Paid media should show directional signal within 21 to 30 days, SEO within 90 to 120 days, and content-led organic demand within six to nine months. What should be visible at 30 days regardless of channel is process quality: research depth, creative volume, testing cadence, and reporting that tells you something you did not already know. Judge process early and outcomes late, because the reverse gets good agencies fired and bad ones retained.

That inversion is the expensive mistake. A business gives a performance agency six months because "marketing takes time" — when a paid account with no signal at week five is already failing. Another business fires an SEO agency at month three for no rankings, when month three is roughly when the foundational work stops being invisible.

Both decisions were made on the wrong clock. If you've already worked through what a ₹1 lakh retainer should deliver, this is the companion question: how long before you can tell whether you're getting it.

Why do channels run on different clocks?

Because they buy fundamentally different things.

Paid media buys attention immediately. The auction is live the day you fund it. The variable is not whether people see the ad but whether the offer, creative, and landing experience convert — and that answer arrives in weeks, because you can buy enough impressions to reach statistical confidence quickly.

SEO buys position in a queue. Google has to crawl the change, index it, test it against user behaviour, and adjust. That process runs on Google's schedule, not yours, and it involves competitors who are also moving. Google's own guidance is explicit that changes take time to show effect and that no provider can guarantee position or timeline.

Content and brand buy familiarity. The mechanism is accumulated exposure across many touches. There is no version of this that completes in a quarter, which is why the long-term effectiveness research published through the IPA consistently separates short-term activation from brand building — they draw from the same budget and answer to entirely different timeframes.

Truth line: Judging every channel on a 90-day clock guarantees you will over-invest in the fast ones and abandon the ones that compound.

What should you actually see at 30, 60 and 90 days?

By day 30 — process, not outcomes

This is the checkpoint that matters most and the one almost nobody defines in advance. Regardless of channel, you should see:

• Research that demonstrates understanding of your buyer, not your industry in general
• Conversion tracking configured and verified, not assumed
• A stated hypothesis about what will work and why
• Creative or content in production, with volume visible
• A first report that names something you did not know

If paid media is running, you should also see early directional signal — cost per click stabilising, click-through rate benchmarking, and the first read on lead quality. Not profitability. Direction.

By day 60 — early outcomes on fast channels

Paid media should be producing leads at a knowable cost, with at least two rounds of creative testing behind it. SEO should show technical fixes shipped, content published, and early impression growth in Search Console even if clicks haven't moved. Impressions rising before clicks is the correct order, not a failure.

By day 90 — pattern, not proof

Paid should be approaching or hitting target cost per qualified lead. SEO should show ranking movement on lower-competition terms and measurable impression growth. You should be able to name which channel is working, which isn't, and what the agency is doing about the second one.

Ninety days is where you can judge trajectory. It is rarely where you can judge return.

The Judgement Window Framework

We use this when setting expectations at the start of digital marketing engagements, because agreeing the clock upfront prevents most mid-engagement conflict.

Every channel has three windows. Write them into the engagement before work starts.

Window 1 — Process (day 30, all channels)
Judged on: research quality, tracking accuracy, creative volume, testing cadence, reporting substance. A failure here is a genuine failure and does not require waiting.

Window 2 — Signal (channel-dependent)
Paid media: day 30. Email and CRO: day 45. SEO: day 90. Social: day 60. Content-led organic: day 120. Judged on: is the metric moving in the right direction, regardless of magnitude.

Window 3 — Outcome (channel-dependent)
Paid media: day 90. Social: day 120. SEO: day 180. Content and brand: day 270. Judged on: cost per qualified lead, pipeline contribution, revenue.

The framework's real value is what it prevents. It stops you cancelling an SEO engagement at day 60 for missing an outcome it was never going to produce, and it stops you tolerating a paid account that showed no process quality at day 30 because "these things take time."

What are the warning signs before the clock runs out?

Poor results alone are not sufficient evidence. Poor process is. Six signals that mean you don't need to wait for the outcome window:

Nobody has asked about your margins. An agency optimising to lead volume without knowing what a customer is worth is guessing at the target.

Creative output has stalled. Same three assets running in week eight means no testing is happening, which means no learning is happening.

Reporting describes rather than diagnoses. Reach, impressions, and engagement with no statement of what was tested, what failed, and what changes next. Covered in more depth in how to actually measure digital marketing ROI.

The same optimisation explanation repeats. "We're still optimising" for three consecutive months is not a plan, it's a holding pattern.

Tracking was never verified. If conversion tracking is broken at day 30, everything reported since is noise. Most accounts we inherit have it misconfigured.

Nobody has disagreed with you. An agency that has never pushed back on a request is executing, not advising. That's a structural problem and it will not improve with time. It's the same distinction explored in growth partner versus vendor.

Four or more of these at day 45 means the engagement is already failing, and waiting for day 90 costs you a quarter you won't recover.

What slows results down that isn't the agency's fault?

Worth stating plainly, because roughly half the engagements that fail do so for reasons on the client side.

Slow approvals. Creative sitting in a WhatsApp thread for nine days is nine days of media spend running on old assets. If your approval turnaround averages a week, your agency is operating at roughly half the cadence you're paying for.

Restricted asset access. No product photography, no customer access, no case study permission, no willingness to appear on camera. Creative quality is capped by raw material.

An offer problem misdiagnosed as a marketing problem. If the product, pricing, or positioning is the constraint, no channel execution fixes it. The diagnostic sequence is in what to do when marketing isn't working.

