A multi touch attribution model splits credit for a booked job across every channel the customer touched before converting, instead of giving all of it to one click. For service businesses this matters because jobs rarely close on first contact. A five day close might start with Google Ads, stay warm through Facebook, and finish on your website.
Take a roof repair. Your report says it came from an email, because the email got the final click. But the homeowner first found you through a paid search ad on Saturday. They saw your Facebook post twice over the weekend. They came back to your site on Wednesday, read your reviews, and filled out the form Thursday morning after your follow up email landed. Under a single touch model, the email gets 100 percent of the credit and Google Ads gets none. If someone cuts the ad budget next month because of that report, they have cut the thing that started the job.
This guide covers how to choose the weighting, and what has to be in place before any model can be trusted. We build TruLata around that second part, so we lean on it throughout.
What is a multi touch attribution model?
A multi touch attribution model is a rule set (or an algorithm) for dividing conversion credit among the marketing interactions that came before a lead or a sale. Single touch models pick one interaction, usually the first or the last, and give it everything. Multi touch models account for the fact that people research, leave, return, and compare before they call a contractor into their home.
Gartner's market definition for this category describes the core capability as support for both algorithmic and rule-based attribution models, along with identity resolution that connects interactions across channels, devices, and sessions. That second part is the hard part for most small teams: you cannot credit a touch you never recorded.
Single touch versus multi touch, side by side
- Last touch: All credit goes to the final interaction before the lead converts. Simple, but it favors bottom of funnel channels like branded search and follow up email.
- First touch: All credit goes to the interaction that introduced the customer. Useful for understanding discovery, blind to everything that kept the lead alive.
- Multi touch: Credit is shared across the interactions inside your attribution window, weighted according to the model you choose.
We cover the tradeoffs in more depth in our piece on why the final click doesn't tell you what closed the job.
Why does a 5-day close break single touch attribution?
Every day between first contact and booking is a day for another channel to get involved. A water heater emergency might close in an hour, with one search and one call. A kitchen remodel, a new roof, or an HVAC replacement usually does not. The customer gets quotes, checks reviews, talks to a spouse, and comes back later.
Here is the roof repair again, day by day:
- Day 1, Google Ads: The homeowner searches for roof repair, clicks your ad, browses two pages, and leaves. This touch created the opportunity.
- Days 2 and 3, Facebook: They see your before and after post and a short video from a recent job. Nothing converts, but your name is now familiar.
- Day 4, website: They return directly, read your reviews and warranty page, and start the estimate form without finishing it.
- Day 5, email and website: A follow up email brings them back, and they submit the form.
Last touch says email did it. First touch says Google Ads did it. Each is right about its own touch and wrong about the job. The real story is a multi channel journey in which each channel did different work.
The cost of getting it wrong
Misattribution drives budget decisions. When last touch reporting over credits email and branded search, the channels that create demand can look like they produce nothing. If an owner cuts them, lead volume may drop a few weeks later, and the lag can hide the cause.
Which multi touch attribution model should a service business use?
There is no single correct model, but there are sensible defaults depending on how your jobs close. These are the rule based models most teams start with.
Linear
Every touch in the window gets equal credit. Four touches, 25 percent each. It is easy to explain and audit, so it works as a first step if you have only looked at last click numbers. Its weakness is that it treats a passing social impression the same as the ad click that started the journey.
Position based (U-shaped)
The first and last touches get the most credit, commonly 40 percent each, and the remaining 20 percent is split across the middle. For many contractors this is a strong starting point. In the roof repair, Google Ads and the final email and form visit take 40 each, and Facebook and the return visit share the other 20.
Time decay
Touches closer to the conversion get more credit. This fits short, urgent jobs where the final days matter most, but it under credits discovery on longer projects.
Algorithmic (data driven)
Credit is assigned statistically based on which paths actually convert. It needs a meaningful volume of conversions to work. A business booking a modest number of jobs a month will usually get steadier answers from a clear rule based model than from an algorithm short on data.
