How to read a geo-grid heat map

Start with the number, not the colour

A heat map is a geo-grid with colours on it: green where a business shows in the map pack, yellow where it is close, red where it is nowhere. The colours make it quick to read and easy to misread. Every square should carry the actual position, because “4th” and “10th” can share a colour and mean very different things. Fourth is one good month from the pack. Tenth is not.

The positions that matter most are 1 to 3. That is the pack itself, the three businesses Google shows under the map before anyone taps “more places”. Everything else is a question of how far away from the pack you are.

Five patterns, and what they usually mean

Real maps are messy, but most fall into a handful of shapes. The shape tells you more than any single square. These are illustrations, drawn to show the shape clearly, not results from a client.

• The bullseye

  Strong at the shop, fading evenly in every direction. Distance is doing what distance does. The fix is more relevance and reputation so you hold on further out: category, services, a steady flow of reviews.

• The blank wall

  Not in the pack even at your own door. That is rarely distance. It usually means Google does not think you match the search: the wrong primary category, the service missing from the profile, or a suspended or unverified profile.

• The half map

  Strong on one side, gone on the other. Look at what is on the empty side: a river crossing, a town line, or a competitor with a strong profile sitting right there. Each has a different fix, so find out which before doing anything.

• The island

  A strong pocket far from the shop. Often an old address still sitting in directories, a page that ranks for that town, or a second profile you did not know about. Worth checking before it is treated as good news.

• The static

  Good and bad points scattered with no shape, and different next week. It often means an unstable pack. In some trades, garage doors above all, that is fake listings appearing and vanishing, which is a problem to report, not a sign your work failed.

The average-position trap

Many reports boil a map down to one figure, the average position. It is tidy, and it hides the thing you care about.

The better single figure is the share of points where you are in the top three: none of nine for Map A, five of nine for Map B. Better still, look at the map. The weekly measurement in our dashboard leads with that share and always shows the grid underneath it.

Comparing this month with last month

One map is a snapshot. Two maps on the same points and the same searches are evidence. When you compare, read in this order:

1. Did the method change? Different points, different searches or a different grid size make the comparison meaningless. Check this first.
2. Which squares crossed into the pack, or fell out of it. That is where calls change.
3. Did a block move, or single squares? A single square going from 3rd to 5th and back is normal wobble. A whole side of the map turning from red to yellow over two months is a trend.
4. Who moved in? If you dropped at a cluster of points, see who took them. A new competitor, or one that just rebuilt its profile, explains a lot.

Blank squares and bad squares are different

Two kinds of square are easy to confuse with a bad result:

• No map pack at that point. For some searches in some places, Google shows no map at all, for anyone. That is not you losing.
• A point that could not be measured. Sometimes a search fails. An honest map leaves that square empty and says so. A careless one records a failure as “not found”, and your map looks worse than your business.

We keep both apart from genuine “not in the top 20” results, because a guess dressed as a measurement is worse than a gap.

From the map to a plan

Reading the map is only useful if it changes what you do next. A rough order that works for most trades:

None of this promises a result. It tells you where the effort is most likely to count, and the next map tells you whether it did. For trades with a strong season, such as landscaping or roofing, start early: the map is slow to move, and a zone won in July does little for a May peak that has already passed.