SEO proposal for clients: build it with the prospect's real rankings before you sign

What a data-driven SEO proposal is (and why it beats the template)

A data-driven SEO proposal is a diagnosis measured against the prospect's own domain before any contract exists: their current rankings by city and device, the top 10 for every keyword they care about, and, if they grant access, their real Search Console queries. None of that is estimated or guessed at: it is measured, timestamped, and presented as it is. It is the difference between telling a business owner "we can improve your SEO" and showing them, in a table, that they currently sit at position 14 in Bogota on mobile and position 9 on desktop, for the exact search they believe already brings them customers.

The template does the opposite. Search for "SEO proposal" or "how to write an SEO proposal" and most of what ranks today is generic and interchangeable: 80-point audits that could apply to any website because they do not really measure any single one in particular, keyword gap lists copied over from an unrelated industry, and pricing that means nothing once converted into a client's actual currency. None of those pages use the prospect's real rankings by city and device. That gap is exactly what a proposal built with a rank tracker fills: it does not replace the consultant's strategy or judgment, but it gives the proposal something no template ever will, which is the prospect's own domain measured today, not a generic stand-in.

The thesis is simple and worth repeating on the first call with any prospect: they sign when they see their own data, not when they read a list of services. A prospect looking at their real position in their real city reacts differently than one reading "we will do a technical audit, keyword research and link building." The first is concrete and verifiable; the second is the same paragraph the other three providers they are comparing already sent them.

What to ask the prospect before you measure: domain, cities, keywords and Search Console access

Before you ever open the app you need a roughly 30-minute discovery call, and the script is shorter than it sounds. Four questions cover it: which cities actually generate revenue (not where they wish they sold, where they already do), which products or services carry enough margin to justify an SEO investment, 5 to 10 searches the prospect believes bring them customers, and from where and on which device they last searched for themselves. That last question sounds trivial and ends up being the one you use the most later, because it almost always reveals the exact bias you are about to correct with data.

On that same call, ask for read-only access to their Search Console. It is a permission, not a password: the prospect adds your email as a user on the property from their own Google account, in a couple of minutes, without sharing anything sensitive. The exact permission type to request and the steps to get there deserve their own full guide, so here the recommendation is simply to always ask, even when you expect to be turned down.

If the prospect does not grant that access — it happens more often than anyone would like, especially with business owners who do not even know Search Console exists — the diagnosis proceeds exactly the same with what you can measure without anyone's permission: public rankings and the site scan. The only thing you lose is the view of real queries and unbranded impressions, which is valuable but not essential for a solid proposal. Do not let missing access stall the work; turn it into the first line of your proposal instead, where you explain what you would show them if they granted it.

How to run an SEO proposal diagnosis in one week, step by step

With the information from the discovery call in hand, the full diagnosis fits inside one week of real work, without that meaning a week of billable hours: it is minutes on day 0 and a daily glance for the rest of the week. Here is the exact sequence, from day 0 to day 7.

1. Day 0 · Discovery call (30 min)

   Domain, 1 to 3 cities that actually generate revenue, 5 to 10 keywords the prospect believes bring them customers, their customers' device, and, if possible, read-only Search Console access.

2. Day 0 · Project and site scan

   You create a project with the prospect's domain and run the site scan, which relies on Google's public results the same way rankings do: it needs neither ownership nor site verification. In seconds you get up to 50 keywords they already rank for, with volume and difficulty — a first look at what already works, before you have even added the keywords from the call. This is exactly the features you use in the diagnosis: site scan, Search Console and rankings by city.

3. Day 0 · Add the keywords by real city

   The 5 to 10 keywords from the call, each tracked in the city where the prospect actually generates revenue. Quota example: 8 keywords across 2 cities is 16 units on Pro, or 32 if you track them on “Both”; Free gives you 6 units on a single device and a single project — enough for 6 keywords in one city, not much more.

4. Day 0 · First measurement, almost instant

   With 1 or 2 new keywords, the measurement runs in seconds; add the full batch and it is ready in a few minutes. By day 0 you already have a position by city, the prospect's URL that ranks (if it appears in the top 30), and the current top 10 for every keyword. That is your day-zero snapshot, the one you compare the rest of the week against.

