When a scenario leaves problem markets, the Divestments page works out the smallest set of sites that would clear them. You tell it how much overlap you're willing to live with and which sites matter most; it solves for the rest.

You need a saved scenario first - see how to set up a Local scenario.

Step 1 - Pick the scenario to resolve

Scenario to resolve lists the project's saved scenarios. Choose one and Sloc shows what it was run on - the data version, the subset, the parties and the catchment - so you know which analysis you're remedying.

Step 2 - Set how much overlap you'll accept

Maximum risk tolerance is the bar the plan has to clear:

• None - no overlaps - no overlap may remain anywhere. The strictest, and the biggest remedy.

• Low delta risk - only markets where the merger adds very little may remain.

• Low risk - resolve the high and medium risk markets.

• Medium risk - resolve only the high risk ones.

The bands come from the thresholds you set in Metrics & risk, so set those before you come here.

Step 3 - Choose the remedy approach

• Do not acquire Target sites - carve sites out of the deal before it completes.

• Divest Acquirer sites only - keep every target site and sell the acquirer's own.

• Divest Acquirer or Target sites - whichever resolves the overlap with the fewest sites.

Step 4 - Say which sites you'd rather keep

Several different remedies usually clear the same overlaps. Site prioritisation decides which one Sloc proposes.

• Treat all sites equally - the smallest effective remedy, no favourites.

• Set site importance manually - give sites, or whole categories of site, a value. The solver prefers plans that give up less total value, so two sites worth 1 each go before one worth 3.

• A site field reflects importance - use a number already in your data, such as revenue or floorspace.

Step 5 - Run it

Click Run divestment plan. The results show what had to be fixed and what fixes it.

Problem scope counts the markets that breach your tolerance and how many of each party's sites are caught up in them. Proposed remedy counts the sites the plan gives up, split the same way - so you can see at a glance whether the remedy falls on the acquirer or the target.

Clusters

Problem markets that share sites have to be solved together, so Sloc groups them into clusters. Each cluster shows how many problem markets it holds and how many sites the remedy touches. Open one and it names the sites to let go and the sites kept, and tells you whether there was one optimal answer or several equally good ones.

Clusters are the unit to argue about: one cluster is one local problem, and its remedy stands or falls on its own.

Looking inside a cluster

Clicking a cluster - on the map or in the list - does both at once: the map zooms to it, and its card opens beside it. That pairing is the quickest way to answer "where is this, and what does it cost?", because the geography and the site list are on screen together.

Click a market inside it and a card gives you that market in summary: whose site it is centred on, the catchment used, whether the site is Kept or given up, and - the part that matters - how its risk changes, shown as the rating before the remedy, an arrow, and the rating after.

View details opens the market in full: the sites in it on a map, the headline figures with their deltas, and the shares as a chart or a table. It has its own Pre-remedy / Post-remedy toggle and a selector at the top left for moving between the markets in that cluster.

Switch it to Post-remedy and you see what the plan does to this market specifically - here a 100% combined share becomes No overlap. That is the evidence that the remedy works, market by market, rather than in aggregate.

Pre-remedy and post-remedy

The toggle above the map switches between the problem and the fix. Post-remedy redraws the markets as they'd be once the plan is done - resolved markets in green, with the divested sites marked - which is the clearest way to show that the remedy works.

Save the plan

Click the save icon, name the plan and describe it. Saved plans sit against the project like scenarios do, so you can keep a strict plan and a lighter one side by side and compare what each costs.

Pro tips

• Run the strictest tolerance first. Knowing the worst case makes every softer plan easier to argue for.

• Name plans after the tolerance and approach - "No overlaps, targets only" - so the list reads as a ladder of options.

• A site you can't realistically sell belongs in site importance, not in your head. Give it a high value and let the solver work around it.

• Exporting the plan is covered in how to download divestment results.