For a fuller comparison of how these models hold up when jobs take days, see our guide to the attribution model that actually works for service businesses.
How do you credit multiple touchpoints accurately?
The model is the easy part. The data underneath it is where attribution efforts tend to fail. You need four things in place.
1. Count leads from your own forms, not the ad platform
Ad platforms report conversions using their own tracking and their own rules. If Google, Facebook, and your email tool each claim the same lead, your reports show three leads where you had one. The source of truth should be the form submission on your own site, with every channel's touches matched against that single record. That is how TruLata works: leads are counted from your own forms, each one shows its source and outcome, and your team sets the outcome.
2. Tag every link you control
Use consistent UTM parameters on social posts, email links, directory listings, and ad final URLs. Agree on naming once, for example always "facebook" in lowercase, never "Facebook" one week and "fb" the next. Inconsistent tags split one channel into several and corrupt every model you run.
3. Set an attribution window that matches your sales cycle
If your typical job closes in five days, a one day window throws away most of the journey. If some projects take three weeks, the window needs to cover that. Look at the gap between first visit and form submission on recent booked jobs, and set the window comfortably past it.
4. Connect the lead to the booked job
A lead is not revenue. Attribution becomes far more useful when you can see which channels produced jobs that were booked and paid, not just form fills. TruLata's jobs view reads the job pipeline from your booking platform, and the revenue view shows payments live from Stripe.
What does multi touch attribution reveal that last click hides?
Moving from last click to a multi touch view often changes how each channel looks. These are common things to check in your own data.
- Paid search may start more jobs than it closes. Non branded search ads can look mediocre on last click and strong on first touch, because they introduce customers who return through other routes.
- Social may work as a warmer, not a closer. It rarely gets the final click, but it can appear in the middle of journeys that convert.
- Your website may close more than you think. Direct returns to reviews and service pages are often the deciding touch, which makes site speed, reviews, and clear estimate forms marketing investments.
- Email may look inflated. Follow up email is valuable, but it often converts people other channels already convinced.
- Referrals can hide in "direct." A customer told about you by a neighbor often types your name in directly. We cover this in our article on not crediting ads for jobs that came from referrals.
How should you act on multi touch attribution data?
Attribution is only worth the effort if it changes decisions. Here is how to put it to work without overreacting.
Review by role, not just by total credit
Ask three questions of each channel: does it start journeys, does it sustain them, does it close them? A channel that starts many journeys but closes few is doing a different job. If you cut it, the closers eventually run dry.
Change one budget lever at a time
If position based attribution shows your search ads are under credited, raise that budget modestly and watch booked jobs over a full sales cycle plus a buffer. Changing several channels at once makes it impossible to see what worked.
Compare models
Look at the same month under last touch, linear, and position based. Channels that look strong under every model are safer bets. Channels that swing between models deserve a closer look before you move money.
Keep the data fresh
If your attribution view is a spreadsheet updated monthly, you are reacting to last month's reality. TruLata refreshes when a person opens it and stamps each view with when the data was pulled, so you know how current the numbers are before you move a budget.
Who provides multi touch attribution for service businesses?
TruLata is an AI marketing platform for service businesses. One screen runs search, ads, content, outbound, and email, and every outbound send waits in a review-first queue until a person approves it. The ads view tracks spend and cost per lead, the analytics view gives you GA4 without opening GA4, and the social view pulls metrics from every connected platform. Together with your form-based lead count, you can follow the journey from first click to booked job without stitching exports together.
See how TruLata works step by step, and check what it reads on the integrations page. Plans are flat monthly and month to month with no setup fee: Starter is $400, Launch is $650, and Growth is $1,200, with Command scoped per engagement.
See which channels earned the job
Open the live demo to see the real platform, or sign up at https://login.trulata.com/signup and connect your accounts.