Sales follow-up gaps. Leads arriving and sitting unworked for two days will convert at a fraction of their potential. The agency's number looks bad; the failure is downstream.

Budget below the channel's floor. ₹40,000 of monthly media spend cannot produce statistically meaningful testing data in 30 days. The clock stretches because the data is thin, not because the work is slow. Related arithmetic in why cheap retainers cost more.

How should this be written into the engagement?

Three additions to your scope document, all of which protect both sides.

Agree the windows in writing before work starts. Which channels, which signal date, which outcome date, judged on which metric. Ambiguity here always resolves into a disagreement later.

Define what a failed checkpoint triggers. Not automatic termination — a structured conversation with a named remediation plan and a follow-up date. Most engagements that die could have been fixed at day 45 if anyone had raised it.

Specify client obligations against the same clock. Approval turnaround, asset provision, access to sales data. An engagement that only holds the agency to a timeline will produce arguments about attribution the moment something slips. The mechanics are covered in how to write an SOW that holds.

What if the agency asks for more time?

Sometimes reasonable, sometimes a stalling tactic. Three questions separate them.

"What specifically will be different in the next 60 days?" A credible answer names a change in approach — a new creative direction, a channel reallocation, a technical fix shipping. A weak answer describes more of the same with more optimism.

"What have you learned that you did not know at the start?" Agencies doing real work accumulate specific knowledge about your buyer, your funnel, and your competitors. Agencies going through motions have general observations.

"If this were your business, would you keep spending?" The answer to this is usually visible in how quickly it comes.

An agency that answers all three well has probably earned the extension. One that deflects on the second has not.

How Midgrow structures this

We build complete growth systems rather than selling channels as line items, which means we set the clock explicitly and early — because the alternative is being judged against an expectation nobody wrote down.

• Windows agreed in writing before work begins, channel by channel, with the metric named.
• Conversion tracking audited in week one. Not week six. If measurement is wrong, every subsequent report is fiction.
• A day-30 process review covering research, tracking, creative volume, and what we've learned — not outcomes, because outcomes at day 30 are noise on most channels.
• Reporting that names what failed. A report with no failures in it is a report that isn't testing anything.
• Client obligations stated in the same document — approval windows, asset access, sales data — because the clock runs on both sides.

That spans performance marketing, social media systems, SEO, and the AEO and GEO layer determining whether AI assistants recommend you at all.

The proof is public rather than promised. We generated 10,890 leads at 11.3x ROI for a solar EPC client and delivered 585 percent organic growth with first-position rankings for Autosys Solar — the second of those in energy, a sector where the outcome window genuinely runs to six months and both sides knew it from day one.

Book a 45-minute growth diagnostic. If you're mid-engagement and unsure whether to continue, bring your last three reports. We'll tell you whether the problem is the clock, the process, or the brief. Start the conversation.

Frequently asked questions

How long does it take for digital marketing to show results in India?
It depends entirely on the channel. Paid media shows directional signal in 21 to 30 days and reliable cost per lead by day 90. SEO shows early movement at 90 to 120 days and meaningful commercial rankings at six months or beyond. Social builds over 60 to 120 days. Content-led organic demand typically takes six to nine months to become material.

When should I fire a marketing agency?
When process quality is poor, not when early results are. Poor process is visible at day 30: no research depth, unverified tracking, stalled creative output, reporting that describes instead of diagnoses, and no learning accumulating. Those are structural failures that do not improve with time. Weak early results on a long-cycle channel are usually just the channel behaving normally.

Is three months long enough to judge an SEO agency?
Not for rankings, but long enough to judge the work. At 90 days you should see technical issues fixed, content published, internal linking improved, and impression growth in Search Console even if clicks haven't moved. Impressions rising before clicks is the correct sequence. If none of that has happened by day 90, the problem is execution rather than timeline.

Why do impressions rise before clicks in Search Console?
Because Google first tests your pages at lower positions where they are seen but rarely clicked. As those pages accumulate engagement signals and authority, positions improve and clicks follow. Rising impressions with flat clicks at month three is a normal, healthy early pattern. The same pattern at month nine is a problem with relevance or click-through rate.

What should I expect from a marketing agency in the first 30 days?
Research demonstrating understanding of your buyer specifically, conversion tracking configured and verified, a stated hypothesis about what will work, creative or content visibly in production, and a first report containing something you did not already know. Outcomes at day 30 are noise on most channels. Process at day 30 is entirely judgeable.

Can slow approvals from my side affect results?
Substantially. Creative sitting unapproved for a week means media spend continues running on older assets and the testing cycle halves. If your average approval turnaround is seven days, you are effectively operating at half the cadence your retainer funds. Approval windows should be written into the scope document alongside the agency's obligations.

What if my agency keeps saying they are still optimising?
Ask what specifically will be different in the next 60 days, and what they have learned that they did not know at the start. Credible answers name a change in approach and specific knowledge about your buyer or funnel. The same optimisation explanation repeating for three consecutive months is a holding pattern, not a plan.

Should the judgement timeline be in the contract?
Yes. Agree the signal and outcome windows per channel in writing before work starts, along with the metric each is judged on and what a failed checkpoint triggers — ideally a structured remediation conversation rather than automatic termination. Most engagements that end badly could have been repaired at day 45 if anyone had a defined moment to raise it.