5. Day 0-1 · Search Console, if you have access

   Unbranded queries with impressions, the pages receiving them, and average position — used only as context, never presented as "current position": Search Console's average position and the real position a tracker measures are not the same number.

6. Days 1 to 6 · Daily monitoring

   With the 14-day Pro trial you measure every day and watch how much each keyword oscillates day to day — a position that is stable on day 0 and jumps five spots by day 3 tells you something different than one that never moves at all. You also run the AI-powered keyword analysis to spot intent mismatches between what people search and what your title and URL suggest against that keyword's top 10. Your working hypotheses come out of that.

7. Day 7 · Close the diagnosis

   Build the baseline table, prioritize 3 working hypotheses, write down clearly what you will not promise, and export the CSV and the PDF (with your name and color, no logo) to attach to the proposal.

One nuance worth stating plainly: the "one week" duration is not a hard technical minimum, it is the time you need to confirm that the day-0 position was not a fluke. If the prospect is in a hurry and you need to present sooner, the day-0 snapshot already works for the diagnosis — position, top 10 and the ranking URL are all complete from the first minute — what you sacrifice is certainty that the position is stable, and that is something worth telling the prospect directly in the proposal, not hiding.

The test that disarms the client: "I'm already first" against their real city and their phone

Every consultant who has ever sold SEO knows the line: "but when I search, I'm already first." It is almost always true and almost always misleading at the same time. The prospect searched from their own office, logged into their own Google account, and probably already clicked their own site before. Google personalizes on those signals, so their "position 1" describes what they themselves see, not what a new customer sees searching from home, elsewhere in the city, with zero history with that brand. Switching to incognito mode does not fully fix it either: incognito mode does not erase location or device, which are still the consultant's own.

The way to disarm that objection is not to explain it, it is to show it. You measure the keyword in the prospect's real city — not your office's city — and, on Pro, on mobile and desktop at once, since Free measures each keyword on a single device. The result gets presented in a three-column table the prospect can read in ten seconds: what they believe, against what mobile measures in their city, against what desktop measures in their city.

There is no need to dramatize the gap or oversell it in front of the client: position 9 or 11 is not a disaster either, it is simply different from what the prospect believed, and that distance between belief and data is exactly what opens the conversation about why they need you and not someone who will just repeat what they already think. Browser personalization, and why incognito mode does not fully solve it, is a topic that deserves its own space, so it stays at that mention here.

Today's top 10 for each keyword: who currently holds the spot you're about to fight for

Every keyword in the diagnosis comes with its current top 10, available on both Free and Pro, and that list is the second most convincing data point for a prospect because it puts a name and a face on the competitors currently taking that space away from them. Sorting those ten results into four buckets — direct competitor, marketplace, media outlet or directory — tells you far more about a keyword's real difficulty than any generic "keyword difficulty" score pulled from a tool.

If the top 10 for "dental implants Medellín" is full of other local clinics, the conversation is about differentiation and original content. If instead two health marketplaces and a medical directory occupy half the top 10, the conversation changes: you are not only competing against other clinics, you are competing against platforms with budget and a dedicated team behind their SEO, and that is worth telling the prospect with that same bluntness, because it sets a realistic time expectation before anyone signs anything.

This top 10 of your own tracked keywords is not the same thing as researching a competitor's entire keyword universe: figuring out every other search a rival domain ranks for needs a dedicated keyword gap tool like Ahrefs, Semrush or Dinorank, which Rankiamos does not replace. How to read deeper patterns in the top 10, and what to do once you discover a competitor is beating you across several searches at once, is covered in the guide on reading who is outranking you on each keyword.

One clarification worth stating directly in the proposal itself: competitor history — how those competitors moved over time, who entered and who left — is a Pro feature that starts accruing data the day the prospect signs and the project becomes real. During the diagnosis week you see today's snapshot, not months of footage; that part waits until after the signature.

