Eleven RFP responses received. Scored against a weighted matrix. Three agencies on the pitch stage.
The build that followed will run into the mid six figures. The selection consulting was a rounding error against it. The job was to make sure the right agency was chosen for the right reasons, and to make sure every agency that didn’t get picked walked away with feedback they could use.
The starting point
The RFP from the previous case study went to the long list. Responses landed 3 weeks later. Eleven came back in full. Some were tight, well structured, twenty pages of focused detail. Some were a hundred and twenty pages of marketing copy with the answer to the actual question buried in a footnote. The job was to read all of them with the same eye, score them on the same criteria and end up with a defensible shortlist.
The marketing lead didn’t have the bandwidth to read eleven RFPs cover to cover and score them consistently. The partners had even less. That work needed to happen outside the in-house team, by someone with a scoring framework already built and no relationship with any of the agencies in the room.
The approach
1. The scoring matrix came first
Before I opened the first response, I built the scoring matrix from the RFP itself. Every question already had a weight attached, between 2 and 5, reflecting how much it mattered to the firm. The 5s were the questions where the answer would make or break the engagement: how the rebuild would be approached at scale, how the agency would integrate with the firm’s CRM and marketing automation, how it would handle GDPR, the budget, the timeline, the security certifications and the CMS recommendation.
Forty one questions, weighted. Each answer scored 1 to 5 against the rationale documented in plain English so the partner team could see why an agency had got the score it had got. Cumulative weighted score per agency. Top three on weighted score progress to the pitch day. The rest get feedback.
The scoring matrix had to stand up because the conversation that followed would be partner driven. Marketing leads at law firms can defend a recommendation if they can show their working. They can’t defend “I thought their pitch deck was nice.” The matrix was the working.
2. Reading every response with the same eye
Eight working days to read all eleven responses, score every answer, write the rationale and tabulate the totals.
What you find when you read eleven RFP responses in succession is that they cluster into three groups. The top group answers the question that was asked, with specifics, with examples and within the word limit. The middle group answers a slightly different question, the one they wished you’d asked, with case studies that don’t quite match the brief. The bottom group hasn’t read the brief properly. One agency skipped the timeline question entirely. Another said “this is not an official budget” and left the cost section blank. Both were incumbents on adjacent accounts and had assumed the relationship would carry them. It didn’t.
A few specifics worth naming. The agencies that scored highest had three things in common.
They held actual security certifications, not “aligned to the principles of.” For a firm handling regulated client data across two jurisdictions, this question carried a weight of 5 and the gap between holding ISO 27001 and being “in process” was a full two points on a five point scale.
They quoted a fixed price, broken down by phase, with hours and rates legible. The agencies that quoted time and materials, or a budget range with 27% variance, lost points immediately. A firm being asked to defend a six figure spend to a partnership meeting can’t open with “the cost is somewhere between this and that.”
They had specific named references with contact details, not “available on request after shortlisting.” One agency declined to provide references at all until a contract was signed. That answer was so unusual it became a talking point in its own right.
3. The shortlist conversation
Two hours, in the marketing lead’s office, with the partners who’d signed off the RFP. Scoring matrix on screen. Top five agencies on a slide. Justification per agency, with the questions they’d answered well and the questions they hadn’t.
The conversation was sharper than it would have been without the matrix. Partners are trained to ask for the evidence behind a recommendation. The matrix gave them the evidence in the format they expected. Disagreements happened, and got resolved on the data rather than on impressions.
We landed on three for the pitch day. A close fourth went on the reserve list with explicit feedback that we’d come back to them if either of the top three pulled out.
4. The pitch day
Three agencies. One day in the firm’s regional office. Forty five minutes per pitch, fifteen minutes of questions, thirty minutes between to debrief. The partners sat in. The marketing lead chaired. I sat in the corner of the room and took notes.
The questions in the pitch day were written from the parts of each agency’s RFP response that had scored lowest. If an agency had been vague on integrations, that’s what we asked them about. If an agency had been thin on the timeline, that’s what we asked them about. The pitches weren’t an opportunity for the agencies to repeat their RFP responses. They were an opportunity for the agencies to address the gaps in them.
The debriefs between pitches were the part the marketing lead later said was the most useful. Twenty minutes with the partners off the record, talking about what we’d just heard, what was missing, what had landed, what to ask next. By the time the third agency left the room, the decision was almost made. The decision call the following morning ratified it formally.
