SEO Trial Period: How to Test an Agency in 90 Days

1. Introduction: Ninety Days Beats a Twelve-Month Leap of Faith

Deciding whether to hire an SEO agency at all is one question. This guide answers the next one: how do you hire without betting a full year of budget on a supplier you have met twice?

Twelve-month contracts are the industry norm because SEO genuinely takes time. But a twelve-month lock also means that if the agency is wrong for you, you find out in month four and keep paying until month twelve.

A structured trial breaks that trade-off. It respects the fact that SEO takes months to show revenue while still giving you a fair way to judge the work early. Google says SEO needs four months to a year to show benefits, so a trial cannot be about results. It has to be about process and evidence.

The video below covers the hiring side of the same decision, which is useful context before you set the trial terms.

2. What Is an SEO Trial Period, Exactly?

Quick Answer: An SEO trial period is a short paid engagement, normally ninety days, with a written deliverable list, a measured starting baseline, agreed day-90 pass marks and a clean exit at the end. It is a full working quarter with a shorter commitment, not a discount package or a sales demo from your SEO agency.

The word "trial" causes most of the confusion. Owners hear it and picture a free sample. Agencies hear it and picture unpaid work. Both readings produce a bad quarter. A properly built trial has four parts:

• A paid, full-strength scope. The same work the agency would do in months one to three of a normal retainer, not a thinner version.
• A measured baseline. Traffic, rankings, indexation, technical errors and enquiries recorded in writing before anyone touches the site.
• Pass marks agreed in advance. Specific numbers or deliverables that decide continue-or-stop at day 90.
• A clean exit. No auto-renewal, no penalty, and all assets handed over if you walk away.

That is closer to a monthly retainer than to a one-off SEO project. The only real difference is the length of the commitment, and the fact that the review date sits in the contract instead of becoming an argument later.

Key takeaway: A trial is a normal quarter of SEO with the review date, pass marks and exit terms written down before work starts. Anything thinner is a demo, not a trial.

3. Why Ninety Days Is the Right Length

Quick Answer: Thirty days only proves an agency can run an audit. Sixty days ends just as the first fixes are being indexed. Ninety days covers a full cycle of audit, implementation, re-crawl and measurement, which is the shortest window that shows whether an agency can actually ship inside the normal SEO timeline.

Each month in a pilot does a different job:

• Month one is diagnosis. Access, baseline, technical audit, keyword map. Almost any agency looks competent here.
• Month two is implementation. Fixes deployed, pages rewritten. This is where weak suppliers start missing dates.
• Month three is evidence. Google re-crawls, indexation moves, early non-brand keywords appear. You finally see cause and effect.

Cutting the trial to sixty days removes month three, the only month that tells you anything Google has confirmed. That is why a short trial flatters a weak agency: it ends while everything is still a promise.

Ninety days also stops anyone hiding behind "SEO takes time" indefinitely, a line that shades quickly into the vague guarantees no honest agency makes.

Key takeaway: Ninety days is the shortest window containing a full audit-implement-recrawl cycle. Sixty days ends before Google has responded to anything.

4. What Actually Ships Inside a 90-Day SEO Pilot

Quick Answer: Across ZenWeb's Malaysian SME pilots, baseline setup, technical audit and keyword mapping ship in over nine cases in ten. Published content and Google Business Profile work ship in roughly eight. Link acquisition lands first placements in only about four in ten, which is why it should never be a pass mark. Expect the same sequence during SEO agency onboarding.

Workstream activity by month across ninety-day SEO pilots for Malaysian SMEs, with the share of pilots in which each workstream was completed by day ninety.

Workstream	Month 1	Month 2	Month 3	Done by day 90
Baseline and tracking setup	Recorded	—	Re-measured	97%
Technical audit and fix list	Delivered	Fixes shipped	Re-crawl check	94%
Keyword map and page plan	Drafted	Approved	In use	91%
Money-page rewrites	—	6–10 pages	4–8 pages	88%
Google Business Profile clean-up	Audited	Corrected	Posts and Q&A	82%
New content published	—	2–4 pieces	3–5 pieces	76%
Link acquisition placements	—	Outreach starts	First placements	41%

Based on ZenWeb's client sample of 500+ Malaysian SME accounts (2024–2026).

