What you are buying, in plain terms.

Technical fixes, content built for queries that have been screened for real commercial intent, and the earned coverage that makes the rankings hold. Produces: position band movement you can see keyword by keyword, and the pages that caused it.

Account structure, keyword and audience selection, landing pages, and the ongoing job of cutting what does not convert. Produces: cost per inquiry by campaign, and a shrinking list of things you are paying for that never worked.

The pages and articles that answer what your buyers ask before they are ready to buy, written by people who have read your product rather than a brief about it. Produces: assets that keep earning after they are published, which is the only kind worth commissioning.

Tracking that is actually correct, then the unglamorous work of improving what the traffic does when it arrives. Produces: a conversion rate that moves, and reporting you can hand to a finance director without a translator.

We build and maintain sites as part of the marketing rather than referring it out, which means a fix identified on Monday can ship on Monday. Produces: no change request queue between finding a problem and solving it, which is worth more than it sounds.

You work with the principals. There is no account manager relaying questions to a delivery team, which is the arrangement that makes most agency relationships feel slow for reasons nobody can quite name.

Every line should have an output and a rough timescale attached. "Ongoing SEO" is not a deliverable. "Technical fixes and eight pages built against screened keywords, with band movement expected from month three" is one, and it can be checked.

Ask for the name and ask whether that person is in the meetings. Agencies that sell with principals and deliver with juniors do not lie about this if asked directly, because the honest answer is on the org chart.

The single most useful question on this list, and the one that most often produces a pause. Ad spend, in nearly every case, and often photography, video, translation and development beyond some threshold. If the answer is "nothing, it is all included", the scope has not really been written yet.

A service worth buying has a version of "this channel is not going to work for you". If nobody can name one, you are buying activity rather than judgment, and you will find out which in about nine months.

Ask to see a real report, redacted. The tell is whether it contains anything that did not work. Reporting that only shows movement in the right direction is not reporting, it is retention.

Content, links, tracking configuration, account access and the site itself should be yours and should stay yours. Anyone whose results are hosted on their own domains, or whose ad accounts are not in your name, has built a reason for you not to leave that has nothing to do with performance.

Two numbers people conflate. Media spend goes straight to Google or Meta and is yours. The cost of the work is separate — a salary, a freelancer, or a retainer. Ours is a flat monthly fee with no long-term contract, and the mix inside it moves with the data rather than with what was scoped in month one. We would rather quote against your numbers than publish a table that fits nobody.

No, and you probably should not. The most common sensible starting point is one channel run properly plus the tracking to prove it, and the second most common is fixing the site before anything is pointed at it. Buying six channels at once is the standard way a first-year budget disappears without producing evidence about any of them.

No long-term one. The average client stays about seven years, and the reason we can say that is that none of them have to. Notice periods are the industry's substitute for retention and they are a bad one — an agency that needs twelve months of lock-in is telling you what it expects month nine to look like.

Paid can produce inquiries in the first week. Conversion work on the site can move within a few weeks because it applies to traffic you already have. Organic typically shows band movement within a quarter and clicks in the second or third. A proposal promising organic results in thirty days is describing paid traffic or a penalty being lifted.

You do — content, links, tracking configuration, ad accounts, the site. Accounts are set up in your name from the start rather than migrated later, which is a small administrative difference at the beginning and an enormous one if you ever leave.

Most of our better arrangements are exactly that rather than a handover. The split that works is usually product knowledge in-house and the long-feedback-loop work outside, which is the argument in the 2010 passage above with the threshold moved.