5. The feedback to the unsuccessful
Eight agencies didn’t make the pitch day. Two more pitched and didn’t win. Ten phone calls to make. Each one a structured conversation. What scored well, what scored less well, what specifically would have changed the outcome.
This part of the engagement matters more than it looks. The legal sector is small. The agency market that serves it is smaller. Every agency that pitched on this RFP will pitch on the next one and the one after. Walking away from a process with constructive feedback, rather than a one line rejection email, changes how those agencies show up next time. It also changes how they talk about the firm in the market.
A few specifics from the feedback calls.
The incumbent that scored 3.9 out of 5 despite knowing the firm better than anyone got a longer call. The honest feedback was that incumbent advantage isn’t a substitute for answering the RFP properly. Skipping the timeline question and refusing to commit to a budget were unforced errors that no relationship would have absorbed.
The agency that quoted the highest price by some distance and required a proprietary CMS licence at over £80,000 a year got feedback on value proposition. Their technology was strong. The pricing model meant they ruled themselves out of consideration before the technical review started.
Where it landed
By the end of month 2, the firm had appointed its agency for the build. The contract was signed against the named deliverables in the RFP, with the agency’s response as a contractual appendix. The build kicked off with no scope dispute and no re-quoting. The discovery work from the previous engagement carried straight into the build.
The agency selection process took eight weeks from RFP submission to contract signing.
The marketing lead got three things out of the process they didn’t have before.
A decision the partners signed off on the first time of asking. The scoring matrix carried the meeting. There was no second round of partner debate.
A contract that referred directly to the agency’s own RFP response. If the agency under delivered on anything they’d committed to in writing, the contract gave the firm a route to recompense without renegotiation.
A market reputation as a firm that runs procurement well. The agencies that didn’t win on this RFP will be in the room on the next one. Their experience of this process makes them more likely to put their best people on the next pitch.
Real business outcomes
- RFP responses received; 11
- Agencies shortlisted for pitch;3
- Days from submission to appointment;48
- Disputes over scope after contract signing; 0
- Re-quoting events after kick off; 0
- Build budget range across the 11 responses; £128k to £540k+
The variance in that final row matters more than any single number. The same brief, with the same word limits and the same questions, came back with cost answers that ranged across a factor of four. Discovering that variance during the RFP, not during the build, was the whole point of the process.
What this means in practice
A bad agency selection process costs more than the agency selection consulting ever could. The two ways it costs you are well documented.
The first is appointing the wrong agency. Six months in, the relationship breaks down. The contract has no teeth because the original brief was loose. The firm starts again, with the calendar against it and the partner team less patient than the first time.
The second is appointing the right agency on the wrong terms. The build comes in late and over budget because the agency priced what they could see and rebid the scope as the work revealed itself. The change orders pile up. The marketing lead spends three months arguing about scope rather than three months running the project.
Both failure modes are functions of how the agency was selected, not how the agency works. Running the selection on a weighted scoring matrix that ties back to the original RFP, with a pitch day that probes the gaps in each agency’s response, with a contract that names the response as an appendix, removes the conditions under which either failure mode is likely to happen.
The consulting cost value of those eight weeks of process is the difference between a build that lands on time and a build that doesn’t.
What I’d tell a marketing lead about to do this
Five things, in order.
- Build the scoring matrix before you read the first response. The weights belong to the RFP, not to your impressions of the agencies as their proposals come in.
- Score with rationale, not with stars. A score of 4 with two sentences of reasoning beats a score of 5 with no reasoning at every partner meeting you take that score into.
- Use the pitch day to probe the gaps, not to repeat the brief. The agencies have already had thirty pages to make their case. Twenty minutes of questions on the parts they didn’t address tells you more than another forty five minute deck.
- Give the unsuccessful agencies constructive feedback. The market is small. Your next RFP will land in the same inboxes. How you ran this one decides who shows up for the next one.
- Sign the contract with the agency’s RFP response attached as a contractual document. The agency wrote the words. They can defend them. If the build slips, you have a starting point for the conversation.
About this engagement
Eight week engagement. Followed agency selection from RFP submission to contract signing. Pitch day attended in person at the firm’s regional office.