Which SEO tools track Google local packs across British cities?

What to take away

Why local pack rank tracking fails outside London

Google local packs behave differently depending on where the searcher stands. In central London, a dense grid of searches often returns the same handful of businesses. Outside the capital, the pack can change street by street, and sometimes postcode by postcode.

Bristol is a good example. A search from Clifton can return a different pack from the same query run in Bedminster, even though both sit inside the same city boundary. A tool that checks one central point will report one position and call it the city ranking.

Leeds spreads its commercial districts across the city centre, Headingley and the ring road business parks. Glasgow adds another wrinkle: Scottish search results sometimes differ from English ones for the same query, particularly for services tied to local regulation or public bodies.

Most rank tracking tools were built for national campaigns. They check a keyword from a city-level location and return a single number. That number is real, but it describes one point, not the city.

Local pack tracking also depends on what the tool counts. Some tools record the position of a business inside the pack. Others record whether the business appears at all. Those are different metrics, and mixing them up leads to bad decisions.

Personalisation makes this harder. Google weighs proximity, search history and device. A tool cannot remove all of that, but it can control location and device, which is where the useful comparison starts.

For multi-location brands, the practical question is not "what do we rank" but "where do we rank, and for whom". That is a coverage question, and it needs a tool that samples across a geography rather than a single pin.

How we tested named rank trackers in Bristol, Leeds and Glasgow

We ran the same set of local queries through five tools across the three cities. The tools were Semrush, using its rank tracking features, Moz Local, STAT Search Analytics, Local Falcon and Local Viking.

Each tool was set to track the same ten keywords per city. The keywords covered trades, hospitality, professional services and retail, so the packs were not all dominated by the same business types.

We fixed the device to mobile for the primary run and repeated it on desktop. We set the location to the city centre first, then to three outlying points per city: Clifton, Bedminster and Fishponds in Bristol; Headingley, Chapel Allerton and Holbeck in Leeds; Finnieston, Dennistoun and Govan in Glasgow.

We recorded four things per result: whether the business appeared in the pack, its position in the pack, whether the pack appeared at all, and whether the tool reported a different answer from a manual check on the same day.

The manual check was the control. One researcher ran each query on a clean mobile device with location services on, in the actual place, and logged the pack by hand.

Any tool that claims local pack coverage should survive that comparison. Where it did not, we noted the failure point rather than the score. A tool that is wrong in a predictable way is more useful than one that is wrong at random.

We also logged how each tool described its own local coverage. Several vendors explain local search in terms of reaching nearby customers, which is the right frame, but the product documentation often says less about sampling density than the marketing does.

The method matters as much as the headline claim, and a product comparison needs a named source is part of that.

Bristol results: where each tracker missed the local pack

Bristol produced the widest spread between the city centre and the outer points. For queries such as emergency plumbing and same-day dental appointments, the pack changed between Clifton and Bedminster in roughly half of our checks.

Semrush tracked the city centre accurately and matched the manual check most of the time there. At the outer points it reported the city-level position, which sometimes differed from the local pack by several places.

Moz Local performed well on business listing accuracy, which is its core strength, but it is not a grid rank tracker. It told us whether listings were consistent, not where the pack placed a business at a given point.

STAT Search Analytics handled the location variation better than the point-based tools. Its strength is sampling, and that showed in Bristol, where the pack is genuinely fragmented.

Local Falcon and Local Viking gave the most granular picture. Local Falcon sampled a five-by-five grid across the city and produced a heatmap that matched our manual checks closely. Local Viking used a coarser grid and smoothed over the Clifton and Bedminster differences.

Failure points in Bristol: point-based tools under-reported pack volatility, Local Viking over-smoothed, and every tool struggled with queries where the pack appeared on mobile but not on desktop.

A worked example shows why this matters. A Bristol dental practice tracked one keyword from the city centre and saw position two in the pack. The same query from Bedminster returned no pack entry at all.

The city-centre number was true and useless. After switching to a five-by-five grid, the practice found that its visibility dropped sharply south of the river, which changed where it spent its local budget.

