The First Question I Ask a Client Is What Last Year's Revenue Was.

When a global hospitality group came to us, I didn't start by asking which keywords they wanted. I asked what revenue they closed last year and what the target is for this year. From that answer, the next questions follow naturally. You can then ask whether the gap is in basket size or penetration in a particular country, which markets under-contributed, and which ones already hit their number. Only after all that does a search strategy have anything real to attach to.

Hardly anyone asks about revenue first. I recently talked about this on the DailySEO ID podcast, where we argued that SEO should focus on driving revenue, not just improving rankings. I estimated that 99% of SEO professionals never begin by asking about revenue, and looking at the proposals I see, even that might be an overestimate.

What the packaged proposal leaves out

The failure has a familiar shape: an agency selling search like a food stall sells noodles. Package A gets you five backlinks, package B ten, and the premium tier ranges from ten to twenty keywords a year. Somewhere in the deck is a technical audit exported straight from Semrush or Ahrefs, a Core Web Vitals score, and a list of meta titles that run too long. None of that is wrong on its own; it is just beside the point because you are here and the business is over there.

Twenty keywords a year won't tell a founder what their company looks like in three years. A backlink package doesn't answer whether the business will grow from 5 billion in revenue to 15 billion. So the client signs, the campaign runs, the report arrives full of positions and sessions, and nothing connects to the number the board actually tracks.

Where ranking actually belongs

Ranking is not the goal and never was. It is a user indicator that shows a campaign is working, placing it below the goals and KPIs rather than above them. That is how the Goals Triangle I use is built. I put ranking that far down because user interest moves. Even when query volume holds steady, the intent behind it shifts, and behaviour around it shifts faster. A goal must be stable enough to steer by for a year, and search demand is not. People misread this constantly, so the two claims need to be separated. Saying ranking is not a goal is not the same as saying ranking doesn't matter. It matters a great deal, as long as it sits in the position matching what it can actually tell you.

Why search budgets here stay small

Look at where mature Indonesian brands put their money, and you will find the largest line is paid rather than organic. Inside organic, more goes to media relations than to search. Search sits in the budget like a stepchild, and I would argue it earned that spot honestly because the people running it couldn't show a return on their investment.

Antipathy compounds from there. I have friends who are brand managers at large FMCG companies, and several have concluded they would rather run ads or work with KOLs. You can verify this without commissioning a survey by visiting the brand sites across the FMCG portfolio in this market and seeing how many are maintained.

Pay carries the same gap. Comparing what a search practitioner earns here with that in Singapore or the US puts you somewhere between the first and the ninetieth floor. One engagement taught me how that works. A Singapore company hired me partly because I was cheap, and in year two they mentioned their previous agency charged three times as much. Better work for a smaller invoice is flattering for one conversation and corrosive for the decade that follows.

AI has taken the entry-level execution work

For years, a junior could enter this industry through execution: fixing meta descriptions, running the crawl, tidying the report, and learning business logic later. But that door is closing. The arithmetic is clear. A business can hire a junior analyst or pay a couple hundred dollars a month for a Claude subscription and get the crawl, analysis, and deck. Connect MCP to the CMS, Search Console, and Analytics, and a technical audit takes ten minutes. What comes out is not raw data but something you can take into a board meeting. I will be direct about what this means at my agency: I will not hire a junior SEO now because execution is handled.

That leaves a harder question for anyone entering the field now. My answer has nothing to do with which tools they learn. Three moves, in the order I would make them.

• Build the experience yourself. Buy a domain, launch sites in three different industries, run your own campaigns, and break your own things. Nobody gets to say they have never touched a Next.js site anymore.
• Look at campaigns from the business side before the search side. Ask what the product is, who buys it, and where the company sits today.
• Then go to a real business, as an employee or consultant, and describe the opportunity you see in search and how you can take them there.

Years of service have stopped functioning as a credential, since you can have thirty years in search and still hold no working model of how AI search picks an answer. What I want to know is whether someone understands how search works right now, which is also why a junior with real judgment can outrun a veteran who stopped learning in 2019. None of this replaces the person holding the standard, because garbage in, garbage out still governs everything, and an AI-assisted campaign is only as good as the practitioner driving it, since validating output means knowing roughly what the answer should be before you see it.

What people are actually typing

The behaviour change makes the revenue argument more urgent, not less. Traditional search ran on short queries like cheap shoes, on head terms with high volume and a scramble to rank for them, and anyone who came up optimising for that pattern learned to treat volume as the prize.

Watch someone use an assistant now, and you'll see they don't type cheap shoes; they type that they want running shoes under a million rupiah, durable in wet weather, comfortable enough for a marathon, and available with cash on delivery. One message puts the entire set of requirements on the table, and that isn't the future; that's today.

Content built for the broad head term answers none of it, while content built from the audience persona, from the pain points, and from the queries those specific people type answers all of it. That sequence is why we built StoryMint: we start with who the audience is before anything gets written, which puts it in a different category from a keyword database. Put the user first and SEO follows.

The question to open with instead

The pitch doesn't need to change; only the opening question does, and it isn't which keywords do you want to rank for but what your business objective is and how search can contribute to it. Everything downstream of that- the technical work, the content investment, and the AI visibility measurement- gets easier to prioritise once the answer exists, because you finally know what you're optimising toward.

If your search reporting still stops at positions and sessions, that's the gap to close first, and it's the work we do on the measurement side. Full conversation on the DailySEO ID podcast, SEO Harus Bicara Revenue, Bukan Cuma Ranking, roughly an hour, in Indonesian.