Answered straight

How to compare three SEO proposals that look nothing alike

The short answer

Convert every proposal into hours before you compare anything. Phrases like "unlimited keywords" and "20 backlinks a month" aren't units — turn them into working hours, divide by the fee to get a blended rate, then score all three against eight weighted criteria. Any proposal with no baseline is disqualified.

Updated 26 July 2026 · Written by the Last Agency team · See what SEO actually costs

The short version

  • Nobody quotes in the same units, so the first job is arithmetic, not judgement. Hours per rupee is the only common denominator.
  • "Unlimited keywords" is worth zero points. Keyword tracking is free; the work isn't.
  • Thirty backlinks a month at ₹25,000 is arithmetically impossible with earned links. Something else is being sold to you.
  • One omission disqualifies a proposal regardless of price: no baseline number, frozen before work starts.

Three quotes, three sets of units

You asked three agencies for a proposal. One came back with a keyword count. One came back with a deliverables list. One came back with a team structure and a number of hours. The fees are ₹25,000, ₹95,000 and ₹75,000, and there is no honest way to say which is cheapest, because they aren't quoting the same thing.

The examples below are composites — the three shapes we see most often in the Indian market, not real quotes from real agencies. Use them as a template for your own three.

The method has three steps: convert deliverables to hours, normalise all three onto one sheet, then score. It takes about ninety minutes and it's the highest-return ninety minutes in the whole buying process. Do it before the follow-up calls, not after — once you've met the people, you'll be scoring how much you liked them, and likeability is uncorrelated with whether your organic leads move.

  • Proposal A — ₹25,000/mo. Unlimited keywords, 30 backlinks a month, 8 blogs, monthly technical audit, dedicated account manager.
  • Proposal B — ₹95,000/mo. Four long-form articles, technical SEO retained, links via "digital PR", quarterly strategy review, 12-month term.
  • Proposal C — ₹75,000/mo. 45–60 hours itemised by discipline, three named people, sample of the last 20 links placed, month-to-month after the first quarter, one measured number written into the contract.

Step one: turn every phrase into hours

Proposals are written in marketing units. Delivery happens in hours. These are reasonable conversions for competent work in India — argue with the numbers, but make the agency argue with them too.

Converting proposal language into working hours.
What the proposal saysWhat it usually meansHours to assume
"Unlimited keywords"Rank tracking is priced per keyword and costs almost nothing. The optimisation work is identical regardless of how many you track.0 — it's not a deliverable
"30 backlinks a month"A genuinely earned placement takes 3–8 hours of research, pitching and follow-up, at a hit rate of maybe 5–15% on cold outreach.90–240 hrs if earned; ~0 if bought
"8 blogs a month, 1,000 words"Brief and research 1–2 hrs, draft 2–3, edit 1, internal linking and publish 1.40–56 hrs
"Monthly technical audit"Re-running a crawl is quick. Fixing what it finds is the work, and it's usually excluded.1–2 hrs audit; fixes separate
"On-page optimisation, 10 pages"Titles, meta and headings is one thing. Rewriting a page to match intent is another.5–15 hrs, or 30–50 if it's real
"Dedicated account manager"Often a coordinator who forwards reports, not the strategist who made the plan.2–4 hrs — ask who owns the plan

Step two: put all three on one sheet

Now every proposal is expressed the same way. This is usually the moment the cheap one stops looking cheap and the expensive one stops looking expensive.

Build it as a spreadsheet with the quotes as columns and these seven rows. It takes twenty minutes and it does something no amount of reading can: it forces each agency's claims into a form where they can contradict each other. Proposal A's blended rate is the giveaway — nobody in India staffs senior outreach at ₹180 an hour, so either the deliverables mean something other than what they say, or somebody didn't do the arithmetic before sending the document.

The same three proposals, normalised.
LineA — ₹25,000B — ₹95,000C — ₹75,000
Implied working hours140–300 (as written)55–7045–60 (stated)
Blended rate per hour₹80–₹180 — impossible₹1,350–₹1,700₹1,250–₹1,650
Content8 posts, writer unnamed4 articles, editor named4 articles, brief + edit itemised
Links30/mo, source unstated"Digital PR", no samplesSample of last 20 placements
TechnicalAudit only, fixes excludedRetained, dev not includedRetained, dev sprint scoped
BaselineNoneNoneTrailing-90-day organic leads
Contract6-month lock-in12-month lock-inMonthly after quarter one

The eight criteria, and what each weight is doing

Weights matter more than scores. Most buyers weight price at 40% and measurement at nothing, which is exactly backwards — price is the one variable you can renegotiate later, and measurement is the one you can't retrofit. Score each criterion 1 to 5, multiply by the weight, add it up.