Leeds results: grid tracking versus postcode-level rank tracking

Leeds suited postcode-level tracking better than Bristol did. The city has distinct commercial clusters, and a business with one branch in Headingley and another in Holbeck needs two separate answers, not one city average.

Postcode-level rank tracking gives exactly that. You define the locations that matter to the business and the tool reports a position for each. That is simpler to interpret than a heatmap and easier to explain to a client.

Grid rank tracking gives a fuller picture. It shows the shape of visibility across the city, including the streets where a business drops out of the pack entirely. For a single-site business trying to dominate one catchment, the grid is more informative.

The trade-off is cost and volume. A tight grid across a large city means many more checks per keyword, and most tools price on checks. A grid that is too coarse produces a map that looks reassuring and hides the gaps.

A five-by-five grid is 25 checks per keyword; a seven-by-seven grid is 49; a nine-by-nine grid is 81. Postcode tracking is one check per defined postcode, so six branches means six checks per keyword.

Typical published starting prices for dedicated grid tools are in the low tens of pounds per month for a small campaign. The per-check rate drives the bill once the grid and keyword list grow. Postcode trackers often start at similar monthly prices, but the bill scales with each added location.

In Leeds, the grid tools agreed with each other more closely than in Bristol. The pack was more stable across the city, so a moderate grid was enough. The postcode tools matched the manual checks at each defined point but told us nothing about the space between them.

Failure points in Leeds: postcode tools missed pockets between the defined points, Local Viking defaulted to a grid too coarse for a city of this size, and device switching changed pack membership for several retail queries.

Glasgow results: Scottish local packs and map pack volatility

Glasgow was the least stable of the three cities. Pack membership changed more often between runs, and several queries returned noticeably different businesses from the equivalent English search.

Some of that is expected. Scottish results can differ for services connected to devolved matters, and local packs for public-facing services reflect that. A tool that tracks only English locations will not show it.

Semrush and STAT both allowed Scottish locations to be set explicitly, and both reported Glasgow positions that matched our manual checks at the city centre. The difference appeared at the outer points, where pack churn was highest.

Local Falcon and Local Viking coped well with churn because they sample repeatedly. The weakness was timing. A grid run that takes hours can span a period when the pack changes, so the heatmap blends two different states.

Glasgow also exposed a device problem. Several queries returned a pack on mobile with three entries and on desktop with a different set. Tools that default to desktop reported a version of the pack that most searchers never see.

Failure points in Glasgow: slow grid runs blended pack states, desktop defaults misrepresented mobile packs, and location settings that only offered UK-wide or England-level options could not represent the city properly.

Which rank tracking tools held up and which failed

Semrush is the strongest all-rounder for teams that want rank tracking and local search data in one place. Its local SEO features cover rankings, listings and reviews, which matters if you are fixing pack visibility rather than only measuring it.

Moz Local is not a rank tracker and should not be judged as one. It is a listings and presence tool. Teams often buy it expecting pack positions and then discover it does something else. Its local SEO product is worth having alongside a tracker, not instead of one.

STAT Search Analytics held up best on location sampling. Its search analytics platform is built around flexible sampling, and that is exactly what local pack tracking needs outside London.

Local Falcon and Local Viking split the result. Local Falcon was more accurate and more expensive. Local Viking was easier to run and quietly wrong in the places that mattered.

Before committing budget, it is worth asking which tools are worth buying for a UK team for this specific job.

Settings UK teams should change before trusting local pack data

Most of the failures we found were settings problems, not software defects. Fix these before you report a local pack number to anyone.

Before you trust any of this, decide which metrics to track in SEO tools actually drive decisions for the business. Pack position is one. Share of pack across a catchment is often more useful.

Vendors should be verifiable. Google Search Central documents how SERP features behave, which is the reference point for any claim about local packs. When you score the evidence behind a vendor's local coverage claims, ask for the sampling method, not the marketing page.

Finally, ask what evidence should I check before renewal. A tool that cannot show you where it sampled cannot tell you where you rank.

The wider point is that local pack visibility is about reaching nearby customers at the moment they search, which is what local search solutions are meant to deliver. Tracking is how you find out whether that is happening in Bristol, Leeds and Glasgow, or only in the postcode where the tool happens to look.

Common questions