# 20% Local Intent Number Is 50% Too Low

MIHMORANDUM NO. 1520 | June 4th, 2012Reader Comments (2)

Mike Blumenthal pointed me to this Wall Street Journal article this morning which mentions the oft-cited stat that 20% of all * PC searches* have Local Intent. I just have a quick comment that I’ve made to several people in meetings recently but don’t think I’ve blogged about.

## Even based on information revealed directly by the search engines, the overall number cited for Local Intent searches should be **30%**.

**Here’s why:**

In my five minutes of searching, I couldn’t find a definitive stat as to the percentage of mobile searches (Martijn, maybe Comscore could help me out? :) ). But does anyone doubt that this will be **anything less than 20%** by the end of the year? That’s certainly what most projections were back in 2010. Google itself states that mobile search has quadrupled since then. And multiple studies peg mobile paid traffic around 25%, suggesting that if anything, that 20% number will be too low by December.

Google says 40% of mobile searches have Local Intent; Microsoft cites 53%. Let’s split the difference and say 46.5%.

Let me see if I remember my 8th-grade algebra lesson:

.20(x) + .465(.20(x)) = **.293(x), or 29.3%.**

David, a few more statistics supporting your theory:

– 30% of desktop search in the US has local intent, 40% in the UK (according to Yahoo)

– 77% of mobile search has local intent (according to xA)

I think the number “20” is very approximate and is, as Google said, “safe” number. I suppose the percentage quoted by the Yahoo ex-executive is very accurate. The problem here is that not all local intent searches carry any location data. That’s what Yahoo did – they took into account searches such as for “dry cleaning”, or “plumber” and counted them as local.

Thanks for this post and for the links. It’s very helpful as I’m figuring out the most accurate way to understand the impact of local search.

I do suspect your “8th-grade algebra” is missing one step. See if what I’m thinking makes sense, given your (sensible) assumptions:

46.5% of mobile traffic has local intent. Mobile traffic is 20% of all search traffic. So, 46.5% of 20% is 9.3% (because 0.465 times 0.2 is 0.093).

20% of desktop traffic has local intent. Here’s what I think is the missing step: Desktop traffic is only 80% of all search traffic. So, 20% of 80% is 16% (because 0.2 times 0.8 is 0.16).

In other words, 9.3% of all search traffic is locally intended mobile traffic, while 16% of all search traffic is locally intended desktop traffic.

Add those two percentages together, and you get 25.3%, instead of 29.3%.

I do wish there were clearer base numbers to work with in the equation. On the low end of the spectrum, ComScore reported last July that across all Google, Bing, Yahoo, Ask, and AOL search properties, only 13% of searches had a local intent. On the opposite end, in February, Jon Schepke wrote in a Marketing Profs article, without citing a source, that “80% of all searches conducted via smartphone are for local products and services.” Thanks again for working to cut through the noise.