Read the last row carefully. Link building is the workstream most often sold hard in a pitch and least often delivered inside a quarter.

The rows above double as a shopping list. Any pilot scope missing baseline setup or the re-crawl check has quietly removed the two items that make the review measurable, so check it against what a full monthly SEO retainer should include.

Key takeaway: Audit, tracking, keyword mapping and on-page work reliably ship inside ninety days. Link placements usually do not, so never make them a pass mark.

5. Writing a Trial Scope Both Sides Can Be Judged On

Quick Answer: A trial scope needs counted deliverables, named owners and dates, so "eight pages rewritten by day 60" rather than "on-page optimisation". Add your own obligations too, because half of failed pilots stall on the client side. Apply the same discipline you would use when reviewing an SEO proposal.

Three tests separate a scope you can judge from one you cannot.

• Is it counted? Every deliverable carries a number and a deadline. "Improve site speed" fails; "LCP under 2.5s on the eight money pages by day 75" passes.
• Is it owned? Each line names who does it: agency, your developer, or you. Unowned lines become month-three arguments.
• Is it reversible? You should keep every deliverable if the trial ends, so content, redirects and tracking must live in your own accounts.

The third test catches the most expensive mistake. Work built inside an agency-owned account or tool licence disappears the day you leave, turning a ninety-day trial into a ninety-day loss.

Write your side into the same document. Approval windows, developer availability and access hand-over belong in the scope, because those client responsibilities in an SEO engagement decide how much of the plan can physically ship.

Key takeaway: Counted, owned and reversible. If a scope line fails any of those three tests, rewrite it before you sign, not at the day-90 review.

6. The Leading Indicators That Genuinely Move by Day 90

Quick Answer: Indexation, open technical errors and page speed move first, usually by day 60. Non-brand keywords and clicks move next. Enquiries move last and only slightly. Judge a trial on the early indicators and treat the late ones as direction, which is the same logic that separates real from vanity numbers in an SEO report.

Median values for six SEO leading indicators at baseline, day thirty, day sixty and day ninety across Malaysian SME pilot accounts.

Indicator	Baseline	Day 30	Day 60	Day 90
Target pages indexed	61%	68%	84%	93%
Open technical errors	47	39	18	7
Median LCP on money pages	4.1s	4.0s	2.9s	2.4s
Non-brand keywords in top 20	12	14	21	33
Non-brand clicks per month	240	255	310	420
Non-brand enquiries per month	3	3	5	7

Median values from ZenWeb client tracking across 12 industries, 2024–2026.

The shape of the table matters more than any single number. Everything technical moves between day 30 and day 60; everything commercial moves late and small. An agency showing you enquiry growth at day 30 is showing you seasonality, not SEO.

Set your pass marks on the top three rows and use the bottom three as direction of travel. That is also the honest way to build the reporting pack, as our guide to SEO reporting that proves ROI explains.

Key takeaway: Technical indicators move by day 60, keyword and click indicators by day 90, enquiries barely at all. Set pass marks on what a quarter can actually change.

7. What You Should Not Expect at Day 90

Quick Answer: Do not expect page-one positions on competitive head terms, a measurable revenue lift, a rebuilt backlink profile or recovery from a penalty. Ninety days buys foundations and early signals. An agency promising rankings by day 90 is repeating the guaranteed-ranking pitch in a shorter wrapper.

Four expectations cause almost every disappointed trial:

• Page one for your main money keyword. On a competitive Malaysian term, ninety days moves you into contention, not into the top three.
• A revenue number you can bank. Enquiries move by a handful per month, which is too small a sample to model annual return from.
• A rebuilt backlink profile. Outreach started in month two produces its first placements in month three, at best.
• Recovery from a manual action or core-update hit. Diagnosis fits in a quarter. Recovery normally does not.

Being honest here protects you twice. It stops you rejecting a good agency for missing an impossible target, and stops a weak one selling you that target in the first place.

