Lead generation for consultants: how to build a pipeline beyond referrals

By Alfredo Silva · 8 October 2026 ·10 min read

Referrals give you something genuinely valuable: an introduction with trust already attached.

The difficulty is planning around them. A happy client might recommend you tomorrow, in six months, or only when somebody happens to ask. Your current project, meanwhile, has an end date.

Another route to clients gives you some influence over when conversations start. For an independent consultant the goal is not volume. It is a manageable flow of relevant opportunities that fits the work you can actually deliver.

Start with the next engagement you want to win

"We help businesses grow" leaves a prospective client with far too much to work out.

Describe the situation in which your expertise becomes useful instead. An operations consultant might help a manufacturer coordinate planning across a second site. A sales consultant might help a founder whose team has grown but whose selling still runs through them. A pricing consultant might help a specialist firm work out why its larger projects produce disappointing margins. Those are illustrative, not evidence that any particular company needs help.

Write a short brief for the next engagement:

You can serve several kinds of client. Pick one combination of client, problem and offer for the first prospecting effort anyway, because that is the only way to learn what resonates. Two half-targeted campaigns teach you nothing.

Build a working list you can explain

Start with a number of companies you could realistically investigate and speak to. Ten is a useful planning figure, not an ideal list size.

For each one, record why it fits, what you know, what is still uncertain, and who might own the issue. Company announcements, service pages, leadership updates and vacancies are where those answers usually live.

An observation should produce a hypothesis you can check, not a conclusion:

Swipe the table sideways →

Avoid turning a plausible interpretation into a claim about somebody's business. A new sales manager does not prove sales are struggling, and an opening message that assumes it does will be read as presumptuous by the one person you needed on side.

A research note that holds up follows four steps: Observed, Interpretation, Check, Action. That is also the structure of a good intelligence report, set out at length in what a sales intelligence report should include.

Three routes into conversations

1. Reconnect with people who already know your work

Previous clients, former colleagues and people who once considered working with you are the sensible starting point. Reconnect around the specific work you want now, not in general.

I'm focusing on helping manufacturers get planning and handovers working across multiple sites. Is that something you're seeing among the businesses you work with?

Illustrative, not a tested script. Adapt it to the relationship. What it does well is give someone a concrete situation to recognise, rather than asking them to invent your positioning for you, which is what "let me know if you hear of anything" asks.

2. Build relationships with people who meet the problem first

Think about who sees your prospective client's situation before you do. Depending on the offer that could be an accountant, a software implementer, a recruiter, or a consultant with complementary expertise.

Explain when your work becomes relevant, and learn where theirs starts and stops. A partnership needs a credible reason to benefit both of you and the client. Use the word partner when you mean an actual working relationship, and if you are approaching someone as a potential buyer, say so.

3. Approach selected companies directly

Use the research to ask a relevant question. Refer only to what you have checked, and keep the first message proportionate to what you actually know.

I saw the announcement about your second site. How are you approaching planning and handovers between the two locations?

That opens a conversation. It does not establish a need and it does not entitle you to a meeting. Be ready to say why you are asking and what relevant experience sits behind it.

Give interested people a concrete next step

A broad discovery call is hard for a buyer to evaluate, so make the purpose specific. An operations consultant could offer to walk one handover process and say whether a deeper review is warranted. A pricing consultant could explain how an initial assessment examines scope, delivery effort and margin.

You do not have to give away an engagement to do this. A short conversation can establish whether the problem exists, who owns it, and whether there is a sensible next step. Show relevant proof alongside the invitation: an anonymised deliverable, a clearly explained project, a practical article. Get permission where you need it, protect confidentiality, and label demonstrations as demonstrations.

Make content serve the conversations

Choose questions that come up while somebody is deciding whether to engage you:

• How do we know this problem is serious enough to address?
• What would an assessment actually involve?
• What can we handle ourselves?
• What has to be in place for this work to succeed?

