Local Commerce by the Numbers: Yelp’s “Local Coefficient”

We don’t generally reciprocate the blogospheric practice of publishing posts purely on random items of interest, as our style gravitates more towards exclusive interviews or news items in local media.

But in the spirit of doing this more often, here is a recommended article by our friend Mike Ghaffary at Yelp. The contributed piece in Techcrunch yesterday covers what he’s devised and named the “Local Coefficient”. This is one input in his formula to settle the longstanding debate between the relative market opportunities of online (eCommerce, mCommece) and offline (bricks and mortar).

As we keep saying, the the offline segment is much larger, as it comprises the majority of U.S. retail revenue. eCommerce is only 5 percent of the total. But online is still important: Though these conversions continue to happen offline, increasing levels are being influenced online and in mobile: the phenomenon known as ROBO (research online, buy offline) or O2O (online to offline).

Of course that path to purchase is increasingly convoluted and weaves between of different screens, as opposed to a cut and dry progression from Google search to website, to store. One of our biggest objectives is to track the consumer behavior and advertiser spending throughout the parts of this path to purchase (ads, directional content, product info, mobile payments, etc.).

Back to Ghaffary’s analysis, the local coefficient defines different local business verticals by how likely they are to be disrupted by eCommerce. This includes things like the importance of seeing items or meeting service providers, as well as the availability of online substitutes. It reminds me of the methodology that governs the designation of price elasticity for any given product.

In addition to the local coefficient, the market sizes of all of these verticals are a key input for devising the total current and projected revenue opportunities in online commerce versus offline commerce. It’s an interesting read, and Ghaffary’s ultimate conclusion is that local offline commerce will be 3x that of online and won’t be disrupted significantly for at least the next decade.

It’s a compelling argument, empirically devised, and holds up to reason. We’ll be reaching out to re-connect with Ghaffary since the last time we spoke at our June MLM conference in San Francisco. In the meantime let us know what you think of his construct and the validity of the Local Coefficient.

ScreenHunter_01 Feb. 25 15.49

This Post Has 5 Comments

  1. Painters

    Just to help everyone fully understand this, it would be great to have a legend for the table.

    My guess is that:
    e = ecommerce
    t = ?
    s = ?
    L = Local
    O = Online
    M = Market Size

    O*M = Total online market size
    L*M = Total local market size

    So in this does “local” include the non-online sales as well?

  2. Mike Boland

    Great point. Rather than rehash the entire model, i just linked to the TechCrunch piece where all the elements to the formula are defined better. If you go to the following link, it explains it well. Maybe i’ll add something additional to the post itself too. Thanks for pointing it out.

    http://techcrunch.com/2013/02/24/brick-and-mortar-wins/

  3. Roswell Home Inspector

    Mike,

    Thanks for including the second link.

    It’s interesting to see where they came up with the numbers for the “plumber”. Why didn’t they include all of construction or service industries?

  4. Painters

    Thanks for the link Mike, it makes a lot more sense now.

    Plus, I was way off with my guesses as to what the formula was!

  5. Mike Boland

    Good question regarding the omission of other service verticals. I’ll ask Yelp when i talk to them next week. I bet it was to keep this list somewhat unwieldy for the purposes of his editorial.

Leave a Reply

Your email address will not be published. Required fields are marked *

15 − seven =

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>