Eight weighted criteria for scoring an SEO proposal.
CriterionWeightA 5 looks likeA 1 looks like
The measured number and its baseline20One business metric, frozen on a stated date, written into the contract"We'll improve your rankings"
Named people and the seniority split15Three names, roles, and how many hours each contributes"A dedicated team of experts"
Where the links come from15A sample of recent live placements you can open in a browser"Our private network of publishers"
Hours per rupee12Hours itemised by discipline; blended rate you can checkNo hours anywhere in the document
Technical scope and who ships it12Fixes scoped, dev ownership agreed, exception path for urgent issuesAudit delivered, implementation "client side"
Content ownership chain10Who researches, writes, edits and approves, with turnaround times"SEO-optimised blogs" and a word count
Contract terms10Short notice, no long lock-in, you keep every asset on exit12 months minimum, 90 days' notice, assets on their accounts
Reporting and human contact6A monthly call with the strategist and a report that leads with leadsAn automated dashboard that leads with traffic

Scoring the three

Run the composites through it and the ranking inverts against price in one case and confirms it in another.

Proposal A scores in the low twenties out of 100. Not because ₹25,000 is too little money — a small local site can genuinely be served at that number — but because every line is unfalsifiable. Unlimited keywords, unnamed writers, unsourced links, no baseline. There is nothing in the document you could later prove they didn't do.

Proposal B lands somewhere in the high fifties. The rate is real, the work is probably real, and the people might be excellent. It loses on two things: no baseline, so nobody can say at month nine whether it worked, and a 12-month lock-in that removes your only real remedy. Fixable — ask them to add a measured number and cut the term to six months. If they refuse both, the score is the score.

Proposal C scores in the eighties. It's also shaped like ours, which you should hold against it while reading this page — we designed our proposal around this scoring sheet because we wrote the scoring sheet. Grade us on whether the criteria are the right ones, not on which composite wins.

The omission that should disqualify a proposal outright

No baseline. Not "a weak baseline" — the complete absence of a stated number, frozen on a stated date, that both parties agree the work will be judged against.

Without it, month nine is a debate. The agency shows a traffic chart going up and to the right. You say leads feel the same. Neither of you can win, because nobody wrote down where you started. Every other flaw in a proposal is negotiable; this one makes the whole engagement unfalsifiable by design.

The fix is one paragraph. Name the metric — qualified leads from organic search is the usual right answer, organic revenue if you're transactional. Name where it's measured, which should be your CRM, not analytics. Freeze the trailing 90 days before kickoff. Name the date you'll both look at it again.

That paragraph is why we structure our own SEO engagements around a frozen trailing-90-day organic lead count rather than a ranking promise. We don't guarantee a position — nobody controls Google's index — but we do guarantee movement against your own number, and if we miss it in 90 days we keep working free until we beat it. Any agency can write that clause. Most choose not to, and that choice is information.

What to get in writing before you sign

Seven lines. If an agency can't supply all seven within two working days, you've learned something more useful than any proposal could tell you.

  1. The baseline metric, its current value, the date it was frozen and where it's measured.
  2. Named people, their roles, and the approximate hours each contributes per month.
  3. Hours split across technical, content, links and reporting — see what an SEO retainer includes for the shape this should take.
  4. The last twenty links they placed for any client, as live URLs.
  5. Who ships technical fixes, and what happens when something urgent breaks.
  6. Notice period, minimum term, and confirmation that you own every asset and account on exit — the rest belongs in your SEO contract.
  7. What happens if the number doesn't move. Their answer to this is the single most predictive line in the document.

Related questions.

How do I compare SEO quotes with completely different deliverables?

Convert everything to hours, then divide the fee by the hours to get a blended rate. Four articles, ten technical fixes and twenty links all become numbers of working hours. Once every quote is expressed as hours per rupee, the comparison is arithmetic instead of vibes.

Is the cheapest SEO proposal always the worst?

No. A ₹25,000 quote for a small local site with a narrow keyword set can be perfectly honest. It becomes a problem when the deliverables list is impossible at that price — 30 links a month, unlimited keywords, eight articles. Check whether the arithmetic works before judging the number.

What does "unlimited keywords" actually mean in an SEO proposal?

It means the agency will track as many keywords as you like, which costs them almost nothing. The optimisation work is the same whether you track 20 or 2,000. Score it zero and ask instead how many queries the plan actively works on this quarter.

Should I pick the agency with the most detailed proposal?

Detail is a good proxy for whether they've planned the work, but length isn't. A three-page proposal with named people, itemised hours, real link samples and a frozen baseline beats a forty-page deck of methodology diagrams every time. Grade specificity, not page count.

How much should I weight price when comparing SEO proposals?

Around 12% in this sheet, expressed as hours per rupee rather than the headline fee. Price is the variable you can renegotiate later. Measurement, seniority and link sourcing are the ones you're stuck with, so they carry the weight.

Can I show one agency's proposal to another?

Show the scope sheet, not the document. Sharing a competitor's pricing is bad form and usually breaches their confidentiality line. Sharing your normalised requirements — hours, disciplines, baseline, contract terms — is fair, and it gets you three quotes you can actually compare.

Last slot's open

Make this the last growth call you book.

Grab the free strategy call and walk away with a 90-day growth plan — hired or not. Or just text us. Either way, you'll know exactly how we'd win.

Guaranteed or it's free · No lock-in · Free strategy call