If you want a return figure at the end of the pilot, treat it as a forecast built on the day-90 trajectory rather than a result. Our guide to calculating what SEO is really worth sets out the arithmetic.

Key takeaway: A trial proves capability, not commercial return. Anyone promising page one or a revenue lift by day 90 is selling, not forecasting.

8. What Happens to Malaysian SME Pilots After Day 90

Quick Answer: Pilots that started with a written baseline and an agreed scorecard continue to a full retainer roughly four times out of five. Pilots run on verbal expectations end more often than they continue, and undefined free trials almost always end badly, which mirrors the complaints behind businesses firing their SEO agency.

Share of ninety-day SEO pilots that continued to a full retainer, extended the trial, or ended, grouped by how the trial was set up at signature.

How the trial was set up	Continued to a retainer	Continued	Extended	Ended
Written baseline and agreed scorecard		78%	13%	9%
Written baseline, no scorecard		58%	17%	25%
Verbal expectations only		32%	12%	56%
Undefined free trial		19%	6%	75%

Aggregated from ZenWeb-managed campaigns, Malaysia, 2024–2026. Bar width shows share continuing to a retainer.

The gap between the first and third rows is the whole argument for paperwork. Same agencies, same budgets, same industries. The difference is whether anyone wrote down what success looked like before the work started.

Pilots that end are not all failures either. Some end correctly, because the paperwork made a poor fit visible in month three rather than in month nine, which is usually when a paid third-party SEO audit gets commissioned instead.

Key takeaway: The paperwork, not the agency, is the strongest predictor of how a pilot ends. Baseline plus scorecard roughly doubles the odds of continuing.

9. Exit Terms: Five Clauses That Make a Trial Safe

Quick Answer: Five clauses turn a trial from a risk into a test: no auto-renewal, full asset ownership, complete account hand-back, a documentation deliverable, and a fixed price for continuing. Ownership costs the most if you skip it, so settle who owns your SEO assets before signature.

• No auto-renewal. The trial ends on day 90 unless you actively continue. Silence must not roll you into twelve months.
• You own everything produced. Content, schema, redirects and tracking configuration, assigned to you on payment rather than on renewal.
• Accounts return within seven days. Search Console, Analytics, Tag Manager, Business Profile and CMS access reverted to your named owner.
• Documentation is a deliverable. The audit, keyword map and fix log handed over as files you keep, so your next supplier does not repeat month one at your cost.
• Continuation price fixed upfront. The monthly fee from month four written into the trial agreement, so a good result cannot be repriced.

That last clause is the one owners forget. A trial ending with a happy client and an unfixed price gives the agency every incentive to quote higher in month four.

If you are leaving an incumbent to run this pilot, sequence it carefully using our checklist on changing SEO company without losing rankings, and know in advance what happens when you stop SEO if you decide to pause instead.

Key takeaway: No auto-renewal, full ownership, account hand-back, documentation and a fixed continuation price. Five clauses on one page cap the downside of a bad hire.

10. Free Trial vs Paid Pilot: Why Free Costs More

Quick Answer: A free SEO trial is unpaid work, so it gets junior staff, leftover hours and no implementation. You lose a quarter of calendar time and learn nothing about delivery. The same logic explains why very cheap SEO retainers usually backfire: unpriced work is unprioritised work.

Free trials fail in a predictable pattern. The agency delivers an audit, because an audit is cheap to produce and looks impressive. Then nothing gets implemented, because implementation costs real hours nobody is paying for. At day 90 you hold a PDF and have lost three months.

Two things are genuinely reasonable to ask for free, and the difference matters:

• Fair to expect freeA discovery call, a short opportunity review of your site, and a written proposal with indicative deliverables.
• Not fair to expect freeA full technical audit, a keyword map, content, or any implementation on your live site.

Price the pilot at the agency's normal monthly rate. A discount is fine; free is a warning. If a quote sits far below market, check it against real SEO package tiers in Malaysia before assuming it is a bargain, because the gap is usually explained by what an overseas SEO agency leaves out.

Key takeaway: Free trials buy you an audit and cost you a quarter. Pay the normal monthly rate and shorten the commitment instead of shrinking the fee.

