Four workstreams. Most proposals describe three of them well and the fourth barely, and the fourth is usually the one that decides whether you get named at all.
Here is what each should contain, how to judge it, and what the first ninety days actually look like.
1. Measurement, per prompt
What arrives: the specific questions your buyers ask, run repeatedly across the answer engines, with two figures tracked separately. Whether your business was named, and whether your pages were cited as a source.
Why two figures: they have different causes and need close to opposite work. A business can be read constantly and named almost never. Across the 608 answers we measured in our own category, the brands supplying the reading and the brands getting the recommendation overlap far less than anyone expects. One blended number hides which situation you are in.
What good looks like: a list of questions, an engine against each, a date, and a yes or no in two columns. Boring, checkable, and reproducible by you in a logged-out chat.
What bad looks like: a traffic chart, or a single visibility score with no prompts behind it.
2. Citation and listing work
What arrives: your business claimed, completed and made consistent across the sources your category actually runs on.
We hold citation-source data across 27 industries, scored by the weight each source carries when an engine builds an answer. They are different everywhere. Insurer directories and licensing bodies in healthcare. Established legal directories in law. Review aggregates in home services. Regulator registers and comparison sites in finance.
Why it matters most early: the models re-read these often, so it shows up faster than anything else, and it is the evidence a model uses to decide a business is real rather than a claim.
What bad looks like: Google Business Profile and a handful of general directories. Those are what everyone claims, and in most categories they are not the ones carrying weight.
3. Content built to be quoted
What arrives: pages that answer one real buyer question each, with the answer in the first sentence rather than after three paragraphs of context.
How to judge it: read one and ask whether any single paragraph could be lifted out and dropped into an answer as it stands. If not, it was written to rank rather than to be quoted, and those are different shapes.
What bad looks like: volume. Twenty articles a month is throughput. Ask what the reporting measures instead.
4. Third-party presence
What arrives: you, mentioned in other people's content. Round-ups, comparisons, industry lists, community threads.
Why it is the one that matters: the sources deciding who gets named are mostly not yours.
“We've always grown through referrals - builders who know us pass our name on. That works, but it only reaches people who already know someone in the industry. Within a couple of weeks of the content going live, we had someone contact us directly through the website. That's a channel we didn't have before - and the enquiries have kept coming.
Why it is usually thin in proposals: it is slow, it has a long feedback loop, and it cannot be promised as a deliverable the way an article count can. An agency that omits it entirely is selling you the half of the problem that is easy to invoice.
How the four change by industry
The workstreams are the same everywhere. Their weighting is not, and a proposal that does not vary by category has not been written for you.
Local and home services. Review aggregates and map listings carry unusually high weight, and consistency of name, address and phone across them matters more than anywhere else. Content is the smaller half of the job here. Claiming is the larger.
Professional services and healthcare. Licensing registers, professional bodies and insurer directories are the high-weight sources, and they are slow to update but very durable once correct. Expect more of the budget in workstream two and a longer wait before anything moves.
Ecommerce. Community threads and video reviews do a lot of the work, because that is where people describe products in the words other buyers use. This is the category where workstream four pays back fastest.
B2B software. Comparison sites, integration directories and category pages dominate, and your own documentation matters more than in any other vertical because it is genuinely the best source on what your product does.
Ask any agency which of the four they would weight heaviest for you, and why. A specific answer means they have looked at your category. A balanced answer across all four usually means they have not.
What a monthly report should contain
Six things:
- The prompt list, with named and cited tracked separately, and the change since last month.
- Which competitors were named instead, and on which prompts.
- What was claimed or corrected this month.
- What was published, and which prompt each piece targets.
- What was placed or pitched externally.
- What is next, and why that rather than something else.
If the report is longer than that and still does not contain them, it is a document rather than a report.
Better than a monthly report is a login. A deck is a curated view assembled by the person whose work it describes. A portal is the same view they work from, and the difference matters most in the months where progress is uneven, which is most of them.
What you should have to do
Approve, and answer questions about your own business that only you can answer. That is it.
If you find yourself rewriting drafts, the arrangement is not saving you time and is slower than doing it yourself. Raise it in month two rather than month eight.
The first ninety days, realistically
Month one: measurement baseline, source audit, and the beginning of claiming. Mostly unglamorous. If month one is a strategy document, ask what happens in month two.
Month two: claiming finishes, first content lands, first outreach goes out. Usually the first movement in the citation figure.
Month three: content compounds, mentions begin. This is the earliest point at which judging the work is fair, and the named figure often has not moved yet.
Anyone promising a changed answer in month one is describing something else.
What the upper end looks like
Those are the honest averages. Front-loaded citation work can move much faster than that when a category is unworked.
PROCERT had grown entirely on referrals for years. Within a couple of weeks of the work going live, a stranger inquired directly through the website, and the inquiries kept coming. Fur Magic went from absent to the most-cited brand in their category in three weeks.
Both published with the starting number, the end number and the period: PROCERT and Fur Magic.
One month of deliverables, itemised
What a real month two looks like, from an engagement in a licensed trade.
Measurement. Twenty-two tracked questions, run weekly across five engines, logged out. Named in four, up from one. Cited in nine, up from three. Two competitors named on questions where the client is not, both identified by name.
Claiming. State license record corrected, business name now matching the EIN record exactly. Two review platform profiles claimed. One national directory listing corrected where the address was three years out of date.
Content. Three pages published, one per tracked question, each answering in the first two sentences under a heading matching how the question was asked.
External. One industry round-up pitched, one accepted for the following month, one community thread answered in full without a link.
Next month, and why. The remaining two registers, because the citation figure moved fastest off the first three, and one more external placement, because the named figure has not moved yet and that is the lever that moves it.
Everything in that list is checkable by the client in under twenty minutes. That is the test of whether a deliverable list is real.
See all four, on your own numbers
We will show you the measurement, the sources that carry weight in your category, and the ninety-day order, with your own baseline attached rather than a generic plan.
See which questions name you
A free AI visibility audit on your own buying questions, across all five answer engines.
Book your audit
Ashur Homa
Built and scaled a digital brand to $100M+ in sales with zero ad spend. Has helped businesses generate millions through AI go-to-market strategy. Leads growth at Omni Eclipse.
Connect on LinkedIn
