Lead generation for consultants and agencies: what works

Almost every page ranking for this query is an agency lead magnet, and every one of them ends the same way: a form, and an offer to do it for you.

That is not a criticism of the business model. It is a warning about the advice, because a page written to demonstrate that the writer should be hired will not tell you the part that makes this hard.

Here is the part. You are selling judgement, and judgement cannot be sampled.

Why this is a different problem

A software buyer can try the thing. A trial costs them an hour and no trust. If it works they buy, and your marketing only had to get them to the trial.

Your buyer cannot do that. They are deciding whether your thinking is good, in advance, on incomplete information, with real money and their own credibility attached to being wrong. There is no trial that reduces that risk.

So every technique that works for software leads fails here, and it fails in a specific way: it optimises for volume of contact, when the constraint is evidence of judgement.

The one advantage you have

Software companies cannot demonstrate their product in a cold message. You can.

You can look at a company's actual situation, form a view, and say it. That view is the product. Nothing stops you from giving away a small, true piece of it before anyone has paid you, and almost nobody does, because it feels like working for free.

It is not working for free. It is the only sample you are able to offer.

The message that works is not "we help agencies like yours with X." It is a specific observation about their situation that they had not articulated, followed by what you would look at next. If your judgement is good, that lands as a small demonstration. If it is not, you have learned something important cheaply.

What this means for who you contact

The volume playbook actively hurts you here, for a reason worth spelling out.

To demonstrate judgement you must understand the specific situation. Understanding a specific situation takes real time. So your reachable list is small, and trying to make it big is exactly what turns your differentiated message back into the generic one your prospect already deletes ten of a week.

This is the ceiling described in how to personalize cold email at scale, and for services businesses it binds harder, because the depth required per message is greater.

Twenty contacts a week feels like nothing next to a software team's numbers. It produces more revenue for a services business, because one signed project is worth what a hundred software signups are worth.

The referral question

Every consultant is told that referrals are the answer. They are right, and it is unhelpful advice, because referral is an output, not a channel. You cannot decide to receive more of them.

What you can do is remove the two reasons they do not happen.

Your clients do not know who to refer. They know you are good. They do not have a sentence describing who you help, so they never recognise a match in conversation. Give them the sentence, in the language they would use, and repeat it. This is the highest-return thirty minutes in a services business and most people never do it.

They do not know you have capacity. People do not refer to someone they assume is full. Saying you have room in the next quarter, plainly, unlocks referrals that were already there.

Neither is outreach. Both produce leads.

The trust ladder, in order

Cold to signed is too far in one step for most services buyers. What actually happens is a sequence, and skipping steps is what makes outbound feel like it does not work.

What to publish, specifically

The evidence step is where most practices stall, because the advice stops at build authority, which is not an instruction.

Here is the instruction. Publish the reasoning you would normally only show a client after they pay.

Not a case study. Case studies are structurally unbelievable, because everyone knows they are selected and the numbers are unverifiable. A prospect reads one and learns that you have clients, which they assumed.

What works is a piece of thinking about a problem in your field, done properly, in public. A trade-off you have formed a view on. A common approach you think is wrong and why. A decision procedure you actually use. It works because it is checkable: a reader can evaluate the reasoning itself rather than take your word about an outcome.

This is also the only asset that keeps working while you sleep and the only one that makes a cold message land, because the message says something specific and the published thinking is what a curious recipient finds when they check whether you are worth answering.

Volume does not matter here. One genuinely good piece outperforms twenty competent ones, because the reader is sampling your judgement and one sample is enough.

Pricing the first small project

Step four fails more often on price than on scope, in both directions.

Too cheap and you have signalled that the work is low value, which makes the full project harder to sell rather than easier. Discounting the entry point teaches the buyer what your thinking is worth.

Too expensive and it is not a sample any more, it is a purchase requiring the trust the sample was supposed to build.

The shape that works: priced so that a buyer with budget authority can approve it without a committee, scoped so that it delivers something useful even if they never buy anything else, and defined tightly enough that both sides know when it is finished.

That last clause matters most. An open-ended small project becomes free consulting, and the resentment that produces will cost you the relationship faster than never having started it.

What to measure

Not leads. Conversations with people who could actually buy.

For a solo consultant, three or four of those a month is a functioning pipeline. That number sounds small enough to feel like failure if you have been reading software marketing content, which is most of what exists on this topic.

The leading indicator is the share of your outreach where you could form a specific view before writing. If that share falls, you are stretching for volume and the whole advantage collapses. How to prioritize sales leads covers deciding where that effort goes, and how to research a prospect before a call covers producing the view efficiently.