SEO goals you can actually sign (and what never gets promised)

Google says it plainly in its own guidance for people looking for SEO help, updated June 10, 2026: "No one can guarantee a #1 ranking on Google." It is worth quoting that line directly inside your own proposal — it is the best defense you will have the day a client asks for rank 1 within two months, and it also shows them you know the field well enough not to sell smoke.

Neither rank 1 as a guarantee nor fixed calendar dates. What you can sign are SEO goals framed as ranges with measurable indicators, not promises of a specific outcome. A few reasonable examples: a given number of the diagnosed keywords reaching the top 20 (or the top 10, depending on where each one started) within an agreed working period; a sustained rise in unbranded query impressions in Search Console, not just a good month; new query categories showing up that the prospect's site does not currently receive at all. Each of those indicators is verifiable with the same data you already gathered during the diagnosis, which closes the loop: you measured before signing, you measure the same way afterward.

The baseline table you built during the diagnosis is not just a nice-looking annex, it is the real starting point for the KPI you will be reporting on month after month. The exact conditions worth locking down in the contract — which city, which device, which day of the month, and which tool verifies progress — deserve their own guide, not yet published; once it exists, it will hang directly off this baseline.

The SEO proposal for clients, in six sections that fit five pages

With the diagnosis done, the document practically writes itself. Here are the six sections, in the order a prospect needs to read them:

Here is what section 2 looks like with real data, using the same dental clinic example from earlier:

The PDF you export on Pro attaches directly onto this table: one page with your name, your accent color and a free-text footer for your contact details, no custom logo. If your brand needs a logo on every page, that layout gets built separately, using the CSV as your data source. What the document you hand the client after they sign looks like — the recurring report, not the pre-sale diagnosis — is covered in the guide on the monthly report structure you deliver after signing.

If they sign, or if they don't: what to do with the diagnosis project

If the prospect signs, you do not have to start measuring from scratch: those keywords already carry history from the day of the diagnosis, because tracking is never retroactive on any plan — which is exactly why you add them before signing, not after. From there, competitor history starts accruing from the signature date, along with whatever additional quota you need for the extra keywords coming out of the 90-day plan.

If they don't sign, closing out is just as simple: export the CSV for your records, delete the project, and free up the quota. The Pro plan gives you 5 projects total, so it is worth treating them as a scarce resource and always keeping one open for the next diagnosis, rather than having all of them tied up by active clients and no room left to run the next prospect.

How to split that quota between signed clients and prospects still in diagnosis, once you are juggling several accounts at once — how many units each one gets, when it is worth sacrificing a city — is worked out in full arithmetic in how to split quota between clients and prospects.

The mistakes that lose a proposal even when the diagnosis is good

The diagnosis can be measured perfectly and the proposal can still fail because of how it gets presented. These are the mistakes I see repeated most, roughly in order of frequency.

Sending 50 keywords instead of 10. It feels like more delivered work and it actually dilutes the message: nobody is reading a 50-row table in a first meeting, and the ones that actually mattered get lost among the ones that don't. Presenting Search Console's average position as if it were "the current position." Those are different numbers answering different questions — one is an average across every impression in the period, the other is a position at a specific moment — and confusing them is one of the most common technical mistakes in otherwise well-built proposals; the full arithmetic behind why they diverge is in Search Console's average position and why it isn't the real one.

Measuring from your own city instead of the prospect's. It completely defeats the point of the test from an earlier section: measure from your own office and you end up confirming the exact illusion you were about to prove false. Handing over the PDF with zero interpretation next to it. A document full of numbers with no narrative connecting them generates more questions than it answers, and the prospect will ask you those questions in the meeting, not before. Promising fixed timelines and hiding which tool did the measuring. The second one in particular costs credibility the moment anyone notices: openly stating you measured with Rankiamos, screenshots from the app if needed, comes across as more serious than a table with no visible source.

None of these mistakes are SEO strategy mistakes, they are presentation mistakes, and that is exactly why they are the easiest to fix once you know they exist. The diagnosis already gives you the material; what is left is not wrecking it on delivery. If you want the full picture of everything Rankiamos measures for a small agency or a freelancer, that is the complete catalog behind every data point used in this guide.