“Best business practices.”  The first time I herd this phrase was Sophomore year of college.  I was in business class and we were seated in the round having a discussion about business strategy.  I don’t remember the exact case, but it was about a new company entering a crowded market.  The question was: how should this new entrant compete against the established players?

One of my classmates, who thought he was clever, stood up and said, “Why doesn’t the new business just copy what the other businesses in that market are already doing?”

The class giggled a little bit, because copying

other businesses was obviously not the right answer to the case.  However, the teacher surprised us.  He said, “We call that, best business practices.” – and went on to explain how copying other businesses is something that happens all the time and how it can be a good strategy (or part of a strategy).

To be fair to our professor, by his frame of reference, he was right.  He had spent his entire career working in more traditional industries that experienced a very slow rate of change.  Copying your competitors, in some established industries, was a sustainable strategy in the pre-internet disruption era.

However today – we live in a completely different world.  For the most part, copying your competitors has never been a worse idea.

In today’s world, there are really two kinds of businesses: disruptors and disruptees.

Copying a business that is in the process of getting disrupted is obviously a bad strategy because you too will eventually be subject to that same disruption.

However, copying a business that is doing the disruption can be an equally bad idea.  Especially in high tech, the competitive landscape and market are changing so fast, copying a competitor or set of competitors without deeply understanding the long-term implications of your decisions could land you in a very bad spot.

Don’t get me wrong, it can be very tempting to follow the herd.  As a rational person you figure, “Hey, all of these companies are making money – and I think we can do what they’re doing better.”  At the time, it honestly feels like a no brainer, especially if you’re operating in an industry with lots of other players.  How could all these other companies be wrong?  It’s sometimes not until you see the whole herd running off the cliff that you realize you were right not to follow them.

When you’re on the side of disruption you have to make decisions every day and consider the impact for the next five years.  It’s tempting to take the easy way out and copy other players in the market – especially if those competitors are making lots of money today.  However, you still need to think long and deep about the prospects for the next five years.

In some ways, that’s the secret to running a disruptive company:  you have to make decisions every day in service of the next five years.

If you do a good job – by the time your decisions reach maturity – all of those competitors, who at one point you were tempted to copy, will no longer be in business.

It’s Never Been a More Terrible Idea to Follow the Herd
Tagged on:         
  • Vic Lanzillotti

    I agree with all these insights but will add that in order to make decisions everyday and to execute on them effectively, there is still a “herd” that is important. It is one often overlooked by senior executives. It is their rank and file and middle management employees. Disruptive decisions and overall strategy have to be explained and re-explained to employees in order that they believe in the system and deliver with their daily activities which support the overall strategy. The employee herd must be deeply engaged in order to follow, to believe, and then lead.

  • Vic – couldn’t agree more. I was actually thinking about writing something similar next weekend. Something around the following thesis:

    What I’ve seen in my career is that if you don’t do what you suggest (diligently and clearly explain strategy to employees), employees will naturally start emulating competitors as sort of the “default” thing to do. If senior management isn’t clear with direction – over time – employee behavior will slowly start trending toward emulating competitors out of lack of clear direction. What do you think?

  • I’m not sure I really agree.

    I think I disagree with the (soft) equivalence between “best practice” and “following the herd.” Doing the latter is nearly always sub-optimal; doing the former is usually a great idea.

    However, it’s nearly always uncomfortable to follow best practices. That’s… sort of what makes them best practice? Doing more, harder work, with fewer excuses?

    Also, the best practice changes based on your company development and your situation. What works for Google _will not_ work for a startup; what works for a pre-IPO tech startup won’t work for a publicly traded one, etc.

    By the time “everyone” has caught onto an idea it’s usually good to be doing no-worse-than-that, but also much, much better in at last one focused area.

  • The principle your rebuttal is valid. However I think there are times when best practices can be a bad idea. The best examples I can think of today are when it comes to supply chain management. If you and I were to start a new clothing retailer (we’ll call it – Michael and Andrew’s Amazing Socks) – it might be “best practice” to sell our supply to a series of retailers who in turn sell to the customer. That might be what every other sock company out there is doing and it might be working great for them (this is the “best practice”).

    Here’s the rub though: as a start up, we’re not the established player in the market and (to your point about best practices being different at different levels of maturity) we will have a scale disadvantage vs. the established players. So if we copy them, we will be fighting a perpetual uphill battle. Therefore, if we were to start such a company, we would want to do something innovative, like compress the supply chain and sell direct to consumers (maybe even a monthly subscription model?) – assuming no one else was already saturating that space.

    The argument i’m making here is if you’re not doing something radically different with your business (particularly related to supply chain) – chances are you’re following a herd off a cliff.

  • > The best examples I can think of today are when it comes to supply chain management. If you and I were to start a new clothing retailer (we’ll call it – Michael and Andrew’s Amazing Socks) – it might be “best practice” to sell our supply to a series of retailers who in turn sell to the customer. That might be what every other sock company out there is doing and it might be working great for them (this is the “best practice”).

    Well… I think I’d struggle to call that “best practice” as opposed to e.g. a business model choice.

    If you justify the choices you’re making based on what other people are doing, with no regard to your situation, you’re probably not going to do well.

    I’d say examples of “best practices” in supply chain management include:
    – JIT inventory
    – Minimum-touch warehouse management
    – FIFO / minimize time from production to sale
    – Source local to manufacturing (reduce unreliability from shipping problems)

    If you don’t do any of those – I mean, JIT inventory is super-hard to do right – then your costs increase and it’s difficult to complete. You could price yourself right out of your business model choice because you’re not employing “best practices” that lower cost and improve reliability.

    I guess I have a different definition for best practice. There are *choices* you make that frame the “business problem” you have, and then there are methods/approaches to solve those problems. Those methods comprise “best practices”, relative to the context you’ve placed yourself in.

    In my day to day, for instance, I may run into a classification problem (e.g. true or false). It’s best practice to (a) have a hold out and a validation set, along with training data; (b) use a method that compresses the output space to be in the domain [0, 1] like logistic regression; and (c) make modification to your method based on the problem.

    It doesn’t stop people from doing the “wrong thing”, even somewhat regularly (not having a firm delineation between train/validate/test; throwing an unconstrained linear regression at the problem for feature selection; failing to account for collinearity in linear models; etc), usually if they’re not doing it very frequently or the stakes are lower.

    I wouldn’t say “expressing the problem as a binary classification problem is a best practice” — that’s a choice, driven by the data and the desired outcome. But once you’ve made that choice, there are best practices for achieving the best possible performance within that framework.

  • I think we’re agreeing – aside from my possible mis-use of “best business practice” jargon. Perhaps I should have paid closer attention in MB220 (or whatever class it was!)