11. A 90-Day Pilot Against a Locked 12-Month Contract

Quick Answer: A pilot usually carries a slightly higher monthly fee but caps your exposure at one quarter instead of a year. On typical Malaysian SME numbers you commit roughly RM 10,500 to RM 18,000 rather than RM 33,600 to RM 60,000, a premium worth paying until the supplier is proven. Compare both against current SEO prices in Malaysia.

Read the premium as insurance. Paying ten to twenty per cent more for three months to avoid a possible RM 40,000 mistake is simple arithmetic for most SMEs, and once the supplier is proven, taking the annual discount in month four is the cheaper long-run position.

Businesses running a map-pack-first strategy should apply the same maths to a narrower scope, since local SEO pricing in Malaysia sits well below national campaign budgets.

Key takeaway: The pilot premium is insurance against a year-long mistake. Pay it while the supplier is unproven, then take the annual discount once they are.

12. How to Run the Day-90 Review

Quick Answer: Book the review before the trial starts, score deliverables and indicators separately, then decide continue, extend or exit on the day. A ninety-minute meeting with the baseline document open beats a month of hesitation, and a quick DIY check of your own site beforehand keeps the conversation grounded.

How to score an SEO trial at day 90

Run the review in one sitting, with the signed scope and the baseline numbers on the table.

1. Reopen the baseline document. Pull the numbers recorded before work started, so the comparison rests on evidence rather than memory.
2. Score deliverables first. Go line by line through the scope and mark each shipped, partial or missed. This measures the agency, not Google.
3. Score the leading indicators second. Compare indexation, technical errors, page speed and non-brand keywords against the day-90 pass marks you agreed.
4. Test the explanation of any miss. A specific reason with a date is fine. "SEO takes time" on a missed deliverable is not, because deliverables sit inside the agency's control.
5. Decide on the day. Continue at the fixed price, extend by one month with a written reason, or exit and trigger the hand-back clause. Do not defer.

Most disputes come from mixing steps two and three. Missed deliverables are an agency problem; slow indicators with everything shipped are a market problem, and that distinction is what lets you hold an SEO provider accountable fairly.

Key takeaway: Score deliverables and indicators separately, and decide on the day. Missed deliverables are the agency's fault; slow indicators usually are not.

13. Conclusion: A Trial Tests Process, Not Luck

Quick Answer: Pay a normal monthly rate for ninety days, record a baseline, agree pass marks on leading indicators, and write five exit clauses. That is the whole method for testing an SEO agency without committing a year of budget to a supplier you cannot yet judge.

An SEO trial period does not make SEO faster. It makes the decision cheaper. You still wait months for revenue, but you no longer wait a year to learn whether the people you hired can ship.

Two habits carry the pilot: set the review date at signature rather than at day 85, and take your own side of the deal seriously, because the approvals and access a client owes the agency decide how much of the scope survives the quarter.

If you are still shortlisting, what to expect in the first thirty days of onboarding and the trade-offs of hiring overseas instead of locally are the comparisons worth settling first. ZenWeb runs every new engagement against a written baseline and a day-90 scorecard.

14. Frequently Asked Questions

1. How long should an SEO trial period be?

Ninety days. That covers one full cycle of audit, implementation, re-crawl and measurement, which is the shortest window showing whether an agency can ship rather than only diagnose. Thirty and sixty-day trials end before Google has responded to any of the changes made.

2. Should an SEO trial be free?

No. Unpaid work gets junior staff and no implementation, so a free trial usually produces an audit document and nothing on your live site. Pay the agency's normal monthly rate and shorten the commitment instead. The shorter term is the concession you are asking for, not a lower fee.

3. What results should I expect after 90 days of SEO?

Expect cleaner indexation, most technical errors closed, faster money pages and more non-brand keywords entering the top twenty. Expect only a small movement in enquiries. Do not expect page-one positions on competitive terms or a revenue figure you can bank.

4. Can I keep the work if I end the trial at day 90?

Only if the agreement says so. Content, redirects, schema, tracking setup and documentation should be assigned to you on payment, and all accounts must sit in your name from day one. Without those clauses, work built inside agency-owned accounts leaves with the agency.