The Retention Trophy

J.P. Morgan has named WPP the industry's best at winning new business.

J.P. Morgan also has a 'Neutral' rating on WPP shares.

Both are being honest. That's the interesting bit.

WPP is proud of their top ranking, and loud about it. When good news is otherwise sparse.

In The Two Juries I asked which jury was the better futurist: the creative jury that crowned WPP at Cannes, or the market jury that has removed most of its value.

Here the two juries sit inside the same bank. J.P. Morgan's New Business Review ranks WPP #1 for Q2 2026, for the first half, and for the nine months to Q2. J.P. Morgan's equity research desk maintains WPP at Neutral. Same institution. Same company. Opposite conclusions about what the trophy means.

And like the chart I unpacked in The Mirage Metric, nobody is lying. The data is real. The conditions that produced it are real. But what people see when they look at a new business league table is not what's there. This week, WPP's own channels celebrated the ranking, with CEO Cindy Rose calling it evidence of "real momentum". I want to explain why the desk down the corridor from the league table's authors, looking at the same company, shrugged.

Let's see how WPP H1 results next week illuminate this.

I have stood in the room when the call comes through, too many times. The pitch team crowded around a speakerphone, the pause, the word "congratulations", the roar. Somebody finds warm prosecco. Another starts drafting the press release nobody dared write in advance, for fear of a jinx. The trade press writes it up with a billings number that nobody in the room recognises, and nobody corrects.

Then comes the Monday. Two weeks later, in a smaller room with no champagne, the commercial team walks the leadership through the transition plan: the team to be hired ahead of revenue, the onboarding period the agency is funding, the rate card that won the work, and the year (sometimes two) before the account washes its face. The win was real. The celebration was real. The economics arrive later, quietly, and they are almost never what the press release implied.

This newsletter is about the distance between those two rooms.

The problem of measuring this gap is not the integrity of the scorekeepers. The problem is structural: what the metric can see, and what it cannot. Which brings us to the filters.

Not a single number in new business tables is verifiable.

No client I have ever worked with, nor the consultants who run a pitch, publishes the value of business under review, or the value of the contract that results. No agency does either. Every billings figure in every tracker is an estimate. The estimate travels, a number appears in one trade story, gets repeated, and hardens into fact through repetition rather than confirmation.

Want proof of how soft the ground is? Compare the banks. In the same quarter that J.P. Morgan's estimates crown WPP #1, Bank of America's tracker (built on Comvergence data) puts WPP's Q2 net new business, once retentions are stripped out, at negative $160 million. On the same methodology, Publicis sits at roughly +$1.8 billion. Two respected institutions, tracking the same pitches in the same ninety days, produce a champion and a decliner. Both are defensible. That alone should change how much weight any of us puts on a ranking.

Now apply the second filter, the one that gives this piece its title: how much of the "new" business was business the agency already had?

I'll state it as plainly as I can, from both sides of the table. Clients do not run pitches in order to pay their incumbent more. In my time in the client chair no agency has ever pitched to charge me more, and no procurement team has ever taken an incumbent to a review and emerged with the agency's margin improved. A retention is a defence, conducted at the client's initiation, concluded on the client's terms. By the time a review opens, the relationship has already failed a test and the pitch merely prices the failure.

Look at WPP's Q2 'wins' (including L'Oréal ANZ and Huawei China) where WPP were incumbent. Real work, real relief, and counted as wins by every table. But commercially they are the same account at a thinner margin, delivered by a team that has just spent months on a pitch to keep the work it already did and now has to do more of it, for less, while exhausted. The league table records a trophy. The P&L records a discount. Both are describing the same event.

Even the true wins, the accounts that genuinely moved from another agency, rarely make economic sense inside the first year. The winning agency often funds a transition period. It hires and redeploys ahead of revenue. It over-invests in the first cycle to make the honeymoon succeed because the pitch was won at pricing designed to win a pitch. The win is a cost centre before it is a client, and that assumes the steady-state terms were profitable in the first place.

None of this makes new business worthless. It makes the twelve-month economics of a league-table quarter very different from the headline, which is one reasonable reading of why the equity desk at the very bank publishing the table declines to get excited about topping it.

In every field that manages risk, the real heroes are invisible: the people who quietly prevent the disaster every day, whose success is measured by the absence of an event. Agencies are no different. The best outcome in this business is a client who never appears on any table at all. Never in the pitch tracker, never in the wins column, never in the "billings to defend" chart, because the team serving them has grown the relationship organically, year after year, expanding scope as a partner rather than defending it as a supplier. There is no league table for that. There is only the organic growth line.

So if you want a ranking that actually predicts the future, invert the trophy cabinet and read the risk register. On BofA's estimates, WPP currently has about $3 billion of media billings to defend (the most of any holding company) anchored by the $1.4 billion Coca-Cola review (global excluding the US, Japan and Korea), expected to be awarded in the autumn. By the mere fact of appearing on the list, these clients have told their incumbents something. The hierarchy is brutal and simple. Best: the client who never reviews, grown organically by teams nobody writes about. Next: a genuinely new, genuinely profitable client that pays back inside twelve months. Worst, the retention trophy: the client you kept because you had to, at terms you wouldn't have offered, served by a team that has been through a pitch and now works harder for less.

The tables celebrate the third category as if it were the first.

Which returns us to the corridor at J.P. Morgan. The New Business Review is measuring announcements. The equity desk is measuring economics, the billings at risk, the margin on the retentions, the year-one cost of the switches. Both are doing their jobs well. Only one of them is telling you about the future, and on 6 August, when Cindy Rose presents WPP's first-half results, the first results she will present since laying out the strategy in February, we'll see the two verdicts meet in a single room. The H1 statement is the first answer, and I'll be covering it here.

Brandflow is written by Justin Billingsley, who has spent his career on all three sides of the industry's table: senior client, global agency leader, technology founder. First published 27 July 2026 in the Brandflow newsletter on LinkedIn.