Win local and buying-intent prompts

Local and buying-intent prompts are where a citation is worth the most, because the person asking is about to choose. This is the sequence that wins them.

What you'll learn

• How to build a prompt set that sits next to a purchase
• The fixes that actually move local and buying prompts
• How to prove the win rather than assume it

Step 1: Build the set, not the list

Write six to ten prompts that a buyer would ask in the last hour before choosing. For local, the place goes in the prompt. For buying, the constraint goes in the prompt.

Add the comparison prompts even though they feel uncomfortable. They are where the buyer actually is.

Step 2: Read the grid, not the average

Ask "Which prompts am I losing, and who is taking them?" then open the grid. You are looking for the shape of the loss:

• Losing on one surface only. A retrieval or structure problem. Go to step 3.
• Losing on all surfaces. Either the prompt is too broad to name anyone, or you are genuinely not a candidate in the engines' view. Go to step 4.
• Winning but placed last. You are in the answer and not the recommendation. That is an authority problem.

Step 3: Fix the plumbing first

For local and buying prompts, three fixes do most of the work, in this order:

1. Access. Confirm the crawlers that fetch pages in order to cite them are allowed. This is a two-line file and it silently costs more than anything else on this list.
2. Extractability. The answer to the prompt should exist as plain text near the top of a page, not implied across a page, and not rendered only by JavaScript.
3. Place and trust signals. For local, machine-readable address and contact details, review markup, and links tying your entity to its other profiles.

Check all three in Readiness rather than guessing.

Step 4: Give the engine something to quote

For each losing prompt, the page that wins it usually has three properties:

• It answers the question in a sentence, early, without hedging.
• It says who it is for and who it is not for. Engines quote the qualifier, which is why "best for small teams" gets named and "best for everyone" does not.
• It is specific about place, price band, or constraint. Vagueness is unquotable.

Write that page, or fix the page that should have been it.

Step 5: Ship it through the ledger

Approve the change in the platform before you publish, so a control is captured. Then publish. Then let the platform fetch the page and re-measure. See ship and prove a change.

Without this step you will have an opinion about whether it worked. With it you have a Confirmed moved change and a number.

Step 6: Check the right outcome

Do not judge this work on total citation rate, because informational prompts will dilute it. Judge it on:

• the rate on your buying and comparison prompts specifically
• your placement in those answers, where the answer ranked things
• whether a rival's share on those prompts fell as yours rose

Success looks like

Your buying and comparison prompts cited on more than one surface, with at least one change in the ledger reading Confirmed moved against a control. That is a repeatable result rather than a good month.

Try this in Outercite

Ask the assistant "Where should I focus first, and what is it worth?" and check whether the top row touches a buying-intent prompt. If it does not, your prompt set is probably weighted towards explainers.

Related