I agree with you that the examples you site might benefit from workflow. Compliance and validation practically mandate predictable systems workflow.

    But, you’ll have to admit large enterprises have overdone workflow. It’s not the complex acts of creation that suffer under workflow, it’s all the little department approvals for mundane things. I’ll contend workflow is a kind of corporate ‘dark matter’ — it’s everywhere and no one can see it. Let’s see: project requests, trip approvals, equipment check outs, sample fulfillment, performance reviews, training assignment… It’s all over the enterprise.

  3. Why do I feel like I need to enter a 12-step program for old school workflow now?

    “My name is Kelly and I do see purpose and value in some structured workflow.”

    While I mostly agree with John’s thoughts and Susan’s hearty challenge, my vision of enterprise nirvana allows organic evolution of process and collaboartion to coexist peacefully with some structured framework for collection of and analysis of key metrics.

    As an example, many companies have a very clearly defined – and necessary, flow for terminating employees. While they may wish to invite further commentary from managers and coworkers, the main goal of the HR workflow is to ensure that they can ultimately prove ‘just cause’ for the termination, and may need to rely upon easy comparisons to other similar cases in the past. It doesn’t seem to be such an evil thing when finely crafted policies drive certain processes that don’t invite much of a social element or creativity, and many HR tasks don’t.

    Why does it have to be an either/or? Certainly we’d all like to see more creative & efficient ways to allow for more fluid collaboration, but that also poses very unique challenges with roles, rights and security issues, which is a big part of why we’re still largely dependent on old school, old style workflow.

  4. Great post. “Directionally Correct” as we’d say where I work!

    Working in IT, I will say that whenever one of my colleagues comes in our direction and says, “I am working X and I need a system that does abc”, I reflexively cringe. Inevitably, it involves codifying some supposedly rigid workflow involving defined steps that must be executed in a specific order, involving specific artifacts, involving specific people in specific roles.

    Then, in week 1 of operation, the steps/artifacts/people/roles involved all change. Given that the tools that we use are only extensible by wizard-like IT folks, this really is just a recipe for two things:
    -Job security for IT folks
    -A persistent state of frustration for regular users/people.

    That being said, those workflows MUST exist in some cases. But not in all. I actually blogged about this last year, using my personal experience to outline cases where “Lightweight Workflow” actually works. Here’s that post:

    In a nutshell? Post an article. Ask for comments. Put a “done” tag on it when you are satisfied that the process is complete. You’d be surprised in how many places that this pattern can be applied.

  5. Indeed, Brian is a card-carrying practitioner of ‘work that flows on conversations’. It’s the current ‘untapped potential’ that lies before many to explore.

    As well, Kelly brings up a great point — it’s not either/or, it’s both. How do we know? Because of the optimal design model — the Design Thinking continuum (scroll to graphic The ideal solution will look to optimally leverage all parts of the continuum. Some stuff like termination protocol needs to be a bit more ‘binary code’ or ‘algorithm’ for consistency.

    But to truly tap human potential (where it is available — where individuals have not been so beat down that they still have their brains turned on — requires operating in the heuristic, where edges are just barely discernible, where structure is minimized but also carefully placed.

  6. Are workflows the real culprits? Not that sure…
    I just wrote a post about the issue and think that we must take a higher view on workflows: instead of trying to tie people up in automated processes, we should look into behavioral patterns, then sketch transient ad hoc workflows which support them, adaptive enough to be modified or jettisoned when needed.
    To take on Kelly’s example: imagine that suddenly, a bunch of leaving employees forget to give their badge back. Traditional corporate reaction would be to secure the termination process, for instance by forbidding some action before the badge is back. But what is happening really? The problem is not necessarily with the workflow, it might be HR employee’s fault, door’s dysfunction which allow anybody to enter without badging, teleworking development, or even this new direct access from the parking lot…
    Dealing with customers is an even more workflow-averse realm.. Instead of Instead of narrowing our focus on workflows, let us instead broaden it on emergent patterns. Adaptive case management might well be the missing framework we need.

  7. Back when the earth cooled and I was a new employee, there was very little work-flow automation. There really wasn’t software that could handle it. Processes were simpler then and the approval chain was much shorter. Most workflows meant carrying paper and pen to 2-3 people to get signatures, but also to present or discuss what needed approval. This was collaboration within a workflow. Since those workflows were shorter, they were more efficient. Since actual conversation happened, there was a pass-along of knowledge, too.

    The ability to quickly create complicated, lengthy workflows allowed departments to to create them in multitudes. Ironically, this was under the guise of productivity. It’s almost Orwellian than the stated purpose was the opposite of the actual result.

    The e-workflow tools increase the number of workflows and workflow complexity.

    Social business tools allow people to recover the more natural way to seek approval and pass information. And perhaps it will eliminate some workflows entirely.

    Just a thought to consider — automated workflow is nearly nonexistent in our private lives. Transactions between us and enterprises we do business with are short and simple. We would not tolerate anything otherwise.

    So, back to my post, I did not say we should eliminate ALL automated workflows, but the enterprise and the people who dwell in it would all gain if MOST of them went away. I know as a practitioner, I would certainly be happier.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s