Image's source
14 Aug 2016

Real World Technical Debt

tl;dr Think about a poorly managed construction store, inventory is not alright, products are not the right place, displayed labels does not match you stock namings, you are very close to what technical debt is and why it can be killing you.

Technical debt is a topic that is both very hard to explain and underestimated.

Basically people not coming from a technical background in the startup industry are usually coming from a areas where processes are not mission critical, think of it, marketers, designers, what matters is to organize that event, finish that design, make this printable, those tasks are extremely resilient to poor structuring. You can be good at organizing, you can be bad organizing , it does not ruin your performance as a whole.

Think about healthcare industries, construction work, where it is all about playing in team. And if you miss it, you see it quickly, poor information can kill your patient, bad synchronizing of your construction can lead to dozen of people waiting for the materials or a collapsing building.

As a consequence, I guess that we have a bias giving low consideration to processes.

A friend linked me to Conway's law which states this very appropriate principle:

Organizations which design systems ... are constrained to produce designs which are copies of the communication structures of these organizations.

My story starts with me willing to put some PVC tiles in a room floor with my father. Knowing the brand has bad reputation, we still go to a construction store because it was closer.

We look at a few product, we see opened product boxes, with mixed content (two different models in the same box).

We pick a floor with a nice kind of grey pattern, "wood grey", we go to the first employee around this section.

"We need to cover 14 m²".

"Let me check the reference number of the product. You will need 9 boxes, I still have 12 there, that should be alright".

"Fine, let's go!".

Before being able to leave his desk, the employee is stopped by a colleague in order to solve a "major" problem, a product was sold at "3.98 €" when the label said "3.90 €".

After a few minutes, we finally go back to the shelves, the employee finds the displayed sample of this model, and look for the box, "wood grey" it is. But no box has that label we have "light grey", "dark wood", "stone grey" and others. Well, after checking the product inside the boxes, it seems that "wood grey" is "light grey". He puts aside 6 boxes, one of them has been opened and repackaged with some store branded duct tape. We are missing 3, he starts to look around, no sign of them, he calls a colleague... "Oh yeah maybe they are on top". He goes around and comes back with some machine to fetch them, here are two, one box is opened. At this point my father taps my shoulder and tells me something very smart: "It is perhaps not that bad they don't find it, let's go elsewhere!", he reminds me of the mixed models in an opened package, on the duct tape one we have may have the same kind of problem, or broken tiles, or something else, and we will only discover it only while working.

The employee is not disappointed, silently aknowledging that this is good choice.

On our way to the other store, I was thinking about how I would fix this mess:

  1. First make an exhaustive inventory of what you have:
  • What number of each product?
  • How many pieces are broken / opened?
  1. Understand what causes those inventory discrepancy and rework them

But basically this simple plans misses a few points:

  • Inventory requires a substantial amount of man work, especially because products are mislabeled.
  • The inventory is just a symptom, your organisation is the disease, your processes, your place, your employees, management, market, everything has been contaminated and contribute to the symptoms:

    • Your process are letting this mess stands in peace.
    • The place is probably structured in a way that storage is inconvenient, unsafe or manipulated by customers, and inventory gets "corrupted".
    • Your people are just accepting this as the normality, the one willing to fight have left, the others have given up.
    • Management has seen it working, and believes it will still do.
    • If you look around the population around you is composed of retired people not really in a rush to get their product, all professionals are already dodging you, the market share wounds will take time to heal.

When I see technical debt, it is rarely or never an unexpected outcome, it is the deserved product of a broken environment, like the construction store with a few alterations: faster, more products, store employees are replaced by software components and are very fast to leverage what good or bad they are causing. You get the picture.

You do not fix the technical debt by only altering the software, you need to reshape the organization. Think again about the store example, it is probably easier to get this done from scratch rather than having only the "low expectation" market segment, the broken inventory, the tired staff and the accepting management.

Some say that "a startup broken at its foundation will not survive".

I would not be so strict in my judgement, I have seen some surviving to the sickness, but the very rare that got rid of the disease dit it by scraping a high number of "key" features to keep only the good bases. The only constant is that this technical debt gets superexpensive in the end.

Fräntz Miccoli

This blog is wrapping my ideas and opinions about innovation and entrepreneurship.

Ideas expressed are here to be challenged.

About me The Dark Side Twitter