Answer one of them properly, with an example you can substantiate, and use it in follow-up, on your profile, and wherever buyers evaluate you.

Here is the editorial test worth applying when writing time is scarce: would you be comfortable sending this to a prospective client even if it brought no search traffic at all? If not, write something else.

Protect a small weekly routine

Start with a schedule you can keep while delivering client work. Two hours, illustrative, for maintaining a small account list. Initial setup, sales meetings and proposal writing need their own time on top.

Swipe the table sideways →

Adjust to the complexity of your research. If one account eats the whole research slot, that may be the right call. The schedule is a capacity decision, not a forecast of replies or clients.

During a busy project, cut new prospecting to whatever you can sustain while keeping agreed follow-ups alive. When delivery capacity opens again you then have conversations to return to with context, instead of a cold start.

Track decisions, not activity

A spreadsheet holds the essentials perfectly well: company, person and fit; the evidence and its date; the last interaction; the stage, distinguishing research from replies, meetings, proposals and clients; the next action and when; and the outcome or the reason it closed.

Follow-up needs a purpose. Answer the question asked, honour the timing discussed, or share something relevant to the stated issue. Respect a request to stop. Where there is no response at all, use a limited sequence and then decide whether the account is worth a later look.

An illustrative planning exercise: suppose you want two new engagements. If one in four suitable proposals becomes work, you need eight proposals. If half your qualified conversations reach proposal, that implies sixteen such conversations.

Those rates are not benchmarks and not promises. Replace them with your own history as it accumulates, allow for the time between stages, and remember that a small sample moves about a lot. The exercise reveals workload. It does not forecast which month anybody signs. The same caution applies to pipeline coverage and to your sales cycle.

A weekly planner you can copy

Plain text, so it goes into whatever you already use.

WEEK OF: ______________
========================

Time I actually have this week : ____ hours

FOLLOW-UPS DUE
  [ ] company / person / what I promised / by when
  [ ] 

ACCOUNTS TO INVESTIGATE  (aim: a few, not a list)
  [ ] company -- why it might fit -- what I need to establish
  [ ] 

APPROACHES TO WRITE
  [ ] company -- the observation I am referring to -- the question I am asking
  [ ] 

CLOSED OR PARKED THIS WEEK
  - company -- reason -- revisit when?

WHAT I LEARNED
  - about the offer:
  - about who responds:
  - what to change next week:

No score and no promised number of meetings, because neither would mean anything. The last section is the one that compounds.

Measure without inflating

Track relevant replies, conversations that happened, proposals, wins, and the time each took. Look at why things stalled as well as at the totals.

Keep the stages honest, because this is where small numbers get flattered. From our own published work: in the EOS Implementer engagement, SONAR™ scanned 766 construction businesses and produced 20 verified owners, ranked by the moment their business was in and each carrying a reason to call. Jim called them, and two owners who had never met him agreed to join his next session.

Three numbers, three different things. A scan is not a shortlist and a shortlist is not an accepted invitation, and writing them in one column is how a small result starts sounding like a large one. The research answers for whether the 20 were the right 20. What happens on the call is yours, and it is supposed to be.

A smaller set of companies is worth pursuing when the reasons are clear. The commercial steps after that still have to happen.

When outside help makes sense

Consider it when you know the problem you solve and can handle the client conversations, but research or outreach keeps losing to delivery work. That is a capacity problem with a specific shape, and it is worth naming the missing task before you outsource it.

You might need help identifying relevant accounts, investigating changes, running an agreed outreach process, or keeping the records straight. Keep ownership of the offer, the substantive client conversations and the delivery promises.

SONAR™ addresses the research and the opportunities. RELAY™ addresses the outreach. Which one depends entirely on where your own process is getting stuck, and if you are weighing that against hiring somebody, hiring an SDR against outsourcing sets the two side by side.

Talk to Alfredo

Start this week with one clear offer, a small account list and protected time for follow-up. Build the routine you can keep when the next project starts, because that is the week it usually stops.

← Back to The Signal